228 21 28MB
English Pages XXXII+248 [288] Year 1986
This entirely new work is the last volume to be published in the Oxford History of England, and replaces J. N. L. Myres’s own part of the first volume in the series. Roman Britain and the English Settlements — the classic ‘Collingwood and Myres’. The author returns to the subject of his earlier contri bution — the dark centuries of English history between the collapse of Roman rule in the early fifth century and the emergence of the Anglo-Saxon kingdoms in the seventh — but considers the period afresh in the light of the rapid proliferation of work on the subject in the half-century since the first volume was published. Much new evidence on the literary sources, on the archaeological evidence both in England and on the conti nent, and on place names and other linguis tic developments has led to significant changes in emphasis: Dr Myres now draws attention to some little-understood factors which seem to link Roman Britain with Anglo-Saxon England, and so suggests strands of political and social continuity which may help to explain this complex and traumatic period of our history. 1 his volume takes its place in the series as Volume lb, alongside Peter Salway’s major recent study of Roman Britain (Volume la, 1981; Oxford Paperbacks 1984), which replaced R. G. Collingwood's part of ‘Collingwood and Myres’. J. N. L. Myres is Honorary Student of Christ Church and formerly Bodley’s Librarian in the University of Oxford. Jacket illustration: Anglo-Saxon pot from North Elmham, Norfolk. By permission of the Trustees of the British Museum
THF. O X F O R D H I S T O R Y O F ENGLAND Edited by SIR GEORGE CLARK
T H E O X F O R D H IS T O R Y O F E N G L A N D E d i t e d b y SIR G E O R G E CLARK
IA . R O M A N B R I T A I N By PETER SALWAY, 1981 IB. T H E E N G L I S H S E T T L E M E N T S B y J . N . L. MY R E S , 1985
II. A N G L O - S A X O N E N G L A N D , c . 550-1087 By SIR FRANK STENTON. Third edition, 1971 III. F R O M D O M E S D A Y B O O K T O M A G N A C A R T A . 1087-1216 By AUSTIN L. POOLE. Second edition, 1955 IV . T H E T H I R T E E N T H C E N T U R Y . 1216-1307 By SIR MAURICE POWICKE. Second edition, 1962 V . T H E F O U R T E E N T H C E N T U R Y . 1307-1399 By MAY MCKISACK, 1959
■
V I. T H E F I F T E E N T H C E N T U R Y . 1399-1485 By E. F. JACOB, 1961 V II. T H E E A R L I E R T U D O R S . 1485-1558 By J . D. MACKIE, 1952 V III. T H E R E I G N O F E L I Z A B E T H . 1558-1603 By J. B. BLACK. Second edition, 1959 IX . T H E E A R L Y S T U A R T S . 1603-1660 By GODFREY DAVIES. Second edition, 1959 X . T H E L A T E R S T U A R T S . 1660-1714 By SIR GEORGE CLARK. Second edition, 1956 X L T H E W H I G S U P R E M A C Y . 1714-1760 By BASIL WILLIAMS. Second edition revised by C. II. STUART, 1962 X II. T H E R E I G N O F G E O R G E I I I . 1760-1815 By J . STEVEN WATSON, 1960 X III. T H E A G E O F R E F O R M . 1815-1870 By SIR LLEWELLYN WOODWARD. Second edition, 1962 X IV . E N G L A N D . 1870-1914 By SIR ROBERT ENSOR, 1936 X V . E N G L I S H H I S T O R Y . 1914-1945 By A. J. P. TAYLOR, 1965
T HE
ENGLISH SETTLEMENTS By *
J . N. L. MY R E S s o m e tim e B o d le y 's L ib ra ria n a n d P r e sid e n t o f t h e S o c ie ty o f A n tiq u a rie s
C L A R E N D O N PRESS • O X F O R D 1986
Oxford University Press, Walton Street. Oxford 0 X 2 6DP Oxford New York Toronto Delhi Bombay Calcutta Madras Karachi Kuala Lumpur Singapore Hong Kong Tokyo Nairobi Dar es Salaam Cape Town Melbourne Auckland and associated companies in Beirut Berlin Ibadan Nicosia Oxford is a trade mark o f Oxford University Press Published in the United States by Oxford University Press, New York © Oxford University Press 1986 All rights reserved. No part o f 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 o f Oxford University Press British Library Cataloguing in Publication Data My res. J. N. L. The English settlements. — (Oxford history o f England: IB) I. Great Britain History — Roman period, 55 B.C.-449 A.D. 2. Great Britain — History — Anglo-Saxon period, 449-1066 I Title 942.01 DA 145 ISBN 0-19-821719-6 —
Library o f Congress Cataloging in Publication Data Myres, J. N. L. (John Nowell Linton) The English Settlements. (The Oxford history o f England; IB) Bibliography: p. Includes index. 1. Great Britain History Anglo-Saxon period, 449-1066. 1. Title. 11. Series. DA152.M97 1985 942.01 85-15538 ISBN 0-19-821719-6
Set by Grestun Graphics, Abingdon Printed in Great Britain at the University Press, Oxford by David Stanford, Printer to the University
ACKNOWLEDGEMENTS HIS b o o k co u ld n ev e r have re a c h e d th e p o in t o f p u b lic a tio n w ith o u t th e g e n e ro u s assistan ce o f m an y frie n d s a n d scholars w orking in th e sam e fields w h o have c o n trib u te d in d iffe re n t w ays to its c o m p le tio n . T o a n u m b e r o f th ese I have ex p ressed m y o b lig a tio n in th e te x t o r th e n o te s. But I o w e m u c h m o re th a n I m a y realize to p erso n a l c o n ta c ts o v er m o re th a n fifty y ears w ith th o se , b o th in th is c o u n try and a b ro a d , w ith w h o m 1 h av e o fte n d iscu ssed th e p ro b lem s o f th is age. I am also g re a tly in d e b te d to m y so n , D r. M. T . M yres, o f th e D e p a rtm e n t o f B iology a t C algary U n iv ersity , w ho w as m a in ly resp o n sib le fo r persu ad in g m e to p ersevere w ith th is ta sk a t an age w h en m an y a u th o rs h av e had th e ir fill o f read in g a n d w ritin g , a n d m ay fin d it m o re b u rd e n so m e th a n th e y o n c e d id t o ex p ress th em selv es in tellig ib ly in p rin t. His ex p e rien ce o f u n iv ersity teac h in g in C an ad a, alb eit in a field o f s tu d y fa r rem oved fro m m in e, h a s e n a b le d m e to avoid o b sc u ritie s o f ex p ressio n and to p ro v id e e x p la n a tio n s , esp ecially o f so m e tech n ica l te rm s, w hich m ay be h e lp fu l to s tu d e n ts u n fa m ilia r w ith th e h isto rical and geographical b a c k g ro u n d s in w h ich th e s to ry o f th e English se ttle m e n ts m u st b e set. T h e w elco m e d ec isio n o f th e p u b lish e rs to allow fo r th e first tim e in th is series som e illu stra tio n s in a d d itio n to m ap s has e n a b le d m e to in c lu d e a few visual aid s, to ex p lain th e b asic seq u en ces in th e arc h a e o logical ev id en ce fo r th is p e rio d w hich a rc d iffic u lt e ith e r to e x p o u n d o r to c o m p re h e n d w ith o u t su ch h e lp . F o r p erm issio n to re p ro d u c e fo r th is p u rp o se m aterial alread y p u b lish e d elsew here in d iffe re n t c o n te x ts I am g re a tly in d e b te d t o P ro fe sso r V. I. Evison, M rs. S. C. H aw kes, Dr. M . W elch, a n d M r. W illiam J . R o b e rts IV. T h ey a re o f co u rse in n o w ay re sp o n sib le fo r a n y u n e x p e c te d co n c lu sio n s I m ay h av e d ra w n fro m the o b je c ts th em selv es. I am also m o st g ra te fu l to D elia T w am ley a n d M argaret G o lb y , w ho have ty p e d at d iffe re n t stages a c o n tin u o u sly evolving a n d o fte n ra th e r m essy m a n u sc rip t w ith ad m ira b le ac cu rac y a n d p re cisio n . A b o v e all m y th a n k s are d u e to m y g o o d frie n d s J e a n C ook a n d G race Briggs, w ho have c o lla b o ra te d m o st g en e ro u sly a n d e ffic ie n tly in relieving m e o f nearly all th e tire so m e a n d tim e -c o n su m in g in cid en tals o f a u th o rsh ip . T h e y have id e n tifie d , c h e c k e d , a n d sta n d a rd iz e d m y o fte n w ayw ard re fere n ces, in v estig ated m an y q u erie s, a n d re m o v e d m in o r o b sc u ritie s, d u p lic a tio n s, a n d e rro rs o f all k in d s. While a n y rem ain in g d e fe c ts are e n tire ly m y re sp o n sib ility , it is d u e to th e m th a t th is b o o k has reached a fo rm w h ich m ak es p u b lic a tio n possible. 1 am m o st g ra te fu l fo r all th a t th e y have d o n e to b rin g th is to pass.
T
J . N. L. M.
CONTENTS LIST OF FIGURES
xiv
LIST OF MAPS
XV
ABBREVIATIONS
xvi
INTRODUCTION
xvii
MAPS 1 -4
x x ix -x x x ii 1. T H E N A T U R E O F T H E E V I D E N C E : I. T H E L I T E R A R Y S O U R C E S
Im p o rta n c e o f th e p e rio d C auses o f its o b s c u rity D eficien cies in th e lite ra ry ev id en ce L ack o f contem porary* re c o rd fro m a n illite ra te so c ie ty A b sen ce o f reliab le d a tin g B ed e’s in te re s t in c h ro n o lo g y In a d e q u a c y o f c o n tin e n ta l sources T h e G allic C h ro n icles C o n s ta n tiu s ’ Vita G erm a n i G ildas a n d th e a p p e a l to A etiu s B ede’s c o rre c tio n o f th e d atin g M eaning o f th e A d v e n tu s S a x o n u m G ildas a n d th e obsessio m o n tis B a d o n ici V alu e o f G ild as’ te s tim o n y fo r th e p e rio d o f his o w n life His m e n tio n o f A m b ro siu s: n o m e n tio n o f A rth u r T h e H isto ria B r itto n u m a n d A rth u ria n legends E arly sain ts’ lives a n d o th e r in c id e n ta l m aterial T h e A n g lo -S a x o n C h ro n icle as a so u rc e fo r th is p e rio d
1 2 4 4 5 6 7 7 8 8 9 10 11 12 14 16 18 20
2. T H E N A T U R E O F T H E E V I D E N C E : II. A R C H A E O L O G Y A N D PLA C E-N A M ES A rch aeo lo g ical ev id en ce Its n eg ativ e c h a ra c te r o n th e B ritish side R easo n s fo r th is a n d its sig n ificance A nglo-S axon m a te ria l m a in ly fro m c e m e te rie s M eans o f d a tin g arch aeo lo g ical evidence L ack o f p re cisio n in ev itab le in ty p o lo g ic a l stu d ie s E v id en ce fo r lo n g use o f o b je c ts o f in trin sic value S ig n ifican ce o f w ea r-a n d -tear a n d u n e x p e c te d asso ciatio n s Im p o rta n c e o f fragile o b je c ts o f low cost and sh o rt life su ch as p o tte r y R elativ e ev id en tial value o f c re m a tio n a n d in h u m a tio n fin d s Im p o rta n c e o f arch aeo lo g ical evidence fo r se ttlin g d a te and p ro v e n a n c e o f im m ig ra n t se ttle rs
21 22 23 24 25 25 25 26 27 28 29
Vlll
CONTENTS
P lace-nam e evid en ce U n sa tisfa c to ry n a tu re as h isto ric a l m a te ria l C o n fu sin g e ffe c ts o f o ra l tran sm issio n in illite ra te so cieties L aten ess o f m o st w ritte n fo rm s M istakes d u e to linguistic ignorance o f scribes D ifferen tial survival o f p lace-n am e elem en ts Loss o f villa n am es a n d th o se o f o th e r R o m an o -B ritish c e n tre s C o rru p tio n o r loss o f to w n n a m e s S ignificance o f n a m e s in ceaster, w ie, eccles, f u n t a , e tc . Vicus a n d w ic-ham n am es V icus a n d villa S e ttle m e n t n a m e s in -ingas, -ham , -tu n , e tc . P ro b le m a tic m eaning o f th e -ingas a n d re la te d fo rm s R elatio n o f -ingas n am es to arch aeo lo g ical d istrib u tio n s T h e ir re la tio n to c o m p o u n d s in -ham , -tu n , e tc . S ocial significance o f th ese d istrib u tio n s a n d th e ir ev o lu tio n E n d u rin g co m p le x itie s o f p lace-n am e stu d y
29 29 30 30 30 31 31 31 32 33 35 36 36 37 40 41 44
3. T H E C O N T I N E N T A L B A C K G R O U N D S ax o n s, A ngles, a n d J u t e s o n th e C o n tin e n t acco rd in g to B ede P o sitio n o f th e F risians E vidence o f T a c itu s fo r Frisians, A ngles, a n d E udoses P to le m y a n d th e S ax o n s: th e ir c o n n e c tio n w ith th e C hauci T h e T erp en o n th e N o rth Sea co ast F e d d e rse n W ierde a n d W ijster L and sinkage a n d th e m o v em en t to B ritain C arausius a n d th e S axon S h o re P ressure fro m S candinavia o n th e A ngles, th e J u te s , a n d th e Frisians T h e c u ltu ra l b a c k g ro u n d a n d th e ty p o lo g y o f b ro o c h e s L ong b ro o c h çs a n d cru c ifo rm b ro o c h e s fro m th e n o rth S aucer, a p p lie d , a n d b u t t o n b ro o c h e s in S axon areas E q u al-arm ed b ro o c h e s a n d o th e r S ax o n ty p es S ignificance o f th e d is trib u tio n o f b ro o c h ty p e s in E ngland T h e ir re la tio n sh ip to fa sh io n s in dress F ash io n s in p o tte r y A nglian sty les in d e c o ra tio n a n d th e N o rd se e k u ste n g ru p p e J u tis h sty les in p o tte r y R elatio n s b e tw e e n J u tla n d a n d K en t S ax o n sty les in p o tte r y S c h a le n u m e n , S ta n d fu sssc h a len , a n d B u c k e lu m e n Parallels b e tw e e n c o n tin e n ta l a n d English fashions in p o tte r y S om e p e rso n a l lin k s b e tw e e n p o tte r s S ignificance o f th e d is trib u tio n o f various ty p e s in E ngland
46 47 49 50 51 52 52 53 54 56 57 59 61 62 62 63 64 66 66 66 66 69 72 72
CONTENTS
ix
4 . T H E R O M A N O -B R IT IS H B A C K G R O U N D A N D T H E S A X O N SH O R E E vidence fo r G erm an ic in filtra tio n in R o m an B ritain 74 G erm an ic e le m e n ts in th e R o m an a rm y 75 B atavian a n d T u n g rian c o h o rts o n H a d ria n ’s Wall 75 Irreg u lar u n its o f F risians in th e th ird c e n tu ry 77 T h e n u m e r u s H n a u d ifrid i a t H o u sestead s 77 T h e cives T u ih a n ti a n d H o u sestead s W are 77 G erm an o ffic e rs b u rie d n e a r M alto n 78 Use o f G erm an n am es a n d p e rh a p s language 78 T h e ca re e r o f C arausius 79 T h e A lem an n i in B ritain 80 C ro cu s and C o n s ta n tin e ’s p ro c la m a tio n as e m p e ro r 80 F ra o m a r a n d th e A lem an n i in 3 7 2 80 E v id en ce f o r th e ir po ssib le s e ttle m e n t in Y o rk sh ire 81 G erm an ic a n d late R o m a n arch aeo lo g ical sites in L in co ln sh ire 82 T h e e sta b lish m e n t a n d significance o f th e L itu s S a x o n ic u m 83 E vidence o f A m m ian u s a n d th e N o titia 84 A rch aeo lo g ical trac es o f early G e rm a n ic s e ttle m e n t o n th e L itu s 85 R o m an o -S ax o n p o tte r y a n d its d is trib u tio n in e a ste rn B rita in 89 R easo n s fo r its ab sen ce s o u th o f th e T h am es 91 S o m e d istin ctiv e ex a m p le s o f R o m an o -S ax o n p o tte r y in Essex 93 C o n d itio n s o n th e S ax o n S h o re illu s tra te d a t Venta Ic e n o ru m 97 T he C aisto r-b y -N o rw ich a n d M arkshall A n g lo -S ax o n ce m eteries 98 E arly G e rm a n ic ce m eteries a t o th e r R o m a n to w n s o n th e S axon S h o re 101 5. SAXONS, ANGLES, AND J U T E S ON THE SAXON SHORE G en eral use o f th e te rm S ax o n fo r sea-raiders in L atin a n d C eltic sources W hy d id B ritain b e c o m e E n g land a n d n o t S a x o n y ? G ro w in g d o m in a n c e o f A nglian a n d re la te d p e o p le s in n o rth G erm an y D ating o f early A n g lian s e ttle m e n t in B ritain E vidence o f p o tte r y a n d m etal-w o rk T h e a p p e a l to th e A ngles in A n g lo -S a x o n C hronicle (E ) 4 4 3 (4 4 6 ) Its po ssib le re la tio n to th e H engist s to ry A rch aeo lo g ical ev id en ce fo r A nglian p e n e tra tio n o f th e eastern M idlands D is trib u tio n o f th e g reat c re m a tio n c e m e te rie s C o n tra st w ith th e s itu a tio n in th e so u th -e a st S ignificance o f m ix ed ce m eteries T h e S ax o n in h u m a tio n areas in th e so u th -e a st R elatio n to earlier s e ttle m e n t o f th e S ax o n S h o re
104 106 107 108 108 109 109 110 111 111 112 113 113
X
CONTENTS
P ro b lem o f th e J u t e s B ede's evid en ce T h e ir close re la tio n sh ip to th e A ngles A rch aeo lo g ical links w ith J u tla n d P o tte ry , b ro o c h e s, a n d b ra c te a te s T h e Q u o it B ro o ch S ty le Its re la tio n sh ip to a rtistic d e v e lo p m e n ts o n th e C o n tin e n t Its d a te in B ritain T h e s e ttle m e n t in K e n t in re la tio n to its R o m an b a c k g ro u n d R o m a n ro a d s, to w n s , a n d fo rtresses, villae regales, a n d th e la th e s S ignificance o f p lace-n am e survival F ran k ish in flu en ce o n th e c u ltu re o f K e n t T he a rg u m e n ts fo r a n d against a F ran k ish tak eo v e r T h e p o s itio n o f L o n d o n E vidence fo r its sta tu s in th is p e rio d T h e m eaning o f th e ‘su b -R o m a n tria n g le ' in re la tio n to th e S ax o n S h o re C o n tin u ity at M ucking a n d elsew here C o n d itio n s w ith in L o n d o n ’s w alls a n d th e A ld w y ch d e v e lo p m e n t S ignificance o f th e site o f S t. P au l’s E arly S ax o n p e n e tra tio n o f th e T h am es V alley Its re la tio n sh ip to th e S o u th S axon sectio n o f th e S axon S h o re G e o g ra p h y o f co astal Sussex R o m a n C h ich ester a n d A n d re d e sc e a ste r T h e early S ax o n sites a t A lfristo n a n d H igh D ow n Æ lle o f Sussex as th e first B retw ald a Possible c o n q u e sts in th e T h am es V alley a n d b e y o n d A rch aeo lo g ical evid en ce fro m S u rre y , O x o n ., a n d G lo u ce ste rsh ire T h e S ax o n S h o re a n d th e k in g d o m s o f Sussex, Essex, a n d L indsey T h e laets o f K e n t a n d th e social s tru c tu re o f East A nglia
113 114 114 115 115 116 116 117 122 123 124 126 127 128 129 130 131 132 134 134 135 135 135 135 137 138 139 140 142
6. T H E F O R M A T I O N O F W E S S E X Im p o rta n c e o f th e a c c o u n t o f W est S ax o n o rig in s in th e A n g lo S a x o n C hronicle In d ire c t re fere n ce to fo rtre sse s o f th e S ax o n S hore S ta tu s o f C erdic, a n d his fa m ily o rig in s as ea ld o rm en His n am e a n d its asso ciatio n s D ev elo p m en t o f his in d e p e n d e n t rule in H am p sh ire and W iltshire P o sitio n a t W in ch ester a n d S ilch ester A rch aeo lo g ical c o n tra s t o f H am p sh ire a n d B erkshire w ith W iltshire S ignificance o f C erdicesbeorg in re la tio n to W ansdyke W essex a fte r C erd ic D o u b ts a b o u t C y n ric T h e A n g lo -S a x o n C hronicle s to ry a n d its re la tio n to G ild as W ansdyke a n d th e R idgew ay T h e S ilch ester d y k e s a n d th e S ax o n s o f th e u p p e r T h am es V alley
144 145 146 147 149 149 151 152 153 153 154 155 156
CONTENTS A su g gested lo c a tio n fo r M ons B adonicus A m b ro siu s a n d A m esb u ry C eaw lin a n d th e S ax o n se ttle m e n t o f th e u p p e r T h am es area P o etic b a c k g ro u n d a n d a rtificial d atin g o f his sto ry T he cam paign o f W ib b an d u n 5 6 8 T he B ie d c a n fo r d cam p aig n , a ttrib u te d to 5 7 1 , p ro b a b ly m isd ated Its re la tio n t o th e fifth -c e n tu ry s e ttle m e n t o f th e u p p e r T h a m e s region T o p o g ra p h ic al sig n ifican ce o f th e place-nam es N o m e n tio n o f D o rc h e ste r T h e D eorham cam p aig n o f 5 7 7 E v id en ce fo r surviving B ritish a u th o rity in th e lo w er Severn area T h e F eth a n lea g cam p aig n o f 5 8 4 Its p o ssib le co n se q u en ces a t C u tte slo w e a n d C u d d esd o n D efeat o f C eaw lin a t W o d n esb eo rh 5 9 2 a n d h is d e a th 5 9 3 A m alg am atio n o f th e H am p sh ire a n d T h am es V alley S ax o n s Im p o rta n c e o f C eo lw u lf 5 9 7 -6 1 1 F o u n d a tio n o f th e g re a te r W essex
7.
xi 159 160 162 163 165 166 167 167 167 168 169 169 170 171 172 172 172
THE HUMBRENSES AND THE NORTH
S ignificance o f th e H u m b er b asin as an early fo cu s o f s e ttle m e n t T races o f H u m b re n sia n origins in L in d se y , D eira, a n d B ernicia A rch aeo lo g ical links b e tw e e n all th ese p eo p les B ede’s use o f th e te rm N o rth u m b ria S e ttle m e n t o f L in d sey fro m th e H u m b er W interin g h am a n d W in te rto n P o sitio n a n d sta tu s o f L in co ln P au lin u s a n d S t Paul-in-the-B ail R e la tio n o f L in d sey to th e L itu s S a x o n ic u m P o sitio n o f C aistor-on-thc-W olds, H o rn c a stle , a n d A n ca ster T h e b eg in n in g s o f M ercia S e ttle m e n t u p th e T re n t valley a n d th e Fosse Way S itu a tio n a t N ew ark a n d L eicester S e ttle m e n t o f D eira a n d th e V ale o f Y ork C o n c e n tra tio n o n a n d b elo w th e W olds b e tw e e n Brough (P etuaria) a n d M alto n Im p o rta n c e o f th e S a n c to n c e m e te ry a n d its c o n tin e n ta l c o n n e c tio n s , a n d o f p agan G o o d m a n h a m R e la tio n o f D eira to su b -R o m an a n d A nglian Y o rk Survival o f R o m a n b u ild in g s , A d jac en t c e m e te rie s S itu a tio n in n o rth Y o rk sh ire a n d D urham T h e b a ttle o f C a tra e th Ida 5 4 7 a n d th e b eg in n in g s o f A nglian B ernicia S e ttle m e n t u p th e T ees, T y n e , T w e e d , a n d C o q u e t
174 175 175 175 176 177 178 179 179 180 182 182 182 187 189 190 195 195 196 197 197 198 198
xii
CONTENTS
Y eavering a n d th e villa regalis A d G efrin N ative resistan ce to A nglo-S axon s e ttle m e n t B ritish p o e m s a n d tra d itio n s W idespread survival o f B ritish p o p u la tio n a n d in s titu tio n s
199 199 199 200
8. C H A N G E A N D D E C A Y E c o n o m ic co n se q u en ces o f th e R o m a n o c c u p a tio n o f B ritain T h e Pax R o m a n a a n d a m o n e y e c o n o m y G ro w th o f c o m m e rc e , in d u stry , a n d p o p u la tio n E x p lo ita tio n o f n a tu ra l reso u rces T im b e r, m in e ra ls, a g ric u ltu re ‘T h e G o ld e n A ge’ o f R o m a n B ritain R ising p o p u la tio n largely u n c h e c k e d b y p o litica l in se c u rity D ev elo p m en t o f S ax o n S h o re as stim u lu s to e c o n o m ic g ro w th O verseas tra d e a n d G erm an im m ig ratio n C learance o f w o o d lan d s: use o f tim b e r fo r building and c e n tra l h e a tin g E x p an sio n o f a g ric u ltu re fo r th e arm y a n d th e new u rb a n so c ie ty In d u stria l m a n u fa c tu re s, p o tte r y , te x tile s, m etals E x p o rt tra d e in th e f o u r th c e n tu ry B ritain fo r a tim e free fro m co n se q u en ces o f p o litic a l ch a o s L ate fo u rth -c e n tu ry R o m a n p o litica l collapse B reakdow n o f m o n e y e c o n o m y Increasing b a rb a ria n in tru sio n E vidence o f G ildas fo r e ffe c ts o n to w n s a n d villas U rb a n a n d ru ra l in s titu tio n s w re ck ed d u rin g fifth c e n tu ry D isap p earan ce o f u rb a n C h ristia n ity so o n a fte r G e rn ia n u s’ visits 4 2 9 , 4 4 7 D isap p earan ce o f c o u n tr y house life a n d villae sta te s Loss o r c o rru p tio n o f place-nam es show s lack o f m easures fo r reg u latin g b a rb a ria n s e ttle m e n t C areer o f A m b ro siu s re m e m b e red as e x c e p tio n a l am o n g B ritish lead ers N a tu re o f u rb a n c o n tin u ity in th e so u th -e a st C o n tra s t w ith s itu a tio n in w est. S ocial a n d p o litic a l significance o f G ild as’ ty r a n ts E vidence o f C ad b u ry V o rtig e rn ’s fo rtre ss b u ild in g Survival o f B ritish p o p u la tio n w ith so m e p o litic a l, religious, a n d social s tru c tu re in w est a n d n o r th Its d isa p p e a ra n c e a lm o st w ith o u t c u ltu ra l tra c e in th e new G erm an ic so ciety o f s o u th , ea st, a n d c e n tra l E ngland
202 202 202 203 203 204 204 204 204 205 205 205 205 206 206 206 206 208 208 209 210 211 212 213 215 215 216 216 217 217
CONTENTS
Xlll
APPENDICES I
(a) T h e C o n q u est o f K e n t, as re c o rd e d in th e A nglo-S axon C h ro n icle (A) (b )T h e C o n q u est o f S u ssex, as re c o rd e d in th e A nglo-S axon C h ro n ic le II T h e W est S ax o n A n n als fro m th e A n g lo -S ax o n C h ro n icle III T h e d a te o f th e ‘O bsessio M o n tis B ad o n ici’
220 221 222
BIBLIOGRAPHY
224
INDEX
237
220
FIGURES Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig.
1 2 3 4 5 6 7 8 9 10 11
D ev elo p m en t o f th e cru c ifo rm b ro o c h C ircu lar b ro o c h e s o f various ty p e s E nglish a n d c o n tin e n ta l p o tte r y o f A nglian ty p e s J u tis h p o tte r y S ax o n S te h e n d e B o gen p o tte r y S a x o n B u c k e lu rn e n R o m a n o -S a x o n p o tte r y R o m an o -S ax o n u rn fro m B illericay, Essex R o m an o -S ax o n u rn fro m C h elm sfo rd , E ssex T h e Q u o it B rooch S ty le E ngraved b ro n z e p la te
58 60 65 67 70 71 90 94 95 118 120
SOURCES A c k n o w le d g e m e n ts a re d u e fo r figures ta k e n fro m th e fo llo w in g sources: Fig. 1: Fig. 2 g • h : A berg, T he A n g lo -S a x o n s in E n g la n d (U ppsala, 1 926) Fig. 2 a - / : W elch, E arly A n g lo -S a x o n S u sse x , B A R 112 (1 9 8 3 ); c b y c o u rte s y o f W orthing M useum a n d A rt G a lle ry , d ra w in g b y Mrs P. C lark e; o th e r d raw in g s b y N. G riffiths Figs. 3 , 5 a n d 6: M y res, in P B A 5 6 (1 9 7 0 ) Fig. 4 : M yres, C orpus o f A n g lo -S a x o n P o tte r y o f th e Pagan Period (C am b rid g e, 1977) Figs. 7 -9 : R o b e rts , R o m a n o -S a x o n P o tte r y , B A R 106 (1 9 8 2 ) Fig. 10 a - f : H aw k es, in A rch a eo lo g ia 98 (1 9 6 1 ); b d ra w n b y C. O. W aterh o u se a n d re p ro d u c e d b y p erm issio n o f th e T ru stee s o f th e B ritish M useum ; o th e r d raw in g s b y S. C . H aw kes g - h : W elch, o p . c i t .; d raw in g s b y N . G riffith s Fig. 11: E vison, F ifth -C e n tu ry In va sio n s S o u th o f th e T h a m es (1 9 6 5 ); a fte r R ig o llo t, p i. XII
MAPS 1. T h e c o n tin e n ta l b a c k g ro u n d t o th e English s e ttle m e n ts
xx ix
2 . T h e S ax o n S h o re a n d its h in te rla n d
xxx
3. T h e g ro w th o f W essex 4 . T h e H u m b ren ses
xx x i x x x ii
5. D is trib u tio n o f c o m p a ra b le p o tte r y in J u tla n d and so u th -e a st E n g lan d
68
6. D is trib u tio n o f R o m an o -S ax o n p o tte r y
88
ABBREVIATIONS A n t. J . A r c h . A e l. A rch . J .
T h e A n tiq u a r ie s J o u r n a l A r c h a e o lo g ia A e lia n a T h e A r c h a e o lo g ic a l J o u r n a l
BAR
B ritish A rchaeological R e p o rt
B r it. CS EHR
B r i ta n n i a C a r t u l a r i u m S a x o n i c u m , e d . W. d e G . B irch E n g li s h H i s t o r i c a l R e v i e w
EP-N S
English P lace-N am e S ociety B ede, H is to r ia E c c le s ia s tic a N en n iu s, H i s t o r i a B r i t t o n u m
HE H is t. B r itt. J E P -N S M ed. A rch . MGH ------- ( A A ) C h r o n . m i n . -------- ( A A ) E p . N o t . D ig . O c c . P a n . L a t. V e t. PBA R IB
J o u r n a l o f t h e E n g l i s h P la c e - N a m e S o c i e t y M e d ie v a l A r c h a e o lo g y M o n u m e n ta G e r m a n ia e H is to r ic a --------( A u c t o r e s A n t i q u i s s i m i ) C h r o n ic a m in o r a -------- E p i s t o l a e N o t i t i a D ig n ita tu m O c c id e n ta lis P a n e g y r i c i L a t i n i V e te r e s P r o c e e d in g s o f th e B r itis h A c a d e m y T h e R o m a n I n s c r i p t i o n s o f B r i t a i n , R . G.
C o llin g w o o d a n d R . P. W right TRH S
T r a n s a c tio n s o f th e R o y a l H is to r ic a l S o c ie ty
B I B L IO G RA PH IC A L NOTE F o r b o o k s c ite d in fo o tn o te s a n d in th e B iblio g rap h y n o p lace o f p u b lic a tio n is given fo r th o se p u b lish e d in L o n d o n , n o r fo r th e series B ritish A rch aeo lo g ical R e p o rts, p u b lish e d in O x fo rd .
INTRODUCTION HEN Volume I o f the O xford History o f England was published in 1936 it contained, in addition to R. G. Collingwood’s full-length study o f Roman Britain, a final section o f five chapters in which I endeavoured to summarize the current state o f knowledge on the English settlem ents which took place between the collapse o f Roman rule in the fifth century and the emergence o f the AngloSaxon kingdoms in the seventh. I had been asked to contrib ute this section in order to fill a gap between Collingwood’s account o f Roman Britain in Volume I and the massive study o f Anglo-Saxon England which Professor F. M. (later Sir Frank) Stenton was to write as Volume II in the series, for neither Collingwood nor Stenton wished to concern them selves in detail w ith the dark centuries betwreen. It would have been natural for m y chapters on this obscure period to have formed rather a prologue to Anglo-Saxon England than an epilogue to Rom an Britain. Their purpose was to look forward to the future and it was in these years th a t the foundations o f England were laid in the collapsing ruins of the Roman world. But Collingwood’s Rom an Britain was to be ready for the press some years before S tenton’s AngloSaxon England and, being shorter than S ten to n ’s book was likely to prove, had more space that could be spared to accom m odate what had to be said about the English settle ments. This more or less fortuitous circumstance led to Volume I of the O xford History becoming know n to generations of history students as ‘Collingwood and Myres’. The implication which the phrase carried, that it was a cooperative product of jo in t authorship, was entirely erroneous. Collingwood himself was eager to make this clear from the start. The first sentence of his preface reads ‘This volume is not a w ork o f collabor ation’, and it was not in fact until his te x t was in final draft that I learnt that he proposed to pursue the story o f Roman Britain far beyond its generally accepted term ination in the early years o f the fifth century to include a sub-Roman epilogue in the A rthurian age. That he did so made it difficult
W
XVlll
INTRODUCTION
to provide any coherent account o f Britain in the fifth century in cither part o f the book. On the one side the collapse o f Roman Britain makes little sense w ithout a detailed study o f those barbarian forces which eventually overwhelmed it, and on the other the activities o f those forces are themselves barely intelligible except against a back ground o f the political and economic changes that were affecting the sub-Roman world in the fifth century. W ithout more effective collaboration between the m utually exclusive viewpoints provided by the tw o authors o f ‘Collingwood and Myres’ it may have been difficult for the student o f these dark centuries to form a comprehensive and just estim ate of all the factors that turned Roman Britain into Christian England. In spite o f these disadvantages, perhaps more apparent to me as part-author than to the general reader, the book was well received and had an unusually long run. Owing to the war and to Collingwood’s early death it was never substan tially revised and in the course o f over forty years b o th parts of it inevitably became very out o f date, and the whole thing is now little b etter than a period piece. A vast am ount of fresh inform ation on every aspect o f Roman Britain has becom e available during the past generation, and this has not only necessitated a radically new approach to almost every aspect of the subject as discussed by Collingwood, b u t has also made it impossible to cover it adequately except at considerably greater length. For all these reasons it became very desirable for the new Volume I o f the Oxford History by Peter Salway to be devoted to Rom an Britain only, w ithout the need to make space for a revised version o f my section on The English Settlem ents. That Salway, like Collingwood, has followed the fading ghost o f Roman Britain into final extinction in a Europe dom inated by Clovis and the expanding kingdom o f the Franks makes it even more desirable th at, in spite of the inevitable duplication o f back ground which this chronological overlap entails, my revised account o f the English settlem ents should appear separately as the start o f som ething new. Stenton’s Anglo-Saxon England was published in 1943, seven years later than ‘Collingwood and Myres’. It was in evitable that he should find it impossible to take up the story
INTRODUCTION
XIX
at the point at which I had left it, just before the coming of Christianity and the emergence o f th e Anglo-Saxon kingdoms in the seventh century-. In his preface Stenton gave the year 550 as the approxim ate point from which he proposed to begin. But in fact he found it necessary to trace the major Anglo-Saxon kingdoms back to their misty origins in the fifth century, and thus to cover again in his own way the whole period included in m y English Settlem ents. What distinguished his treatm ent from mine was his concentration on the docu m entary and place-name evidence to the virtual exclusion of the archaeological material which I had endeavoured to combine w ith it. With Salway’s Rom an Britain extending to the early sixth century and with Stenton tracing Anglo-Saxon England back into the fifth it might well be thought that the Oxford History has no need o f any independent study o f w hat might appear a non-existent gap between them . The devotion of a whole new volume, however slim, to the elucidation of m atters at least partly' considered already by b o th , certainly requires some explanation. It can be found in the fact that neither the new Rom an Britain nor the old Anglo-Saxon England provides any adequate up-to-date reassessment o f the topics to which I devoted my' chapters on the English Settle m ents in 1936. At least as much new material o f im portance has accumulated on these m atters in the last forty years as that which has required an entirely new assessment o f Roman Britain. Much o f this has been due, directly o r indirectly, on the Anglo-Saxon as on the Romano-British side, to the development o f ideas and the opening up o f new research projects whose origins can be traced back to Rom an Britain and the English Settlem ents. To take only my own studies, m y publication o f the major Anglo-Saxon cemeteries of Caistor-by-Norwich, Markshall, and Sancton (1973), AngloSaxon Pottery and the Settlem ent o f England (1969), Corpus o f Anglo-Saxon Pottery (1977), and many shorter pieces,1 all arose from thoughts originally generated while I was writing The English Settlem ents. 1 There is a full list up to 1976 in the Bibliography to my Corpus o f AngloSaxon Pottery o f the Pagan Period (Cambridge, 1977), I. p. xxxii. More recent pieces can be found in the bibliography of my writings included in V. I. Evison (ed.), Angles, Saxons, and Jutes: essays presented to J. N. L. Myres (Oxford, 1981).
XX
INTRODUCTION
Both in England and on the C ontinent very great advances have been made since 1945 in our knowledge o f the archae ology o f the Völkerwanderungszeit, and in our understanding of its historical implications. In 1936 I had little on which to build so far as the English material went except for R. A. Sm ith's purely descriptive county surveys in the Victoria County History published between 1900 and 1912, Baldwin Brown’s substantial chapters in The A rts in Early England III and IV (1915), and E. T. Leeds’ pioneer study, Archaeology o f the Anglo-Saxon Settlem ents (1913), all o f which, though containing much m atter o f basic im portance, were becoming out-dated by subsequent discoveries. None o f these scholars were professional historians, and their w ork had made little impression on historical writing. The only recent attem pt at a synthesis o f the archaeological material was contained in N. Aberg, The Anglo-Saxons in England (1926) and deali mainly with the classification o f brooches and other m etal work. On the C ontinent also most o f the im portant cemeteries in north Germany, the Low Countries, and Scandinavia wrere still unpublished, and there had been no comprehensive attem p t to base historical conclusions on the archaeological evidence since P lettke’s Ursprung und Ausbreitung der Angeln und Sachsen appeared in 1920. In particular no effective use had been made for historical purposes of the great collections o f pottery either in English o r in continental museums. It was only after 1936, and particularly after 1945, that this situation began to change on both sides o f the North Sea. Karl Waller's publication in 1938 o f the great cemetery at Galgenberg bei Cuxhaven began this process just before the war, but the appalling war-time losses o f archaeological material in the German museums stim ulated a notable drive to publish what was left as soon as conditions made such a programme possible. One unexpected consequence o f the war was the deter m ination o f archaeologists concerned with the Völkerw anderungszeit on the Continent and in England to resume and develop their com m on interests in the problems o f that age. A most useful forum for this purpose was provided by the enlightened initiative o f Karl Waller in the establishment of the Arbeitsgemeinschaft für Sachsenforschung based at
INTRODUCTION
XXI
first on his own museum at Cuxhaven and continued after his death by Albert Genrich o f Hanover. It was at Cuxhaven th at the first Sachsensymposion, inaugurating a series o f such informal gatherings, took place in 1949. Subsequent meetings held annually not only in Germany but in Holland, Belgium, the Scandinavian countries, and (three times) in England have been most useful in prom oting the grow th o f personal contacts, and the exchange o f ideas, publications, and visits between the most active scholars in these fields in all countries bordering the N orth Sea.1 Numerous publications, partly stim ulated by the Sachsensymposion, have not only made available much o f the surviving evidence from pre-war museum collections, b u t have included the results o f im p o rtan t recent excavations on settlem ent and cem etery sites such as Wijster in Holland, Feddersen Wierde near Bremer haven, and Liebenau (Hanover), which have produced material directly relevant to the Anglo-Saxon settlem ents in England. The fresh emphasis on evidence from occupationsites as distinct from cemeteries, throw ing new light on the life-style and dom estic economy in addition to the burial customs of the pagan Anglo-Saxon peoples, has been one of the most fruitful developments o f the post-war period. In 1936 almost the only village site o f this kind that had been excavated in England was th a t at Sutton C ourtenay, Oxon. (form erly Berks.) where E. T. Leeds had explored a number o f Grubenhäuser and published his discoveries in Arc ha eo logia 72 and 76 (1923 and 1927). No example was know n in this country at that time o f the more substantial halls o r longhouses which were a regular feature o f the contemporary7 Terpen, or m ound settlem ents, on the marshy coastlands of Holland and north Germany. Since then this gap in our knowledge has been largely filled not only by the recognition o f many further groups o f Grubenhäuser all over eastern England, but by the exploration o f major concentrations of such habitations and workshops associated with rectangular hall-houses at, for example, Mucking, Essex, and West Stow Heath, Suffolk. The inform ation which archaeology can now supply towards an understanding o f the pagan Saxon age is 1 More detail on the earlier stages of the post-war recovery of archaeological study in Germany and north-western Europe in general can be found in my Anglo-Saxon Pottery and the Settlem ent o f England (Oxford, 1969), 8-11.
xxu
INTRODUCTION
thus not only vastly greater in quantity b u t also very much broader in range and quality than it was before 1936. The discovery and excavation o f the ship burial at Sutton Hoo, Suffolk, in the weeks immediately preceding the o u t break o f war in 1939, was in time to receive brief m ention in Stenton’s Anglo-Saxon England in 1943. But his account was necessarily composed while the magnificent grave goods from this richest o f all princely pagan-style burials o f the Völkerwanderungszeit were still inaccessible in war-time storage. It needs substantial revision in the light o f the definitive publication o f the finds by R. L. S. Bruce-Mitford between 1976 and 1984. Nor is it only in archaeology that there have been notable developm ents in the non-literary evidence for this age. The study o f place-names was already contributing much infor m ation relevant to the English settlem ents before 1936, though it has never been easy to interpret with confidence in historical terms. In the early chapters o f Anglo-Saxon England Stenton, whose knowledge o f the subject as then understood was profound, made a much wider use of placenames than I had done. He was able to draw not only on the ever-increasing material being published in the county volumes o f the English Place-Name Society, b u t on that which he had him self assembled for his presidential addresses to the Royal Historical Society delivered annually on aspects of this subject from 1938 to 1940. Since S tenton’s time there have been developments in place-name studies due not merely to the continuing expansion o f its published raw material, but to various essays in its interpretation which may have a profound bearing on its value for historical purposes in this period. These will require fuller examination at later stages in this b o o k .1 They are m entioned here as an indication of the need for a fundam ental reassessment o f the place-name element in the basic evidence for the course and character of the English settlem ents. These are some o f the reasons which have made a quite new approach to the problems o f the English settlem ents, radically different from that which I took m ore than forty years ago, essential, if this basic phase in the history of 1 Especially pp. 37-43
INTRODUCTION
X X lll
England is to be discussed with proper reference to the present state o f knowledge. It is never easy for an author to judge after such a lapse o f time how much o f w hat he then w rote is w orth incorporating in a reassessment based on so much fresh evidence. There are aspects o f the story on which m y own views do not now differ much if at all from those which I then expressed. I have had no hesitation in retaining in such cases passages w ritten before 1936, both long and short, and even single sentences and turns o f phrase, with little or no m odification even if they appear here in a dif ferent context. But in essentials this is a new book whose general arrange m ent necessarily differs from th a t o f m y chapters in ‘Collingwood and Myres’. This is required partly to take account of necessary- changes o f emphasis caused by the differential growth of knowledge on different topics. But partly also it is needed because more has now to be said about the back ground of native society in the later days o f Roman Britain w ith which the Anglo-Saxon settlers found themselves con fronted. These m atters, as then understood, had been covered by Collingwood in his part o f Rom an Britain and the English Settlem ents, and there was no call for me to go over them again in mine. But in a separate volume such as this some account of this background must be given to achieve a proper understanding o f the circumstances in which the settlem ent took place in various parts o f the country. This may involve some duplication o f ground covered in the later chapters of Salway’s new Rom an Britain. Such duplication cannot be avoided if due weight is to be given, as it m ust, to all the evidence, mostly unrecognized in 1936, which suggests a degree of overlap and fusion between the last age o f Roman Britain and the earliest Anglo-Saxon settlem ents which was too little appreciated at that time. It is indeed mainly to allow for a proper discussion o f this overlap that a separate volume at this point in the O xford History o f England is now required. One point needs emphasis. Like all its fellow volumes in the O xford History o f England, the present book is written primarily for students o f history, n o t for specialists in archae ology, linguistics, place-name studies, o r any other related subjects. Their conclusions m ust o f course be utilized, and
XXXV
INTRODUCTION
indeed on occasion criticized, wherever it seems that they have a relevant contribution to bring to the elucidation of this phase o f English history. But the plan o f this series, and in particular its limited provision o f illustrations, has made it, for example, impossible to go deeply into the archaeological discussion o f the numerous categories o f material artefacts which form the evidence on which the conclusions of that branch o f knowledge rest. The significance o f these con clusions can only be grasped, and their meaning can only be correctly appreciated, with the visual aids which abundant illustrations can alone supply. It has therefore often been necessary to state the conclusions to which I have come on many aspects o f this time w ithout providing the reader with the detailed grounds which have led me to them or enabling him to appreciate visually much o f the evidence on which my interpretation rests. In the Bibliography I have given further references to relevant illustrated accounts of the material under discussion. But in the text I have only attem pted to describe its m ore distinctive and significant features in broad outline, and I have done this only where it has been essential to the course o f an historical argument. I have made no attem pt to provide any general guide to AngloSaxon, sub-Roman, or any other relevant category o f Dark Age antiquities. This book is intended to be a history book and nothing else. For somewhat similar reasons I have om itted all technical discussion o f the linguistic problems which arise at every turn in the search for material relevant to an understanding of this age. This is not because they are unim portant but because I am not properly equipped to handle them . The material and psychological changes which accompanied the general substitution o f a Germanic mode o f speech for one based either, as in Gaul, on Vulgar Latin or, as in the Celtic lands, on varieties o f the Brittonic tongues, must have had profound consequences for the structure o f society in all those parts o f the country where it occurred. The most obvious is the transform ation o r replacement o f place-names, and this is the only aspect o f the m atter on which I have felt com petent here and there to com m ent. But language and history in all societies m ust always be deeply intertw ined, and such traum atic changes in speech as took place in Britain
INTRODUCTION
XXV
in the centuries covered by this book should be capable of providing many clues to the corresponding changes in the population and its way o f life. But linguistic change is also a factor which directly affected the survival value o f all contem porary literary evidence. How m uch, for example, would Anglo-Saxon settlers be able to learn about Roman Britain if it was true that they encountered ‘very few people who talked any sort o f Latin at all’ in the parts which they occupied? And if ‘Vortigern, around 450, could not have understood Aneirin around 600, though Gildas, living through the first half o f the sixth century, could probably have understood b o th ',1 w hat were the chances for the meaningful survival o f any historical, social, o r economic inform ation recorded at this time in the lands where Brittonic speech prevailed? Considerations o f this kind are clearly of direct relevance to the subject m atter o f this book, b u t their discussion cannot be profitably undertaken by any but those possessed o f a high degree o f linguistic expertise. It may perhaps be thought by some readers that I have placed too much emphasis on m atters, whether archaeological or historical, th a t directly arise from the topographical distribution o f the early Anglo-Saxon settlem ents. To stress these aspects o f the period has, indeed, been my deliberate intention. I have always felt th a t an essential prerequisite to any understanding o f this age must lie in an attem pt to see the countryside, and the collapsing ruins o f Roman Britain which it contained, from the same point o f view as that which presented itself to the newcomers from the Continent. Such an intim ate and revealing understanding o f the con tem porary topography can only be achieved by detailed personal inspection undertaken on foot or horseback, o r by boat, for in the fifth century no other means of transport were available. Only in these ways can the extent, varying quality, and interrelationships o f the .different parts o f low land Britain be appreciated by the m odem student in terms which correspond, as near as may be, to the impression they must have made on immigrant folk from the C ontinent in early Anglo-Saxon times. 1 K. H. Jackson, Language and History in Early Britain (Edinburgh, 1953), 261 and 690; and sec my review o f this important book, EHR 70 (1955), 630-3.
XXVI
INTRODUCTION
To this end I devoted much o f my leisure, as an under graduate and a young college tu to r in the tw enty years before 1939, to extensive travels on foot through the relevant parts o f Roman Britain and the principal areas o f early AngloSaxon settlem ent. Thus I walked the whole line o f H adrian’s Wall and examined the sites o f all the late Roman signal stations on the Yorkshire coast between the Humber and the Tees. I visited all the Saxon Shore forts o f wrhÍGh traces survive, betw een Brancaster on the n o rth Norfolk coast and Pevensey in Sussex. I walked the whole length o f the Fosse Way from Devonshire to Lincoln, and thence north on the Roman road to the Humber crossing at Brough (Petuaria), and on past Sancton and Goodm anham to Malton and the Vale o f Pickering. In Wessex, I made myself familiar on num erous occasions with the chalklands o f Hampshire and Wiltshire, and came to know very well their main links w ith the upper Thames Valley, by walking the O xford to Winchester route, the Berkshire ridgeway across the Wansdyke frontier and the Vale o f Pewsey on to Salisbury Plain, and the Roman roads from Cirencester to Silchester and Win chester. In the south-east, I walked the so-called Pilgrims' Way along the N orth Downs from Winchester to Canterbury and the tracks along the South Downs from the Cuckmere Valley towards Chichester. I crossed the Weald by the Roman Stane Street from Chichester to the southern approaches of London. N orth and east o f the upper Thames basin, I walked the Akeman Street from Cirencester to St. Albans and most of the Icknield Way route along the Chilterns from the Goring gap towards East Anglia and the Wash. I once had ambitions to penetrate Britain by boat up all the main water-ways on the east and south coasts. It seemed to me im portant to obtain a visual impression o f how the land looks as one approaches the principal harbours and estuaries from the sea. For this purpose I studied both the lower and the upper reaches o f the Thames from below Greenwich past London to O xford and beyond, and attem pted to visualize hoW the valley would have appeared before the upper part o f the river was artificially controlled by locks. I made a fascinating trip up the Humber and the Yorkshire Ouse as far as York, and noted with interest how the principal tributaries from the west all seem to join the main river
INTRODUCTION
X X Vil
facing upstream ; their m ouths are thus not at all obvious to inquisitive mariners sailing by, especially when deeply shrouded in reed beds. I also know the Yare Valley from Great Yarm outh (and its earlier m outh, now blocked by shingle, at Rom an Caister) through Breydon Water to Norwich and nearby Caistor ( Venta Icenorum) on its tributary, the Tas. But life has not been long enough to carry out as much of this programme as I had hoped. I have never found an oppor tunity to make such a water-borne study o f the T rent, or of the im portant Anglo-Saxon routes into the Midlands provided by the valleys o f the Welland, the Nene, and the Bedfordshire Ouse th a t flow through the Fens into the Wash. Even so, I can claim to have visited by these old-fashioned means a great m ajority o f the places in this country whose names appear in the index to this book, and I retain a vivid visual memory o f the present appearance o f a very large num ber o f them and their surroundings. And the fact th at I have studied and drawn Anglo-Saxon pottery in some ninety English museums and private collections d o tted over all the face of the land5 has also contributed greatly to such familiarity as I have with the landscape, for on these visits I have generally found time to look at their setting in the surrounding countryside. I cannot claim a comparable knowledge o f the topography of the continental homelands. But I have worked in m ost of the principal museums in Belgium, Holland, north Germany, and Scandinavia. I have observed from the air, and to a lesser degree by rail and on foot, the lower reaches o f the Weser and Elbe valleys, the neighbourhood o f Kiel and Schleswig, and much o f Denmark. A slow stopping train from Bremen to Cuxhaven once provided me with fascinating glimpses of the numerous Terpen, many still carrying m odern villages, that are scattered over the marshy coastlands between the Elbe and Weser estuaries, from which so many o f our Saxon forbears must have come. Several o f m y own friends, and many other contemporaries of m y generation, made more extensive and adventurous peram bulations on foot, b o th in this country and abroad, Details are listed in my Corpus, I. p. xiii.
X X V lll
INTRODUCTION
often with m ore ambitious objectives than I set myself. I have mentioned w hat I personally attem pted in this field only to explain how it is that I have come to think of the historical and archaeological problems o f this age as set always against the immensely informative background o f the English landscape. W ithout some personal appreciation o f the relevant features o f that landscape, and o f its fading RomanoBritish contents in the fifth and sixth centuries, no one can hope to achieve an adequate understanding o f the course and character o f the English settlements. It is, however, unfortunately true that such an under standing is now far harder to obtain by the pedestrian means which it was natural for me to adopt over sixty years ago. Long-distance walking on the older main roads, and even the minor roads, o f England has been rendered almost intolerable by the proliferation o f m otor traffic. The building o f m otor ways, often sited across rather than along, the natural grain of the country, and the construction o f bypasses, specifically designed to divert travellers away from the age-old centres of population, have made it much more difficult to com prehend, in the old way, the traditional patterns o f human settlem ent. My generation o f historians was perhaps the last for whom it was comparatively easy to identify and appreciate the meaning o f these traditional patterns, unhampered by the physical distortions now imposed on them by the require ments o f m otor transport. C h rist C h u rch , O x fo r d Ju n e 1985
J . N. L. M Y R E S
M a p 1 . T h e c o n t i n e n t a l b a c k g r o u n d t o t h e E n g li s h s e t t l e m e n t s
M a p 2 . T h e S a x o n S h o r e a n d its h in te r la n d
M ap 3 . T h e g r o w th o f W essex
M ap 4 . T h e H u m b re n se s
1
T H E N A T U R E OF T H E E V I D E N C E : I. T H E L I T E R A R Y S O U R C E S period o f some tw o centuries which lies between the collapse o f Roman government in Britain and the arrival of St. Augustine in A D 597 has long been recognized as the most difficult and obscure in the history of this country. Between Rom an Britain and Christian England there is indeed a great gulf fixed, a void o f confusion which remains a standing challenge to historical inquiry. Within these two centuries changes m ore profound and far-reaching than in any other corresponding period took place: these changes modified the physical character o f the people, altered the fundam ental structure o f the language, laid the basis o f many of our institutions, and made possible an econom ic ex ploitation o f natural resources on a scale scarcely attem pted in prehistoric o r even in Roman times. And as the result of this dom estic exploitation came first the possibility and then the problem of a population expanding towards the limits of subsistence, and becoming in its turn the centre o f expansion overseas. Many other and later factors have determ ined the direction, the character, and the extent o f that expansion, b u t it is not too fantastic to believe that its beginnings lay in this period, in the silent and strenuous conversion o f the abandoned arable land and remaining forests o f Roman Britain into the cultivated fields and profitable swine pasture of countless Anglo-Saxon communities. At the time, o f course, nothing could have been further from m en’s minds than such a conception o f the meaning of events. To those brought up in the Roman world the forces of civilization appeared to be giving way to barbarism, and, in so far as the civilized attem pted consciously to appreciate the causes of their troubles, they not unnaturally attributed them to divine vengeance on one another’s sins. Nor were the illiterate barbarians themselves, so far as we can penetrate to
T
he
2
I. T H E L I T E R A R Y S O U R C E S
the thoughts which they were themselves unable to express in perm anent form, conscious in any way o f their destiny. Long after this period was over they were still giving names to places in the forests and fens which show that they regarded them and their contents with superstitious terror, as areas peopled w ith all the supernatural forces o f evil. That n o t withstanding they continued to fell the forests and to drain the fens is a fact whose significance only the later historian can appreciate. All that was apparent by the end o f the sixth century was the destruction o f the wrhole fabric of Roman imperialism in Britain, the disappearance o f its civil and military adm inistration and of many o f the arts of life. Instead there was being built up a group o f precariously founded barbarian kingdoms whose rulers were still living largely on the spoil o f their neighbours, even if many of their dependants, along with the surviving rem nants o f the British population, were already slowly settling down to the cease less routine o f subsistence agriculture. The purpose o f this book is to inquire by w hat stages these changes were brought about and how deep they penetrated into the structure o f society. In the past, and even in very recent times, the most diverse answers have been given to these questions. Some historians have regarded the destruc tion of the Romano-British world and o f the native popu lation as virtually total. Others have treated the Anglo-Saxon intruders as imposing little more than the superficial veneer of a new language and a conquering elite on a British popu lation that remained basically in place, however economically depressed and culturally deprived. How does it come about that such widely different interpretations are possible, and that the changes which certainly occurred should be so diffi cult to understand? The best way to find an answer to this question is to look at the sort o f evidence on which our knowledge o f the period rests, and to contrast it with that on which we rely for an understanding o f the times before and after these dark centuries. O ur knowledge o f Roman Britain is based on an unequal blend o f two main groups o f evidence. Since Britain became part o f the Roman empire, the broad outlines o f its history arc known from the historians who described at dif ferent periods the story o f that empire. But since Britain was
TYPES OF EVIDENCE
3
rem ote from the M editerranean lands which were the centres of Roman political life, and since the works o f few Roman historians have survived, and not one o f them after Julius Caesar had certainly visited Britain, these literary sources provide only the barest outline o f a story, lacking in all the detail which only local sources o f knowledge and under standing could provide. Fortunately, however, that story can be supplem ented from the results o f archaeological survey and excavation. The character o f Roman remains is such that a maximum o f historical inform ation can be derived from them. There are three principal reasons for this. Firstly, Roman remains are abundant, easy to recognize, and readily classifiable, so that the evidence derived from individual sites, w hether forts, tow ns, villas, o r villages, can be constantly checked by reference to other examples o f their own and other classes. Secondly, Roman provincial culture was very well supplied with material goods, in such durable materials as bone, metal, glass, and p o ttery ; many categories o f these goods, being more o r less mass-produced, are closely datable. Thirdly, it was a literate society operating a money econom y, and the occurrence, in some q uantity, o f coins and inscrip tions provides the basis on which such close dating, whether of individual groups o f objects or o f the buildings and other features in association with which they occur, can rest. Although much more remains to be learnt, and there are still many points o f uncertainty in the precise dating o f events, a broad and convincing picture o f the political, social, and economic history o f Britain under Roman rule has been gradually built up from these sources. If this evidence is compared with th a t upon which our knowledge o f Anglo-Saxon England from the seventh century onwards depends, it is clear th a t a great change o f emphasis has occurred. There has been a notable improvement in the quality and quantity of the literary sources, m ost o f which were w ritten in England and often give detailed and con tem porary accounts o f the political history and the activities of the most influential personalities, sometimes w ritten from first-hand inform ation. On the social and institutional side the evidence o f chroniclers and biographers is supplemented by administrative docum ents such as the law codes o f the kings and the land charters which they granted to churches
4
I. T H E L I T E R A R Y S O U R C E S
and laymen. Archaeological evidence, apart from that comprised in surviving buildings, is not on the whole so informative for this age as it is for Roman times. But in recent years excavation in the lowest medieval levels of such towns and ports as York, Winchester, London, and Southam pton has revealed much evidence for the renewal of urban life and for the economic developments in trade and industry on which it rested. It can thus be seen that both for Roman Britain and for later Anglo-Saxon England first hand sources, whether literary or m aterial, are available in sufficient volume to form a reasonably firm basis for the writing o f history. But for the two intervening centuries to which this volume is devoted there are no contem porary sources o f either kind approaching in value those available for the periods before and after. Both the literary and the archaeological evidence, moreover, are rendered more difficult o f interpretation as a whole by their division into two unrelated groups representing the traditions and the material remains o f invaders and invaded. On the literary side the Anglo-Saxons have left no contem porary evidence for the settlem ent at all, for the very good reason th a t the invaders were illiterate. It was not until after their conversion to Christianity that they began to record in writing some o f the oral traditions that had been handed down from their pagan ancestors. A t first these mostly concerned the exploits o f their kings, remembered in the form o f heroic poetry and authenticated by attachm ent to the genealogical fram ework o f royal pedigrees. When, for example, the Anglo-Saxon Chronicle was compiled in its present form in the reign o f King Alfred late in the ninth century, it incorporated three series o f annals attached to dates in the fifth and sixth centuries which offer the outline of a traditional story covering the settlem ent o f Kent, Sussex, and Wessex. Claims have been made, no doubt rightly, on philological and other grounds that certain o f these annals go back in w ritten form to the seventh century, and thus almost to the living memory’ o f m en who took part in the last of these events shortly before the spread of literacy made w ritten record widely possible. Even so, the annalistic format adopted for the Chronicle must have required the attachm ent of dates, which at best are traditional and at worst arbitrary,
THE A N G L O -S A X O N CHRONICLE
5
to incidents, and groups o f incidents, culled from the sagas of kings and heroes, an art-form notoriously indifferent to precise chronology. The point is well illustrated by the fact that the earliest West Saxon annals appear to contain the remains o f two ver sions of w hat was essentially the same story whose incidents are now separated by gaps o f nineteen years.1 This, while it should make one wary o f attaching too much im portance to the exact dates recorded for individual incidents, greatly strengthens the case for a very high antiquity in the stories themselves. Such an unconscious com bination o f two versions in Alfred’s chronicle show-s how much had been done long before his time in the rather unintelligent blending of w hat were evidently very ancient tales. More will need to be said later on about the value o f these tales as historical evidence when the settlem ent o f Kent, Sussex, and Wessex is considered in detail. Here it is only necessary to emphasize that their im portance is not greatly diminished by doubts on the reliability o f the precise dates to which they are attached. For this reason there is no need to pay undue attention to the signs o f what has been regarded in the past as a deliberately artificial arrangement of some events at intervals o f four and eight years, an arrange m ent which has been thought to derive from the sequence of leap years noted in Easter tables. Even if this was so—and there has been a recent tendency among historians to mini mize or even deny this kind o f apparent artificiality in the dating12*7—it would do no more than illustrate what is already 1 S ee A p p e n d ix II a n d K. H a rriso n , F ra m e w o rk o f A n g lo -S a x o n H isto ry (C a m b rid g e, 1 9 7 6 ), 1 2 7 . T h e g ap o f n in e te e n y e ars is lik e ly to have arisen fro m th e tw o v e rsio n s b e in g d a te d w ith in tw o successive n in e te e n -y e a r E a s te r c y c le s, o n e r u n n in g fro m 4 9 4 to 5 1 2 , a n d th e o th e r fro m 5 1 3 t o 5 3 1 . T h e d u p lic a tio n is n o t o b v io u s fro m t h e A n g lo -S ax o n M SS o f th e C h ro n icle a lo n e , b u t it b e c o m e s c le a r i f th e a n n a ls a re c o m p a re d w ith Æ th e lw e a rd ’s L a tin tra n s la tio n . T h e M S h e used m u s t have h a d a n a d d itio n a l e n try a t 5 0 0 re c o rd in g th e c o n q u e s t o f W essex w hich c o rre s p o n d s c lo se ly t o th e C h ro n icle e n tr y o f 5 1 9 d a tin g th e y e a r w h e n C e rd ic a n d C y n ric ‘t o o k th e k in g d o m ’. It is v e ry lik ely th a t th is 5 0 0 e n tr y in Æ th e lw e a rd ’s M S w as d e lib e ra te ly su p p re sse d b y th e c o m p ile rs o f th e e x istin g A n glo-S ax o n t e x t , w h o se n sed th a t it m u st b e a d u p lic a te o f th a t fo r 5 1 9 . See a lso K . H a rriso n , ‘E arly W essex a n n a ls in th e A n g lo -S ax o n C h ro n ic le ’, E H R 8 6 (1 9 7 1 ) , 5 2 7 -3 3 . 7 K . H a rris o n , F r a m e w o r k , 1 4 7 - 5 0 ; see a lso C. W. J o n e s , S a in ts ' L iv es a n d C h ro n icles in E a rly E n g la n d (N e w Y o rk , 1 9 4 7 ), 1 9 1 , w h o n o te s th a t th e se q u en c e o f d a te s a t in te rv a ls o f fo u r a n d e ig h t y e ars d o e s n o t b e a r a n y fix ed re la tio n sh ip t o t h a t o f le a p years.
I. T H E L I T E R A R Y S O U R C E S
6
obvious. In an illiterate age when past events are handed down mainly by oral tradition in heroic verse, their exact dating is o f no contem porary consequence, even if means were available to ascertain it. At best a rough sequence of memorable incidents may be formed by their association w ith personalities whose position in or relation to a royal pedigree was known. These difficulties in the establishment of exact dates for events before living memory were familiar enough to the Venerable Bede. His profound interest in chronological systems made him eager to fit some o f these tales into a correct time sequence in the early part o f his Ecclesiastical History o f the English N ation, finished in 731. But when he set himself to tabulate in the chronological summary at the end o f his w ork1 the most memorable dated events o f the period with which this volume is concerned, he found exactly nine. These included two eclipses o f the sun (useful as pegs on which occurrences otherwise undated can sometimes be hung), three m atters o f ecclesiastical interest in Ireland, Scotland, and Rome, one which connects the sack of Rome by Alaric in 410 with the end o f Roman rule in Britain, and tw o which give imprecise time-signals for the arrival of the Angles in this country7.2 He notes only one specific event, and that not w ithout ambiguity, which is directly relevant to the chronology o f the English settlem ents.3 For w hat he evidently took to be the crucial period o f early settlem ent between 449 and 538 there were apparently in Bede’s con sidered judgem ent no datable occurrences to be recorded at all. In estim ating the part which precise dates can be expected to play in the story' it is sensible to recall the salutary scepticism o f Bede. N or is this lack o f reliable dates in the traditions of the invaders effectively replaced by the occasional references to 1
HE
V. 2 4 .
7 ‘T h e A n g les c a m e to B rita in in th e reig n o f M a rcia n a n d V a le n tin ia n w ho re ig n e d sev en y e a rs fro m 4 4 9 . . . . W hen th e A u g u stin ia n m issio n re a c h e d B rita in in 5 9 7 , it w as a b o u t 1 5 0 y e ars a fte r th e arriv al o f th e A ngles in B rita in .’ s ‘In 5 4 7 Id a b e g an t o reig n fro m w h o m th e ro y a l s to c k o f th e N o rth u m b ria n s is d e riv e d , a n d re m a in e d K in g fo r tw elv e y e a r s .’ B ede d o e s n o t te ll u s w h e th e r h e t o o k th is e v en t t o re p re s e n t th e first A n g lian in c u rs io n s in to B e rn ic ia o r w h e th e r it in te re s te d h im o n ly as a fix ed p o in t in th e g en ealo g y o f t h e N o rth u m b r ia n ro y a l fam ily .
DIFFICULTIES OF DATING
7
the progress o f the English settlers which occur either in continental sources or in the works o f native writers from the parts of Britain where some degree o f literacy was maintained in ecclesiastical circles beyond the reach o f the earlier barbarian intrusions. As already m entioned, Bede had infor m ation which linked the collapse o f Rom an rule in Britain with Alaric’s sack o f Rome in 410, and it was known from the sixth-century Byzantine historian Zosimus,1 who had good earlier sources, that the em peror Honorius had formally declined a request for Roman military assistance addressed to him by the local urban authorities in Britain in circum stances which implied th a t the provincial adm inistration was no longer operative. Zosimus goes so far as to say that the Britons ‘fighting for themselves freed their cities from the attacking barbarians’, and after repudiating external authority ‘were living on their own w ithout obeying Roman laws’.2 Procopius, also writing in the sixth century in reference to these same events, states clearly that from this tim e the Romans never recovered Britain which continued to be ruled by ‘ty ran ts’, the usual term for self-styled local emperors.3 As the fifth century progressed, references in continental sources to events in Britain, w'hether closely datable o r not, become rarer and less reliable. Much has been made o f the entries in tw o related South Gaulish chronicles, both o f which probably used near-contem porary inform ation. Both purport to record the com pletion o f the Saxon conquest of Britain in the decade or so before 450.4 These are certainly exagger ated claims, not least because there is near-contemporary 1 Z o sim u s, e d . L. M en d elsso h n (1 8 8 7 ), vi. 10. 3 Ib id . v i. 5 . 3 P ro c o p iu s, D e B ell. V a n d ., c d . J . H a u ry ( 1 9 0 5 - 1 3 ) , I. ii. 38. 4 M G H ( A A ) ix . 661 s u b a n n o 4 3 8 /9 : B rita n n ia e a R o m a n is a m issa e in d ic io n e m S a x o n u m c e d u n t; ib id . 6 6 0 s u b a n n o 4 4 1 /2 (b u t p ro b a b ly f o r 4 4 5 /6 ) : B rita n n ia e u s q u e a d h o c te m p u s variis c la d ib u s e v e n tib u s q u e la ta e in d ic io n e m S a x o n u m re d ig u n tu r. I t h a s b e e n su g g e ste d w ith so m e p la u s ib ility t h a t w h ile it is v e ry u n lik e ly th a t c h ro n ic le rs w ritin g in th e s o u th o f F ra n c e c o u ld have h a d a c c u ra te in fo rm a tio n o n th e e ffe c tiv e sp re a d o f A n g lo -S ax o n a u th o r ity o v e r B rita in as a w h o le in th e 4 4 0 s , th e y c o u ld h a v e k n o w n fro m M e d ite rra n e a n m e rc h a n ts th e e x te n t t o w h ic h t h e tra d e ro u te s h a d b e e n d is ru p te d b y p ira c y a n d even b y th e o c c u p a tio n o f e sse n tia l p o r ts a n d stra te g ic c o a s tla n d s , fo r e x a m p le th o se h ith e r to c o n tr o lle d f o r th e E m p ire b y t h e f o r ts o f th e S a x o n S h o re . T h is c o u ld easily have given a n e x a g g e ra te d view o f th e e x te n t o f S a x o n p o w e r. H a rriso n , F ra m ew o rk 2 7 . I a m n o t c o n v in c e d b y a tte m p ts re c e n tly m a d e to d isc re d it th e a u th o r ity o f th e s e a n n a ls re la tin g to B rita in as la te r in te rp o la tio n s : se c M. M iller, B rit. 9 (1 9 7 8 ), 3 1 5 -1 8 .
8
I. T H E L I T E R A R Y S O U R C E S
evidence from C onstantius’ Vita Sancti Germani that the saint was able to visit Britain as late as the 440s for dis cussions on theological issues with the sub-Roman authorities. Although the talks were interrupted by Saxon raiders their operations are not portrayed as anything like a serious military take-over. On the other hand such contacts between Britain and the C ontinent were evidently becoming increas ingly precarious at this time. They must have broken down almost completely soon after, except in the south-west. From these parts relations were maintained by substantial, move m ents of the British population across the Channel which were probably already beginning the process which soon converted Rom an Armorica into B rittany.1 The last dated contact between sub-Roman Britain and the authorities in Roman Gaul is not recorded in a conti nental source. It is m entioned in a docum ent o f which Gildas, a British priest writing in the second quarter o f the sixth century in these western parts, has preserved a paraphrase in his De Excidio et conquestu Britanniae. It was apparently a formal appeal for aid against barbarian inroads addressed to Aetius, ruler o f Roman Gaul, and dated betw een 446 and 453 by a reference to his third consulship. Its failure seems to have led quickly to a settlem ent o f friendly barbarians as foederati in the eastern part o f Britain, and this in turn acquired significance for later writers as the m om ent of the A dventus Saxonum in Britanniam.12 Gildas had no means o f dating the third consulship of Aetius, whom he calls Agitius,3 and probably had little idea 1 It m u s t b e sig n ific a n t t h a t w h e re a s t h e B ritish C h u rc h a c c e p te d t h e ch an g e m a d e in 4 5 5 in c a lc u la tin g th e d a te o f E a s te r in th e w e s t, it d id n o t a p p a re n tly receiv e an y la te r a lte ra tio n s th a t w ere a d o p te d in G a u l o n th is c o m p lic a te d m a tte r . A M a n su e tu s 'B ish o p o f th e B rito n s ’ is a m o n g th e sig n a to rie s o f th e A cts o f a c o n tin e n ta l C o u n cil in 4 6 1 , w h ic h su g g ests th a t th e m ig ra tio n w as th e n re c en t a n d th e n e w c o m e rs h a d n o t y e t b e e n a b s o rb e d in to re g u la r d io c e se s. G ild a s, e d . M . W in te rb o tto m (1 9 7 8 ) , 150, re fe rrin g t o M an si, C o u n cils 7 , 9 4 1 . : In w h a t fo llo w s o n th e A d v e n tu s I ow e m u c h t o th e p e n e tra tin g s tu d y o f G ildas S a p ie n s b y C . E . S te v e n s in E H R 5 6 (1 9 4 1 ) , 3 5 3 - 7 3 . H is id eas a rc d is c u ssed a n d d e v e lo p e d in m y a rtic le 'T h e A d v e n tu s S a x o n u m ’ in W . F . G rim es (c d .), A s p e c ts o f A rc h a e o lo g y in B rita in a n d B e y o n d (1 9 5 1 ), 2 2 1 - 4 1 . 3 I t is te m p lin g t o s u p p o s e t h a t th e sp ellin g A g itiu s m a y be d u e t o a h a lf c o n sc io u s c o n fu s io n b e tw e e n A e tiu s a n d A egidius w h o d o m in a te d p a rts o f n o r th e r n G a u l b e tw e e n 4 5 7 a n d 4 6 2 . B u t th e d o c u m e n t p a ra p h ra s e d b y G ildas c a n n o t have b e e n o rig in a lly a d d re sse d t o A eg id iu s, as s u p p o s e d b y S . J o h n s o n , L a te r R o m a n B rita in (1 9 8 0 ) , 112, fo r h e w a s n ever a co n su l.
T H E A D V E N T U S SA X O N U M
9
who he was, o r w hat the circumstances were in which the letter was addressed to him. He was dimly aware that the stages in the breakdow n o f Rom an rule had been marked by a series o f appeals—in Celtic fashion he thought o f them as a triad—culminating in the events around 410 already dis cussed. After this he believed that the Britons had managed for a while on their own to cope with barbarian inroads during w hat historians have term ed a ‘prosperity period’. It was thus natural for Gildas to attach the letter to Aetius to the events surrounding the failure o f the last appeal known to him, thereby placing it before rather than after the ‘pros perity period’ which must belong in the second and third decades of the fifth century. Bede, who used Gildas’ b o o k ,1 knew by 731 that the third consulship of Aetius was in 446, and that this appeal was thus some thirty years o r more later than Gildas had sup posed. But he also knew, probably from his ecclesiastical friends at Canterbury who supplied him with useful infor m ation about the early doings o f the Kentish royal family, that there had been a settlem ent o f federate barbarians dated to the jo in t reigns o f the emperors Marcian and Valentinian, that is, between 450 and 455. There was obviously no room for Gildas’ ‘prosperity period', during which in his view kings rose and fell, sporadic raiding continued, and there was pro gressive corruption o f Church and people, between 446 and 450-55. Although Bede conscientiously attem pts to allow for a condensed version o f it, he had to be content either to treat the A dventus Saxonum as an almost immediate consequence of the failure o f the appeal to Aetius, that is, ‘around 4 4 6 -7 ’, or to place it in the joint reign o f Marcian and Valentinian, between 450 and 455, as his Kentish sources indicated. It is interesting, in connection with the earlier date, that the Anglo-Saxon Chronicle (MS E) records under 443 an appeal to Rome, immediately followed on its rejection by a similar appeal to ‘the nobles o f the Angle race’. This entry, though here dated three years earlier than the letter to Aetius, may refer to the same sequence o f events and so preserve a tradition exactly in parallel with that which led Bede to accept the letter as one alternative clue to the date o f the A dventus Saxonum. 1 M. M iller, ‘B ed e’s use o f G ild a s’, E H R 9 0 (1 9 7 5 ), 2 4 1 -6 1 .
10
I. T H F . L I T E R A R Y S O U R C E S
It is im portant to understand the origin of these dates be cause they show how the A dventus came to be used by subsequent writers as a peg on which to hang the occurrence of later events. Their use in this way meant that the Adventus Saxonum became almost an artificial concept devised for historical purposes and chronological com putation largely independent o f any intrinsic significance that may have originally attached to the events so dated. This is not to say that they can be disregarded as unim portant. The ‘groans of the Britons’ to Aetius are o f the highest consequence as one of the very few occurrences in British history o f this time that can be closely dated, whatever its immediate cause and its effects m ay have been. The Marcian-Valentinian date for federate settlem ent in Kent is also certainly o f high antiquity, for this type of reckoning by imperial reigns goes back to an age when the known succession o f Roman emperors provided a sure guide to the passage o f time. It was familiar, for example, to Gregory o f Tours w riting his History o f the Franks at the end of the sixth century. It is w orth recalling that not long before his time a literate Frankish princess, Bertha, had come to Kent with a retinue, including a bishop as her chaplain, to be married to ÆthcTbert before his accession to the Kentish throne in 560. W ritten records o f early Kentish history may well have been compiled in her circle at C anterbury, attached to dates o f this type, very much earlier than anywhere else in Anglo-Saxon England, indeed within a generation o f the date when Gildas was writing his De Excidio in western Britain. By the time the Anglo-Saxon Chronicle was put together in its present form about three centuries later, this type of dating by reference to the regnal years o f Roman emperors was long obsolete. This no doubt explains why Marcian and Valentinian appear under 449 at the beginning o f the Kentish Annals in the Chronicle in the bungled disguise o f ‘Maurice and V alentines’. This tell-tale corruption o f a time-signal whose meaning had long been forgotten is o f course not only a proof o f the high antiquity o f these Kentish annals them selves, b u t a clear indication that they had followed a line of descent into the Chronicle quite independent o f Bede. No one basing chronicle entries on Bede could possibly have made such a nonsensical mistake.
DATE OF MONS BADONICUS
11
It would thus seem clear th a t the traditional picture of the A dventus around the central years o f the fifth century has come down to us as the beginning o f Anglo-Saxon history, not because these events were necessarily a record o f the earliest English settlem ents that took place, but because they w'ere the earliest such events whose dates were more o r less certainly known. It follows that one should be prepared for evidence, o f whatever kind, th a t may suggest different dates or different parts o f the country as the times and places for earlier Anglo-Saxon settlem ent. It may even turn out that, so far from recording the beginning o f a process, the traditional A dventus stories may be found to be describing the end of it, remembered precisely because these tales were the nearest in time to the days when oral memory could at last be rein forced by written record. There is one other occurrence in Gildas’ generalized account of the Saxon invasions in the fifth century to which he was apparently trying to attach an exact date. This is the obsessio M ontis Badonici, an event which he emphasizes as the turning-point after which there had been no further Anglo-Saxon advance at least until the time when he was writing. He seems to be trying to say that he knows th a t this took place nearly forty-four years ago, because it was in the year of his birth and he is now nearly forty-four.1 U nfortu nately a com bination of Gildas’ tortuous Latinity and a possibly corrupt text makes it not quite certain that this was his intended meaning. Moreover the m atter is made more complex by the fact that the manuscript o f Gildas used by Bede, which must have been w ritten within tw o hundred years of his autograph and was at least three hundred years older than any we now possess, apparently dated Mons Badonicus nearly forty-four years after the A dventus Saxonum , w ithout any indication that this figure was intended by Gildas as his age at the time o f writing. This is, incidentally, a good example o f the tendency already noted among Dark-Age historians to use the A dven tu s, however dated, as a peg to which other events can be attached. If Gildas was already 1 D e E x c id io 2 6 , q u iq u e q u a d ra g esim u s q u a rtu s, u t n o v i, o r d itu r annus, m e n s e iam u n o e m e n s o q u i e t m e a e n a tiv ita tis e s t. See A p p e n d ix III f o r a d is c u ssio n o f th e d a te o f M o n s B a d o n ic u s. I h a v e su g g e ste d a n h is to ric a l c o n te x t fo r it in d isc u ssin g th e fo rm a tio n o f W essex o n p p . 1 5 2 -6 2 .
12
I. T H E L I T E R A R Y S O U R C E S
doing this quite early in the sixth century, as Bede’s text implies, it cannot even be certain that his date for the A d ventus m atched either o f those employed much later by Bede. In spite o f these frustrating uncertainties it remains true that Gildas is the only contem porary w riter who attem pted, however inadequately, a general survey o f British history from the latter days o f Roman rule until the middle of the sixth century.1 When he was writing, Maglocunus (Maelgwn), the dom inant ruler in N orth Wales who died in 547, was still alive, and, if Gildas was right in his apparent belief that he had himself been born in the year of Mons Badonicus some forty-four years earlier, that makes him on any showing a contem porary witness to the situation somewhere in Britain in the first half of the sixth century’. But it is im portant to appreciate the tem poral and spatial lim itations o f his vision. What he says o r implies about the condition o f affairs in that half-century is first-hand contem porary evidence, but only for the parts o f Britain with which he was personally familiar. What he says about the second half o f the fifth century can be based only on what his parents and others o f the previous generation m ay have told him: it may thus be broadly true. But what he says about the century before th a t, say 350-450, is nothing but hearsay, and o f interest mainly as illustrating how little accurate inform ation had come through to educated circles in his generation from even the last years o f the Roman past.2 For the present purpose it is also im portant to appreciate the geographical limits o f his vision. He displays throughout what may be termed a Highland Zone m entality, which viewed the breakdown o f Roman authority from somewhere 1 T h e re is n o n e e d t o ta k e serio u sly th e a tte m p ts t h a t have b e e n m ad e to d e ta c h th e h is to ric a l fro m th e h o r ta to r y se c tio n s o f t h e D e E x c id io a n d t o d ate th e m d iffe re n tly . It has b e e n s h o w n c o n v in c in g ly b y F . K e rlo u é g a n in M . W. B arley a n d R . P. C . H a n so n (e d s .), C h ristia n ity in B rita in 3 0 0 - 7 0 0 (1 9 6 8 ) , 1 5 1 7 6 , th a t th e re is n o lin g u istic d is tin c tio n b e tw e e n th e d iff e r e n t se c tio n s: sy n ta x , p h ra se o lo g y , v o c a b u la ry , figures o f s p e e c h , e tc . a rc id e n tic a l th r o u g h o u t, an d all fit re a d ily in to a n e a rly s ix th -c e n tu ry stage in th e d e v e lo p m e n t o f lite ra ry L a tin , a p p ro p ria te to a reg io n in w h ic h it w as b e c o m in g a le a rn e d to n g u e w hich h ad larg ely c e a se d t o b e th e c o m m o n sp e e c h . ' T h is a p p ro a c h to th e c ritic ism o f G ild as as a n h is to ric a l s o u rc e is v e ry well d e v e lo p e d b y J . M orris in B arley a n d H a n so n , o p . c it. 6 2 -3 .
TH E VALUE O F GILDAS
13
in Western Britain1 in terms o f unsuccessful defence against northern barbarians, the building and loss o f frontier walls. He shows little direct interest in eastern Britain, and the part played by Saxons in the earlier stages o f the collapse of Rom an or sub-Roman authority is virtually ignored.12 When they eventually appear on Gildas’ scene it is not as devastating raiders and invaders o f Lowland Britain, but as federate allies settled there by the ‘proud ty ran t’ to defend his lands against Piets and Scots. Gildas thus placed their coming long after the failure o f the last appeal to Rome and the ensuing ‘prosperity period’. This was the main reason why later writers ignored the possibility o f Saxon settlem ent in the fourth century long before the formalized date attributed to the A dventus Saxonum in the central years of the fifth.3 In summary, then, Gildas can be treated as a credible and most valuable witness to a broad sequence o f events in Britain in the century following the third consulship of Aetius (446-53). His knowledge o f a British appeal at that time rests on a dated docum ent and implies a serious barbarian invasion in the immediately preceding years. This disturbance can naturally be taken as reflected also in the news o f Saxon penetration which prom pted the exaggerated reports of total conquest th a t found their way into the South Gaulish chronicles. Gildas knew also o f a settlem ent o f federate Saxons in eastern Britain about that time, which gave rise to the use by later writers o f 446-7 as a date for the A dventus Saxonum . He attributed this settlem ent to the policy of a superbus tyrannus o r ‘proud ty ran t’, a phrase which seems to be simply a Latin translation o f the word ‘V ortigern’ which 1 T h o u g h h e w ro te w ith d e ta ile d k n o w le d g e o f th e p o litic s o f W ales an d D u m n o n ia an d w as la te r a sso c ia te d w ith B ritta n y . G ildas p ro b a b ly cam e fro m th e n o rth -w e s t n e a r th e C ly d e . T h e e a rlie st L ife s ta te s th a t h e w as A r e c lu ta reg io n e o riu n d u s. A c co rd in g t o J a c k s o n , Language a n d H is to r y , 4 2 , th e sp ellin g o f th is n a m e “c le a rly b e lo n g s to th e six th c e n t u r y 'a n d ‘c a n o n ly c o m e fro m c o n te m p o ra ry m a n u s c rip ts ’. 2 T h e o n ly in d ic a tio n th a t G ild a s w as c o n sc io u s o f th is e a rlie r S a x o n m en a ce is h is m e n tio n o f th e R o m a n s p ro v id in g tu rre s p e r in tervalla a d p r o s p e c tu m m aris (D e E x c id io 1 8 ). W h e th e r th is w as in te n d e d as a re fe re n c e t o S a x o n S h o re fo rts o r to c o a sta l signal s ta tio n s is n o t c le a r, b u t in e ith e r case h e m u st have been th in k in g o f d a n g e r fro m b e y o n d th e N o rth S ea o r t h e C h an n el. 3 I h a v e d e v e lo p e d th is a rg u m e n t f u r th e r in 'T h e A d v e n tu s S a x o n u m ’ in G rim e s , o p . c it. 2 2 2 -4 .
14
I. T H E L I T E R A R Y S O U R C E S
appeared in his Celtic sources and must have been originally rather a title than a personal name. There are hints in his language th a t here to o he tnay have had access to contem porary docum ents. The m ost natural explanation o f his reference to the Saxons coming in ‘three keels’ is that this was the num ber o f ship loads specified in the first formal invitation which led to the foedus o r treaty settlem ent. Gildas also uses in their correct sense technical terms, annona, epimenia, hospites,l which most likely derive from official docum ents relating to the billeting and supply of barbarian foederati. In his view this settlem ent, which he does not precisely locate and was not necessarily in Kent, was followed after a while by further less regulated and less controllable arrivals, and eventually by a general outburst marked by widespread slaughter and destruction. In this episode, whose duration is not specified, Gildas deplores in particular the physical overthrow and ruin o f towns, and the massacre o f Christian clergy, both decisive in bringing about the liquidation o f the most characteristic features of late Roman urban civilization. Although he writes o f all this in melodramatic phraseology employed deliberately to create the maximum impression, Gildas m ust here be treated as a witness to tru th . It would have been fatal to his argument to describe in such terms a situation which his readers knew to have been quite different. Eventually, however, he states that resistance was organized under a leader called Ambrosius Aurelianus, sprung from a family which must have produced one or more o f the tyrants, or local em perors, o f the first half o f the fifth century. After a ding-dong struggle o f unspecified duration he reports th at a decisive British victory was won at a place called Mons Badonicus, and that this had been followed by something over a generation o f comparative peace which was still u n broken, though in Gildas' view seriously threatened, at the date of his writing around 540. In spite o f this precarious peace, however, Gildas is careful to point out that the towns were still deserted, and here too his evidence m ust be accepted as a true description o f the situation as both he and his readers knew it. •
1
D e E x c id io 23.
AMBROSIUS AND ARTHUR
15
Gildas’ sketch o f British history in the century before his own m aturity is as im portant for what it does not portray as for w hat it does. The fate o f the ‘proud ty ra n t’ is not speci fied, b u t he was evidently n o t thought o f as playing any part in the British revival. The whole credit for this is attributed to Ambrosius Aurelianus, whose name and background not only imply an origin in the highest levels o f the RomanoBritish nobility, but suggest a close link o f some kind with the Catholic doctrinal circles centred upon St. Ambrose, the very influential Bishop o f Milan (373-97), whose father was named Aurelius Ambrosius. This parallel in nomenclature must surely be significant. That Vortigern was afraid of Ambrosius as well as o f the Piets and Scots, and o f the threat from Roman Gaul is specifically stated in a passage of Nennius’ Historia B rittonum 1 which m ust come from a near contem porary source. I have suggested elsewhere12 that Vortigern is likely to have been the leader o f the Pelagian party among the British notables whose influence the visits of Germanus were designed to counteract, and th at, in attributing to Ambrosius the successful resistance to the revolted Saxon foederati brought in by Vortigern, Gildas was recording, perhaps half consciously, n o t only a remarkable military revival on the part o f the Britons but the trium ph of Catholic over Pelagian ideology. This aspect o f the m atter is however only indirectly relevant to the story o f the English settlements. What is certainly more significant is that Gildas, in writing the only contem porary narrative o f these m om entous events, has no m ention whatever o f A rthur. His silence is decisive in determining the historical insignificance o f this enigmatic figure. It is inconceivable that Gildas, with his intense interest in the outcom e o f a struggle th a t he believed had been decis ively settled in the year o f his own birth, should n o t have mentioned A rth u r’s part in it had that part been o f any pol itical consequence. The fact is th a t there is no contem porary or near-contem porary evidence for A rthur playing any 1 H ist. B r itt. 3 1 . G u e r th ig im u s re g n a vit in B ritta n ia e t d u m ipse r e g n a b a t. . . u rg e b a tu r a m e tu P ic to r u m S c o tto r u m q u e e t a R o m a n ic o im p e tu n e c n o n a tim o r e A m b r o s ii 2 ‘Pelagius a n d th e e n d o f R o m a n ru le in B rita in ’, J R S 50 (1 9 6 0 ), 2 1 -3 6 , e sp ec ially 3 5 - 6 .
16
I. T H E L I T E R A R Y S O U R C E S
decisive part in these events at all. No figure on the border line of history and m ythology has wasted more of the historian’s time. There arc just enough casual references in later Welsh legend, one o r two o f which may go hack to the seventh century, to suggest that a m an with this late Roman name—Artorius—may have won repute at some ill-defined point o f time and place during the struggle. Hut if we add anything to the bare statem ent that A rthur may have lived and fought the Saxons, we pass at once from history to rom ance.1 This passage was in fact already being made before British scholars collected the materials for the Historia B rittonum early in the ninth century, and the later popularity of this curious com pilation greatly accelerated it. Nennius’ historical m ethod as compiler is best described in his own words. Coacervavi, he writes, om ne quod inveni—T have made a heap o f all that I have found.’ Among the scraps in his heap were some A rthuriana, notably a list o f twelve battles culminating w ith Mons Badonicus in which the defeat o f the enemy is attributed to A rthur’s personal prowess. This list was evidentlyextracted and condensed from epic m atter compiled in A rthur’s honour at some unknown date not necessarily long before the ninth century. Much ingenuity has been devoted to attem pts at localizing the places m entioned and even to devising intelligible campaigns to account for them . But the conclusions offered by various students cover the whole country from Scotland via Lincolnshire and Sussex to the far west, w ithout establishing any convincing historical basis. It is not even certain that all these battles involved Saxons. That the list culminates with Mons Badonicus at once places it under grave suspicion, for the contem porary evidence of Gildas clearly associates this decisive event with Ambrosius, not A rthur. It suggests in fact that the association o f A rthur with all or any o f these scenes o f real o r fictitious conflict may be due rather to wishful thinking by an imaginative panegyrist in possession o f a traditional list o f battle sites than to historical reality. 1 T o d e sc rib e th e w h o le p e rio d o f th re e h u n d re d y e ars fro m 3 5 0 to 6 5 0 as ‘T h e Age o f A r th u r ’, as is d o n e fo r e x a m p le b y J . M orris in h is b o o k w ith th a t title ( 1 9 7 3 ) , show s a t o ta l d isre g a rd o f th e valid h isto ric a l e v id e n ce . S ee m y review in E H R 9 0 (1 9 7 5 ) , 1 1 3 -1 6 .
TH E H ISTO RIA
RR I T T O N UM
17
The Historia B rittohum in its present form dates from an age three hundred years after the time o f Gildas. During these centuries there had been much contact between the British and Saxon peoples. Its compiler was thus able to incorporate in it some materials originally o f Anglo-Saxon origin and dating to the early stages o f the settlem ent. He had, for example, genealogies o f several o f the Anglo-Saxon royal families, and some semi-historical memories o f the conflict between Hengist and Vortigern that led to the perm anent establishm ent of the Anglo-Saxons in the south-east of Britain. He also had notes o f a few otherwise unrecorded incidents, mostly of internal conflict on the British side, which accompanied or followed this settlem ent. For the final collapse of V ortigern’s authority he relied on extracts from a legendary Liber Reati Germani, now lost. This seems to have contained a sort o f picture-book version o f the doctrinal conflict between the Pelagian and Catholic parties among the British notables, personalized as a political struggle betw een Germanus and Vortigern in which the form er finally eliminated the latter, after a series o f melodramatic en counters in the m ountains o f Wales.1 Nennius o r his sources also had some interesting scraps of inform ation in his heap concerning the early stages of Anglian settlem ent in the north o f England in the sixth century. Some o f these relate to tales preserved independently in early Welsh heroic poetry associated with the names of bards such as Aneirin and Taliessin. He was also interested, though in a much less systematic way than Bede, in chrono logical problems, and included a num ber o f calculations whose significance has given rise to much discussion.2 In these he appears to be relating various later occurrences to such fixed points in time as he thought he had for V ortigern’s reign, the A dventus S a xonum , and so on. Very little o f value Sec M y res, ‘Pelagius’, 3 5 . I see n o re a so n t o fo llo w t h e o p in io n o f som e r e c e n t sch o lars th a t th e H istoria B r itto n u m h a s c o n fu s e d t h e h isto ric a l G e rm a n u s o f A u x e rre w ith a n o b sc u re S t. G a rm o n k n o w n o n ly fro m a s c a tte r o f c h u rc h d e d ic a tio n s in v ario u s p a rts o f W ales. C o n s ta n tiu s ’ a c c o u n t c le a rly sh o w s th a t G e rm a n u s m u st have h a d p e rs o n a l e n c o u n te rs in d e b a te w ith P elagian n o ta b le s in B rita in . T h ese d isc u ssio n s c o u ld w ell have b e e n d ra m a tiz e d by e x u b e ra n t Welsh im a g in a tio n in to th e ro m a n tic a d v e n tu re s w hich N e n n iu s p a ra p h ra s e d fro m his re a d in g o f th e L ib e r B e a ti G e r m a n i 2 H ist. B r itt. 66.
18
I. T H E L I T E R A R Y S O U R C E S
can he extracted with certainty from this part o f Nennius’ heap, although some o f the dates are expressed in terms which suggest that they may derive from sources as early as the fifth o r sixth centuries. Such then in bare outline arc the main w ritten sources from which the story o f the English settlem ents has to be deduced. Apart from a few chance references in continental writers primarily interested in other m atters, and that section of Gildas’ narrative that rests on his own experience or his parents’ m em ory, almost the whole o f it belongs in its present form to much later times. It is thus heavily contam i nated with accretions and irrelevancies picked up in the slow progress from these original sources. But here and there some significant piece o f inform ation has come through almost un scathed from the fifth century into these later compilations. Such is the note in the Historia B rittonum o f V ortigern’s policy being based not only on the menace from Piets and Scots, b u t on his dread both o f renewed Roman invasion and o f Ambrosius. Even the m ost unreliable o f the later lives of Celtic saints may contain shafts of light suddenly illuminating a dark corner o f the sub-Roman scene. There is the tale, for example, o f how St. Tatheus was entertained in a dilapidated Roman villa near Chepstow whose owner was still trying to maintain the heating o f his bath-house, if only a t week-ends. Then again, St. Paul Aurelian, visiting in Brittany a Roman town still surrounded by earth ram parts and a lofty stone wall, is said to have found the only signs o f life in it to comprise a sow, a bear, a wild bull, and a hive full of bees and honey housed in a hollow tree.1 Such scenes, like that in which the anonym ous biographer of St. C uthbert portrays the saint sight-seeing in 685 in the ruins o f Roman Carlisle,12 provide dram atic confirm ation for Gildas* picture o f the last days o f urban and rural civilization in Britain as still remembered only half a century before his own birth. But they do nothing to reconstruct even in broad outline a political history o f the centuries covered by this book, or to explain the social and econom ic changes that converted Roman Britain into Christian England. So far as the political 1 T h ese in c id e n ts a re b o th q u o te d b y J . M o rris, A g e o f A r t h u r (1 9 7 3 ) , 253 an d 4 5 8 . 2 A n o n ., V ita S . C u th b e r ti, 37.
SUMMARY OF THE EVIDENCE
19
story is concerned it is n o t too much to say that for the sequence and significance o f the main events o f this time the elements that contribute to historical proof are for the most part lacking. Even when it is certain that an event o f im port ance did occur—say the siege o f Mons Badonicus which seems to have marked a tem porarily decisive check to the progress of Saxon settlem ent—it is often impossible, as in that in stance, to say exactly when o r where it happened, or, for that m atter, exactly what took place. Moreover, such events as seem to be reasonably well authenticated frequently display a reluctance to cohere into a single story. There is, for example, no point o f contact between the early South Saxon and West Saxon annals in the Anglo-Saxon Chronicle and w hat Gildas has to say about Ambrosius Aurelianus and Mons Badonicus, although the tw o scries o f events must have been roughly contem porary and apparently belong to the same quarter of England.1 In the social and economic sphere, Gildas certainly provides good evidence for the destruction in the second half o f the fifth century o f the urban framework which had supported not only the civil adm inistration o f the Romano-British diocese b u t also the Christian organization based until that time essentially on the bishops and clergy of the towns. Gildas may also hint, though less specifically, at the wide spread dislocation o f rural life brought about both by the physical destruction o f villa estates and farmsteads in the countryside, and by the collapse o f the m oney economy on which their viability, as something more than subsistence units, had rested. At any rate this m ust be the meaning behind his casual m ention, almost in an aside, that a time came when the province was deprived o f its regular food supply save only for the resources o f hunting.2 But he gives no details under cither head and indeed m ay not have known any. It is of course especially at the beginning and the end of these tw o dark centuries that the literary evidence has least of certainty to tell. Thus the first half o f the fifth century, after m ost official contacts between Britain and the world of 1 S ee in fra p p . 1 5 8 -6 0 f o r a n a tte m p t t o Find a n h is to ric a l c o n te x t f o r M ons B a d o n ic u s in th e e v o lu tio n o f th e W est S a x o n k in g d o m . 3 D e E x c id io , 19.
20
I. T H E L I T E R A R Y S O U R C E S
Rome had broken dow n, was already beyond the reach of living m em ory in Gildas’ day, and he can tell us nothing coherent or credible about it. What contem porary information has survived from this sub-Roman phase was generated mainly by the ideological conflict between the Pelagian and Augustinian interpretations o f basic Christian doctrine. This conflict seems to have bedevilled civilized thought among the Britons just at the time when unity was above all things essential for success in the struggle with Celtic barbarians and Germanic intruders alike. So it is also at the other end o f the period, after the middle of the sixth century, when Gildas’ narrative ends with dire and well-justified forebodings o f disaster from renewed barbarian onslaughts. Here too contem porary literary evi dence largely fails us for the years o f decisive Saxon conquest when his prophecies were to be so bitterly fulfilled. For this critical tim e there is virtually nothing from w ritten sources for southern Britain except a few brief and tantalizing entries in the Anglo-Saxon Chronicle m ostly culled from a lost saga centred on the exploits o f Ceawlin and his contemporaries in Wessex.1 For the north little literary evidence remains b u t a few even less coherent incidents derived from similar memories o f the wars which established the Anglian dynasty of-Id a in the further parts o f N orthum bria.2 For the great mass o f eastern and central Britain which bore the b ru n t of Anglo-Saxon conquest and settlem ent at this time there is no inform ation from literary sources at all, and we arc dependent wholly on w hat archaeology and place-names have to tell. 1 in fra p p . 1 6 2 -7 1 .
3 in fra p p . 1 9 9 -2 0 1 .
2
T H E N A T U R E O F T H E E V ID EN C E: IT ARCHAEOLOGY AND PLACE-NAMES the archaeological evidence we are at least handling tangible material that comes direct from the years under review. While its quantity in no way compares with that from the Rom an period it is, at any rate on the Anglo-Saxon side, impressive both in its range and its quality. But here to o , as w ith the literary evidence, there is a marked contrast between the material remains left behind by the invaders and the invaded. At first sight it might be supposed that the Britons, heirs to the commercialized culture of Rome with its money economy and wide range of mass-produced goods, would be represented archaeologically, even in decline and disaster, by a mass o f easily recognizable and informative dom estic rubbish. From this one might hope to learn much about the nature and distribution of their surviving centres o f power, and about their success or failure in adapting to the simpler economic conditions and declining standards of living that were now their lot. But this is not so. Apart from a meagre supply o f the simplest types o f brooches, buckles, belt-fittings, and knives, the sub-Roman Britons o f the fifth and sixth centuries appear to have enjoyed—if that is the right w ord—a culture almost as completely devoid o f durable possessions as any culture can be. In Wales and the far west, more o r less beyond the reach of Anglo-Saxon penetration in these centuries, their presence can, it is true, be recognized from a considerable num ber of Christian tom bstones and some fortified or refortified hill top strongholds containing occasionally a scatter o f p o t sherds. These m ay include some from vessels imported at this time by the western sea-ways from the Mediterranean lands. But in the east and centre o f lowland Britain even such tell-tale material indications o f their presence are almost wholly missing. It has required the development o f a highly
W
ith
22
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
specialized technique o f excavation that can only be employed in the most favourable conditions, as in the fields that now cover the site o f Roman Viro conium (W roxeter), to suggest the presence o f a num ber o f tim ber structures th at once housed the last remaining citizens o f the Civitas Cornoviorum almost totally bereft o f durable household goods.1 It would be natural to draw from this surprising lack of recognizable archaeological traces the conclusion that the sub-Roman population o f lowland Britain had mostly dis appeared. It is difficult in any case to resist the impression that their numbers m ust have been very greatly reduced by the casualties o f war, destitution, emigration, and disease, especially among the urban classes least capable of w ith standing hardship. Gildas clearly implies th a t this was so. But before using this negative archaeological testim ony as proof of the once fashionable view that the native population of lowland Britain wras virtually eliminated in these dark centuries, it is well to consider whether this negative record may be interpreted differently. It is probable, indeed certain, that the industries m anu facturing mass-produced goods in durable materials such as pottery, glass, and metal, would be among the first to be hard hit by the collapse o f the money econom y, and by the growing insecurity o f transport whether by road or by water. Later Roman society moreover had become almost wholly dependent on a copious and cheap supply o f consumer durables. The traditional skills required to replace such supplies by the local and domestic enterprise universal in most primitive societies must have been very largely for gotten. As such goods became scarcer through wear, break ages, and loss, it would be natural for many o f them to be replaced in less durable materials, wood, leather, and textiles, which leave little or no trace for the archaeologist. Thus the place o f bronze, iron, glass, and the harder wheel-made products o f the commercial potteries woidd be taken by more perishable home-made products requiring less technical ' P. B arker in P . J . C asey ( c d .), T h e E n d o f R o m a n B rita in , B A R 71 ( 1 9 7 9 ), 1 7 5 - 8 1 . T h e re is a r a th e r im aginative p o r tr a y a l o f th e c e n tre o f p o st-R o m a n V iro co n iu m in P. B arker ( e d .), W ro x e te r R o m a n C ity E x c a v a tio n s 1 9 6 6 - 1 9 8 0 ( n .d .) , fig. 5 . I t m a y be sa lu ta ry to c o m p a re th is re c o n s tru c tio n w ith th e a c tu a l re m a in s as fo u n d . See, e .g ., B r it. 6 (1 9 7 5 ) , 1 0 6 -1 7 a n d figs. 3 - 6 .
SUB-ROMAN A R C H A EO LO G Y
2 3
skill in their m anufacture. There might a t the same time have been a widespread tendency for ruined buildings that had been constructed in brick and stone, more R om ano, to be generally replaced in flimsier materials better suited to the simpler needs o f a less civilized way o f life.1 It must also be rem em bered th a t the latest levels o f any archaeological site are necessarily the least well preserved because they are the most vulnerable to interference and destruction in later times. But w'hen all allowance is made for these factors, it remains difficult, if not impossible, to account for the almost total absence o f sub-Roman archaeological remains in lowland Britain on any other hypothesis than a drastic reduc tion in numbers o f the native population and an equally drastic reduction in the standard o f living of those who remained. Most impressive in this context is the contrast between east and west. One has only to compare the negative evidence from central and south-eastern Britain w ith what is found in Wales and the south-west, scanty though this too may be, to realize that the latter was a society still com paratively vigorous and vital, and, however barbarized, still capable o f maintaining fortified strongholds, centres of Christian culture, and some contacts with the Mediterranean world. But the natives o f lowland Britain have left little or no surviving trace o f any o f these things, nor were they even capable of retaining their own language o r their own placenames in com petition with those o f their conquerors. The archaeological remains o f the Anglo-Saxons during their pagan phase arc o f an altogether different order. It is true that most o f them come from cemeteries rather than from habitation sites, and that in the absence o f contem porary coins and inscriptions nothing is so securely or so precisely datable as one could wish. There is indeed a remarkable contrast in this and other ways with the archaeology of Roman Britain which can perhaps best be appreciated by a 1 T h is c e rta in ly h a p p e n e d a t W ro x e te r. V e ru la m iu m is o n e o f t h e few R o m an c ities th a t h a s p ro d u c e d e v id en ce o f su b s ta n tia l n e w b u ild in g s, involving even so m e d e lib e ra te re a rra n g e m e n t o f th e s tr e e t p la n , as la te as th e m id -fifth c e n tu ry . B u t even th e r e , if C o n s ta n tiu s ’ a c c o u n t o f th e se c o n d v isit o f G e rm a n u s in th e 4 4 0 s can b e tru s te d , th e visiting s a in t seem s to have b e e n lo d g ed in a th a tc h e d h u t , h a rd ly th e a c c o m m o d a tio n fo r a d istin g u is h e d b ish o p fro m G a u l h ad a n y th in g b e tte r b e e n av ailab le.
24
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
simple comparison. I f litilc b u t the cemeteries o f Roman Britain were available for study and if there were no coins and inscriptions to supply dating evidence, the contribution of archaeology to our knowledge o f that phase o f British history would be slight indeed. That however is the position o f the Anglo-Saxon archae ologist. Much can certainly be learnt o f the course and character o f the invasions from the distribution pattern of cemeteries and settlem ents in relation to the geography and geology o f those parts o f the country where they are found. Even more significant docs that distribution pattern appear when superimposed on that o f Roman tow ns, forts, villas, and villages and o f the roads that connected them . In the past Roman Britain and the English settlem ents were too often treated as m utually exclusive periods o f history, each buttended against the other with a minimum o f linkage or overlap between them . This tendency was reinforced by the contrast between the plentiful range o f mass-produced goods evidenced in the excavated ruins o f Roman towns, forts, and villas, and the simpler types o f hand-made tools, weapons, and pottery that come from the Anglo-Saxon cemeteries. The contrast is certainly real and im portant, yet the distribution maps indicate that in m any areas the Anglo-Saxon shows a marked tendency to follow the Romano-British p attern, in a fashion which suggests a considerable degree of temporal as well as of spatial overlap. It is at this point that it becomes necessary’ to consider what means are available for dating the earliest archaeological traces o f an Anglo-Saxon presence in Britain. In the absence of associated coins and inscriptions such dating must rest largely on typological considerations, the points in developing sequences o f the com m oner forms o f artefacts at which English examples occur in quantities sufficient to be sig nificant. For this m ethod to be reliable it is essential that the continental sequences on which it rests should themselves be well-established, and tied with reasonable security to the passage o f time. In the last resort this means a sufficient num ber of links, direct o r indirect, with deposits dated either by associated Roman coins, o f which large num bers were in circulation at this tim e beyond the Imperial frontiers in Europe, or by historical events whose dates are known. In
ANGLO-SAXON ARCHAEOLOGY
2 5
this way a vast interlocking chronological network has been built up by continental scholars for dating the typological sequences o f many sorts o f products. While differences of opinion may persist on the position o f any particular object in its appropriate sequence, and while the grounds forgiving it a rough date m ay often appear individually flimsy, yet the main chronological framework is now so extensive and so firmly established th a t it is m ost unlikely to be overturned, or even to be shifted as a whole either forward or backward by an appreciable period o f time. It must however be recognized that archaeological dating of this kind can never be precise. It may be fairly safe to date the m anufacture o f an object o r a group o f associated objects within a bracket o f seventyfive o r even fifty years. But it is very rarely, if ever, possible to claim th a t any such objects must have been made either before or after a particular year. For this reason archaeological evidence is very seldom of direct assistance to a historian seeking precision in the dating of events to a year o r tw o, o r even a decade. Further uncertainties may arise in dealing with material mainly derived from cemeteries rather than from the debris of occupation. In the stratified levels o f a house or a ditch system it is generally safe to treat m ost o f the material from a sealed stratum as roughly contem porary in use, especially when it consists largely o f fragile objects such as pottery or glass. These are likely to have had a short working life and once broken serve no useful purpose which would justify their removal to a different context. But grave-goods are a different m atter. A much higher proportion o f them consist of tools, weapons, or ornam ents such as brooches, buckles, strap-ends, and so on, which are not only durable and longlasting b u t intrinsically valuable, and thus liable to be reused outside the context for which they were first made. A man may have buried with him objects o f value made at every stage of his working life, n o t to m ention the possibility of inherited heirlooms or loot plundered from his enemies. It is well-known, for example, th a t certain types o f weapons, notably swords, were treasured in barbarian society for sentimental or mystical reasons. It might be several gener ations before a valued object o f this sort found a final restingplace in the grave o f its last owner, perhaps unrelated by
26
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
blood or cultural association with its first. An unusual degree of wear and tear or evidence o f damage o r repair, not to m ention the occasional conversion o f an object to a purpose other than that which it first served, arc all indications of a probable heirloom, or o f a more recent acquisition possibly much older than the date o f the grave in which it is found. This, to take one example, m ust be the explanation for the presence o f the battered remains o f a fourth-century south German brooch in an Angle grave at Londesborough, East Yorks., that is securely dated no earlier than the mid-sixth century by the other objects accompanying the skeleton. Had these other objects not been present it might have been reasonable to date the grave at least a century earlier than its actual period.1 It is now normal for archaeologists to pay far m ore atten tion than used to be given to the extent to which objects, especially those o f high intrinsic value likely to be treasured as heirlooms, exhibit signs o f a long or short period of use before burial. The whole chronology o f the gold bracteates, which form a crucial element in the dating o f the Jutish phase of Kentish archaeology in this period,2 rests very largely on estimates of the extent to which individual pieces exhibit degrees o f wear and tear caused by varying periods of use between the time o f their m anufacture and th a t o f their deposition in the graves from which they were excavated.3 Notw ithstanding these complexities, it is largely from the intensive study o f metal objects from graves that the chrono logical fram ework o f Anglo-Saxon archaeology has been built up. The typological sequences o f all the main varieties of brooches, buckles, strap-ends, and other belt-fittings, as well as weapons such as spear-heads and shield-bosses, are now reasonably well-established. But for the reasons given they may n o t be as trustw orthy evidence for an absolute timescale expressed in the passage o f the years as could be desired by writers o f history. The very durability and intrinsic value o f such goods serve to limit their reliability as pointers to an
' A n t . ] . 4 7 (1 9 6 7 ), 4 3 - 5 0 . 3 S ee d isc u ssio n o n p . 115. 3 S ee S. C. H a w k e s a n d M . P o lla rd , F r ü h m itte la lte r lic h e S t u d ie n , 15 ( 1 9 8 1 ), 3 1 6 -7 0 .
METAL WORK AND POTTERY
27
absolute chronology, and so reduce their value as raw material to the historian. It should be possible to learn more from the study of successive styles and fashions prevailing in cheaper and more fragile products such as pottery. Here the quantities available for exam ination are much greater, their turnover in use must have been much more rapid, the individual pieces have little attraction either as loot or as heirlooms, and the likelihood of their transfer, once broken, to misleading contexts is minimal. For all these reasons pottery should provide a much more sensitive tool for the archaeologist, and thus one capable of making a picture much more sharply in focus for the historian. This in fact has been the function o f pottery in illuminating the study o f most early cultures, and there is no reason why, properly used, it should not provide a similar service in the early Anglo-Saxon age. As w ith the metal-work, most pottery o f this time comes from graves. Remains o f cook-pots and other domestic utensils do o f course occur in considerable quantity as house hold rubbish on occupation sites, b u t very little o f this is sufficiently distinctive to be o f much value for purposes of dating. There was very little commercial production o f house hold pottery at this time, so th a t it is hardly ever possible to study the distribution o f output from one workshop. Most families seem to have been content with home-made containers, generally undecorated and in the coarsest of fabrics. Moreover the normal forms which the pots take were designed to serve continuing dom estic needs, and thus show little tendency to follow any changing fashions that might have significance for dating. It is otherwise with the cemeteries. The rituals o f in hum ation gave scope for the inclusion o f food- and drinkofferings for the dead, contained in pottery vessels often of high quality and the latest fashionable style. Where cremation was practised the whole funerary deposit was normally placed in a pottery urn, whose form and decoration similarly reflected contem porary taste. These varied pottery styles arc invaluable in providing clues to the parts o f the continental homeland from which the different groups o f the invaders came. They can also be arranged, like other grave-goods, in typological series which, for the reasons already m entioned,
28
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
may be easily related to an absolute time-scale. Both the inhum ation and the crem ation p o ttery is thus ultim ately datable by the same criteria as apply to the metal objects with which they arc associated in the graves. It is however w orth noting that these associations are less subject to error or mistake in the case o f inhum ations than cremations. The presence o f a particular pot in a particular grave in direct association with the other grave-goods is norm ally a m atter of easily recorded observation. But the association o f metal objects, often difficult to identify precisely owing to distor tion by heat, with the ashes from a particular cremation urn may sometimes be less certain. Many urns are found collapsed in the ground owing to earth pressure, or broken by other disturbance, and may thus have lost part of their original contents or acquired extraneous objects from the surrounding soil. Moreover there can be no absolute certainty that the original contents gathered from the funeral pyre did not sometimes include objects from tw o o r m ore cremations o f different dates th a t had taken place on the same spot. This is a contingency m ost liable to arise in some o f the very large cemeteries, which seem to have been equipped with regular burning sites that may have been in use over a con siderable period o f time. In spite o f such occasional uncertainties, incidental in one form or another to the interpretation o f all archaeological remains, there can be no doubt that early Anglo-Saxon pottery, and especially th a t from the great cremation cemeteries o f eastern England, forms an invaluable and highly im portant source o f inform ation, whose potential for his torical purposes still remains largely unexplored.1 It will be drawn upon in detail whenever relevant to the m atters dis cussed in the later chapters o f this book. But it may be convenient at this point to draw attention to a few im portant aspects o f this period on which recent study o f the pottery has already throw n fresh and sometimes unexpected light. One such subject is that o f the continental origin of the invaders. Bede's celebrated statem ent that they came from three of the m ost pow erful peoples o f G erm any, the Angles, F u r th e r s tu d y o f th is m a te ria l s h o u ld b e fa c ilita te d b y th e p u b lic a tio n o f m y C o rp u s o f A n g lo -S a x o n P o tte r y in 1977.
V A LU E OF M ATER IA L REMAINS
29
the Saxons, and the Ju te s must be the starting point for this enquiry,1 and the relevant archaeological evidence will be examined in the next chapter. A nother m atter on which the evidence of the p o ttery should throw light is the distribution of the main elements o f the invading peoples in England. This too will receive more detailed consideration in later chapters. A third topic o f great im portance to the historian, and one on which recent study o f the pottery has led to some unexpected revision o f the traditional chronology, is the dating of the earliest phases o f the settlem ent. This will require further exam ination in discussing the relation o f the newcomers to the collapse o f Romano-British society in town and country. In assessing the contribution that can be made by archae ology to the solution o f these problems, the historian has to form his own judgem ent on these and many additional topics which will be brought up as this narrative proceeds. But how ever difficult and uncertain o f interpretation they m ay be, these material remains are o f great significance, if only because they do derive directly from the years under review, and come to us w ithout any interference o r distortion caused by secondary attem pts to explain their meaning. First-hand archaeological evidence, unlike m ost written sources of inform ation about the past, is wholly free from the kinds of error th at are so often introduced by subsequent copyists or the well-meaning but sometimes misguided editors o f literary texts. The same unfortunately cannot be said o f the evidence of place-names. Here we have to deal with m atter th a t has passed through many hands and been subjected to many possible distortions and transform ations on its long and often complex way from the fifth o r sixth century to the tw entieth. Nearly all the names used by the first Englishmen to identify their dwellings and the natural features among which they were set must have arisen more o r less unconsciously in the course of conversation between illiterate folk. In lowland Britain they would have needed great num bers o f such names to carry on their lives w ithout confusion, from the very 1 H E i. 1 5 . T h e relev an ce o f c e ra m ic e v id e n c e t o t h e in te r p r e ta tio n o f B ed e’s s ta te m e n t is m o re fu lly c o n sid e re d in m y R aleigh L e c tu re ‘T h e A n g les, th e S a x o n s, a n d t h e J u t e s ’, P B A 5 6 (1 9 7 0 ) , 1 4 5 -7 4 . S ec a lso m y C o rp u s, i. 1 1 4 -1 8 .
30
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
earliest years o f their settlem ent. More and m ore would be required as they became increasingly familiar w ith their own widening surroundings and ever m ore possessive in their attitudes towards them . But anything up to two hundred years or even more will have passed before, with the coming o f literacy in the seventh century, they can have begun to think of these names as w ritten words. In those tw o centuries sound-changes o f many sorts had been taking place in their speech, some old words had passed out o f com m on use, and others were changing their meaning in the face o f altered circumstances. It m ust often have been the case that the clerks who first tried to express in writing the sounds made by illiterate peasants to identify their own and their neigh bours’ villages will have produced verbal com binations of little obvious meaning to their inform ants o r themselves. They might often bear little relation to the descriptive terms first used to form these names by their ancestors several generations before. It is moreover salutary, if saddening, to remember how very few w ritten place-names have come down to us from the first century o f Anglo-Saxon literacy, and that forms first recorded even in eighth- or ninth-century original docum ents are still precious rarities. In fact vast num bers o f Anglo-Saxon place-names appear in their earliest known written shape in administrative records later than the Norman Conquest, such as Domesday Book, inevitably distorted by French-speaking clerks unfam iliar with the sounds and the sense o f w hat was to them a largely alien tongue. It is n o t surprising, in view o f the difficulties thus inherent in the correct transmission o f these names from early times, that there should be much doubt and disagreement surround ing the interpretation o f the first w ritten forms in which they appear. Some o f these uncertainties are directly relevant to their proper use in various historical contexts, and will there fore require discussion here as and when these contexts arise. But one general conclusion o f great significance emerges at once from this great body o f place-name m aterial. That is its overwhelmingly Anglo-Saxon character. Names o f British 1 A v a lu a b le a tt e m p t to a sse m b le th e p la c e -n a m e e le m e n ts f o r w h ic h e v id en ce e x is ts in th e o ld e s t w ritte n E nglish so u rc e s u p to 7 3 0 c a n b e fo u n d in B. C o x , J E P -N S 8 (1 9 7 6 ) , 1 2 -6 6 .
PLACE-NAMES AS E V ID E N C E
31
origin arc mainly confined in lowland Britain to natural features such as some rivers, prom inent hills, and areas of forest, waste or fen, which required immediate identification in terms equally intelligible to natives and newcomers alike. The names o f a very few tow ns, ports, and fortresses of Rom an origin mainly in the south-east (London, Dover, Lym pne, Reculver, etc.) were taken over more or less unaltered, doubtless because they had long been familiar to seafaring folk o f barbarian as well as native antecedents on both sides o f the Channel and the N orth Sea. Some of the larger Roman tow ns, whose remains, whether inhabited or n o t, must always have been as conspicuous features of the landscape as rivers, hills, and forests, retained their names in a barbarously truncated form, combined generally with ceaster corrupted from Latin castra, to suggest a man-made stronghold.1 Others, like Leicester or Chichester, lost all trace o f their Roman names while still visibly recog nizable as fortified castra. More significantly perhaps, none, with the possible exception o f C anterbury, developed a name based on that o f the native tribe o r civitas o f which it had formerly been the administrative centre, as happened almost universally in northern and central Gaul.12 The names of smaller tow ns, vici, and substantial villages were almost all lost. Lost also—and this must be a m atter o f supreme historical significance—were all the names o f those Romano-British rural estates whose owners had form ed the aristocracy of w ealth, society, and culture, certainly as late as the end of the fourth century, perhaps in some cases far into the fifth. Once again the com plete contrast with Roman Gaul is illuminating. Countless French villages still retain in their 1 As in W in c h este r (V e n ta ) , G lo u c e s te r (G le v u m ), o r C ire n c e ste r (C o rin iu m ). S o m e tim e s as a t C h e ste r, C a isto r-o n -th e-W o ld s, o r C a isto r-b y -N o rw ic h , th e w o rd c e a ste r w as u se d b y its e lf, sh o w in g th a t th e R o m a n o -B ritish n a m e o f th e place h a d b e e n to ta lly f o r g o tte n . C ea ster w a s also o c c a sio n a lly in c o rp o ra te d in th e n a m e s given t o R o m a n v illas, as a t W o o d c h e ste r, G lo u c s., su g g estin g th a t such p la c e s m a y have serv ed a p a ra -m ilita ry p u rp o se in th e ir fin al p h a se s . K . H . J a c k s o n , in L a n g u a g e a n d H is to r y , 2 5 2 , h a s p o in te d o u t th a t c e a ste r as a d e riv a tiv e fro m L atin castra is fo u n d in n o o t h e r G e rm a n ic lan g u a g e. I t is th e r e f o re lik e ly to be a b o rro w in g in to A n g lo -S ax o n fro m L a tin sp e a k e rs in t h e R o m a n to w n s o f B rita in ; th is m ig h t have b e e n h a p p e n in g b e fo re th e fin al co llap se o f R o m a n rule. 2 As a t P aris (L u te tia P a risio ru m ), R e im s (D u r o c p r to r u m R e m o r u m ) , o r A m ie n s (S a m a ro b riva A m b ia n o r u m ).
32
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
present names the adjectival form o f the nom en o f the GalloRoman family to which the estate from which the present settlem ent grew had belonged in the fourth o r fifth century. Not a single Romano-British villa appears to have left its ow ner’s name incorporated in that o f a m odern village in this way. Indeed we do not know with any certainty the names o f any o f the hundreds o f Roman villas whose sites and remains are scattered over the countryside o f lowland Britain. This fact must render extrem ely unlikely the notion som e times proposed that barbarian settlem ent may have taken place in parts of Britain within the framework o f existing estates. Had the m ethod of absorbing barbarian settlers by the legal procedures o f tertiatio and hospitalitas, as practised in Gaul, been followed so extensively in this country, it would m ost likely have left some such traces in o u r village names as are found in one form or another all over France. Even though there was in Britain an almost complete substitution o f Germanic for Romano-British rural placenames at this tim e, recent w ork has shown that a few* of the new names incorporated Latin words which must still have had significant meanings at the time when they came into use. This is an im portant outcom e o f place-name study, for it suggests th a t some at least o f the earliest settlers were sufficiently conversant with spoken Latin to adopt words from that tongue to describe things for which they had no appropriate words o f their own. The m ost obvious cases relate to buildings o r groups o f buildings that had, or appeared to have had, a special function. In addition to ceaster already m entioned, the clearest instances are u ic from Latin vicus, ecles from Latin ecclesia, and fu n ta from Latin fontana. Fontana, giving rise to such names as Fovant, Havant, Cheshunt, etc., is the least significant historically, for it is merely a descriptive term for any structure, large or small, designed to ease the collection o f w ater from a spring or well, w hether it were only a simple well-head on a farm or a decorative fountain in a town. The incorporation o [ fontana in a place-name probably implies either the nearby presence o f Romano-British buildings whose owners had once provided such a facility, even if its usefulness had long outlasted their
LATIN PLACE-NAME ELEMENTS
3 3
own occupation,1 o r a source o f water believed to have some unusual virtue or beneficent quality th a t justified the expense of making it a special architectural feature. Vicus and ecclesia are m ore interesting. Ecclesia, as in Eccles, Eccleshall, Eccleston, etc., seems to imply th e survival, or at least the recognizable remains, o f a Christian church. Ehe distribution o f such names is very distinctive. O f the simple form Eccles there are two examples, b o th adjacent to Roman roads, in Norfolk and there is one in K ent, adjacent to the site o f a substantial Rom an villa. This, like the great establishm ent n o t far away at Lullingstone, may well have been the property o f an aristocratic family which maintained a special building for Christian services. Most o f the other Eccles names, including all the com pound forms, are away in the north and west Midlands, Cumbria, I^mcashirc, West Yorkshire, Cheshire, Staffordshire, and Worcestershire, where there was little Anglo-Saxon penetration before the invaders became Christian. They are thus more likely to derive from early forms o f Welsh eglwys than direct from Latin ecclesia y and may not even refer to churches old enough to have survived from Romano-British times. But the three examples in Norfolk and Kent can only be explained as exceptional survivals o f Roman churches still sufficiently Conspicuous in the early days o f Anglo-Saxon settlem ent to justify specific recognition in this way.2 Even more significant may b e the com bination o f wie from Latin vicus with the early Germanic habitative wrord ham, to produce names now appearing variously on o u r maps as Wickham, Wykeham, Wycombe, and so on. It is natural to interpret these as indicating a settlem ent with the special qualities of a Roman vicus.3 In earlier Roman times vicus had been a formal designation for the smallest units o f local adm inistration in the provinces. It had been used to describe, for exam ple, the lesser towns and some substantial villages, the civil settlem ents that grew up outside military establish ments, and the local sub-divisions o f some large centres of 1 C h c s h u n t, fo r e x a m p le , p ro b a b ly m e a n s 'fo n ta n a a t o r n e a r a c e a ste r’. ‘ K . C a m e ro n , 'E c c le s in E nglish p la c e -n a m e s’ in B arley a n d H a n so n (e d s.), C h ristia n ity in B r ita in , 8 7 - 9 2 . } T h e in te re s t o f th e s e n a m e s was first p o in te d o u t b y M. G e llin g in M ed. A r c h . 11 ( 1 9 6 7 ) , 8 7 - 1 0 4 . See also h e r S ig n p o sts to th e P ast (1 9 7 8 ), 6 7 -7 4 .
34
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
urban population.1 It is probable that by the fourth century vicus had lost m ost o f w hat official legal and administrative meaning it may once have had, though it may still have remained in use as an unofficial status symbol. In any case the word was evidently familiar to the earliest barbarian settlers in the Rom an provinces. In the form wie it was taken over by the Anglo-Saxons and had a long life, acquiring in the process a wide range o f special senses for communities engaged in various agricultural and industrial pursuits. It may be argued th at, in view o f these later developments, the com bination wicham need have no unusual significance historically. But there are reasons for thinking that it does mainly belong to the earliest phases of the settlem ent and may thus have had a special meaning at that time. The mere fact that wie was m ore frequently combined with ham than with any other word for a settlem ent is itself suggestive of an early date. Ham belongs, apparently, to the most primitive phase o f local nom enclature used by the first barbarian settlers, and is thus less likely to have been combined with wie in any o f its later senses. More significant is the distribution o f the wicham names, which tend to follow a m arkedly Roman p attern. They are found n o t merely on, or in the immediate vicinity of, Roman roads, but often in close association with the small Roman towns or other substantial settlem ents which these roads connected. Many o f these were places o f a kind that might properly be described as vici in Roman times. It has thus been suggested that in calling a place wicham the first AngloSaxon arrivals were consciously indicating their awareness th at it had been a focus o f Romano-British population and perhaps o f some administrative activity. It is of course im portant not to press the evidence too far. These names in themselves do not prove either the continued existence o f whatever functions a vicus might still be exer cising when barbarians first settled in its neighbourhood, or even th a t any Romanized Britons were still living there. A ttem pts which have been made to use the wicham places which later became parishes as indicating by their boundaries F o r a d isc u ssio n o f v ic i in lo w la n d B ritain se e S. J o h n s o n in W . R o d w e ll a n d T . R o w le y (e d s .), T h e S m a ll T o w n s o f R o m a n B r ita in , B A R 15 (1 9 7 5 ).
VICUS AND
VILLA
35
the survival o f pre-Saxon administrative units, or to argue from the apparent absence o f early Saxon settlem ent near them for the persistence o f undisturbed Romano-British communities, are going far beyond the evidence. All that these names can be used to suggest is the possible barbarian awareness o f a few still recognizable institutional relics o f the last days o f rural life in Roman Britain. It is certainly interesting, and may well be significant, that whereas vicus was thus taken over into the small repertoire of Latin words used by the Anglo-Saxons in coining their place-names, villa, though it was used in this w'ay extensively in Frankish Gaul, has left little or no trace o f similar adop tion in Britain. It would thus seem that a group o f houses and associated structures, generally sited on a still usable Roman road, might be recognized here and there by early AngloSaxons as being, or having once been, a vicus. On the other hand the remains o f a country house o r large farmstead, once the centre o f an agricultural estate, whether o r n o t closely associated with a surviving road, would not normally be described by them as a villa. This term seems to have had no familiar significance in com m on speech when the new placenames were being form ed.1 Its disappearance is clearly relevant to the similar disappearance already noted, o f villa names derived from those o f the Romano-British families th at had once owned them. This contrast in survival value between vicus and villa in the spoken language o f pre-literate Anglo-Saxon folk is the more remarkable because, when w ritten Latin records came to be made in the seventh century, both vicus and villa were once again familiar terms meaning apparently much the same thing in the formal descriptions o f rural estates. It has recently been argued2 th a t Bede’s usage o f b o th was normally to indicate not single tenem ents b u t substantial blocks of property. These might include smaller communities and were sometimes distinguished as royal possessions o f administrative importance by the use o f such phrases as villa regalis or vicus regius. 1 It d o e s n o t a p p e a r, f o r e x a m p le , a m o n g t h e p la c e -n a m e e le m e n ts fo u n d in E n g lish d o c u m e n ts e a rlie r th a n 7 3 0 , a sse m b le d b y B. C o x , o p . c it. 6 6 ; sec also F. W. M a itlan d , D o m e s d a y B o o k a n d B e y o n d (C a m b rid g e , 1 8 9 7 ), 3 3 3 . 2 See J . C a m p b e ll in P. II. S a w y e r (e d .), N a m e s, W ords, a n d G raves (L eed s, 1 9 7 9 ), 4 3 - 8 .
36
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
It would be tem pting to seek the origin o f such properties in the Roman period, when villa might be a more appropriate description for some o f them than vicus. But it is well to remember the distinct histories o f vicus and villa in the illiterate age that preceded the Anglo-Saxon use o f written record. Vicus could then still be used for a recognizable settlem ent of some kind, though no doubt often for one associated with the personal lordship familiar to the AngloSaxons. Villa on the other hand seems to have gone o u t of common speech altogether, and with it no doubt had gone the institution it had once described. Its revival in literate times must have been largely due to influences from the continuing tradition o f legal Latin preserved in the courts of Frankish Gaul. If this was so, its later use in this country may have little bearing on the possible survival into Anglo-Saxon times of any rural institution which might have been properly called a villa in the last days o f Roman Britain. Most o f the new settlem ent names scattered over the countryside by Anglo-Saxon intm ders can be broadly grouped in three main categories, though these are by no means mutually exclusive. There are firstly folk-names of the -ingas type, which strictly speaking are not place-names at all b u t the names o f com m unities living in certain areas, large or small, which had become identified with them . Secondly there are habitative names such as ham, tun, and other such w’ords indicative o f human settlem ents o r o f the farm buildings, enclosures, agricultural o r industrial works, and so on which such settlem ents required. Names o f this kind arc frequently combined with folk- o r personal names, thus serving to identify the addresses o f their owners or operators. Thirdly there arc names merely descriptive o f the locality, by reference to topographical features such as hills, valleys, streams, and so on, or to the different sorts o f vegetation, fauna, o r flora, which were sufficiently characteristic of the place to be used to identify it. This third category need not be further discussed here, for it is obvious that such descrip tive names could arise at any time in the Anglo-Saxon centuries, and need carry no historical implications limited to the earliest period, although their distribution is clearly relevant to the appearance o f the landscape at the time they were first used. The folk-names and habitative names, in their
THE I N G A S NAMES
37
widest sense, are, on the other hand, pointers to many aspects of the settlem ent period, and it is very im portant to interpret them as correctly as possible, and to extract from them all the historical inform ation which they can properly supply. Until very recently it was the universal assumption among place-name students that the folk-names in -ingas were the earliest group, and that their distribution could be confi dently relied on to indicate areas of primary settlem ent by the Anglo-Saxon peoples in the first century or two after their arrival in this country. This doctrine will be found enshrined not only in the specialist literature b u t in most of the standard histories o f this age. It is implicit in all the early work of the English Place-Name Society, and is accepted w ithout question in the county volumes issued by that admirable body until less than ten years ago. The origins o f this doctrine go back to a very respectable antiquity in the pioneering days o f Anglo-Saxon studies before the middle o f the nineteenth century. J . M. Kemble, for exam ple, used the evidence o f -ingas names when he was developing his mark theory o f the settlem ent, which for many years made a powerful impression on historical thought.1 This view o f the -ingas names survived more or less unscathed the criticisms which eventually led to the mark theory, at least in its more extrem e form, being generally abandoned by historians. There is indeed a great deal to be said for the notion that folk-names o f the -ingas type are likely to have arisen in the earliest stages of the settlem ent, when the com m unities whose names they bear were first imposing themselves on the more or less derelict countryside of Roman Britain. This might well seem the only phase which left room for names to be given which were actually those of the newcomers themselves. The general distribution o f the type w ith its main concentration in the south-eastern counties, Essex, Sussex, and K ent, regions which must have borne the full weight o f barbarian penetration in the earliest days, is m ost naturally explained in this way. O ther consider ations point to the same conclusion. Thus large num bers of early m onothem atic personal names were used to form -ingas i
T h e S a x o n s in E n g la n d (1 8 4 9 ), i. 5 8 - 6 4 .
3 8
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
place-names, and a very high proportion o f them remained as the centres o f substantial estates and large medieval parishes. Others are even referred to as ‘regions' or ‘provinces’ in docum ents preceding the later administrative pattern of shires and hundreds.1 All this points to an archaic quality th at cannot be readily paralleled in any other type of placename commonly used by the English settlers. Yet when the -ingas names are plotted on a map that also carries the pagan Anglo-Saxon cemeteries, there is an odd lack of detailed coherence between the tw o distributions. Both cover in general the same south-eastern quarter of England, w ith few examples o f either to be found north and west of a line from Southam pton to Gloucester and thence north-eastward to the Yorkshire coast. But w ithin these limits there are areas such as Essex which have many -ingas but few cemeteries, while the neighbouring Cambridge region has many cemeteries but few -ingas. In the Thames Valley there is a string o f these ancient names following the river all the wray almost from its estuary past Sonning and Reading to the Goring gap, with very scanty archaeological counter parts except at Reading itself. But westward beyond the Goring gap the picture is reversed w ith numerous early cemeteries and very few -ingas names to m atch them. In commenting on this curious contrast, which became very obvious when the Ordnance Survey’s Map o f Britain in the Dark Ages was first published in 1935, I suggested that the explanation might partly lie in the quite different potential for survival o f the archaeological remains and the original names o f these early com m unities.2 Most archaeological 1 T h u s S o n n in g in th e T h a m e s V alley is c a lle d p ro v in c ia q u a e v o c a tu r S u n n in g e s in a n e ig h th -c e n tu ry c h a r te r (C S 3 4 ). 7 A n t i q u i t y , 9 (1 9 3 5 ) , 4 5 5 - 6 4 . A . G o o d ie r h a s p o in te d o u t (M ed. A rc h . 28 ( 1 9 8 4 ), 1 -2 1 ) w h a t c o u ld b e a sig n ific a n t lin k b e tw e e n th e lo c a tio n o f so m e p a g an c e m e te rie s a n d th e e s ta te b o u n d a rie s re c o rd e d in c h a rte rs w h ic h m a y p reserv e v e ry e a rly te rrito ria l a rra n g e m e n ts . If m a n y e a rly c e m e te rie s w e re th u s p la c e d on th e m arg in s r a th e r th a n n e a r th e c e n tre s o f th e te rrito rie s n a m e d a f t e r -ingas c o m m u n itie s , th a t w o u ld h e lp t o e x p la in th e p a u c ity o f o b v io u s to p o g ra p h ic a l lin k s b e tw e e n t h e a rc h a e o lo g ic a l an d p la c e -n a m e e v id en ce fo r th e s e e a rly s e ttle m e n t p a tte rn s . G o o d ie r h a s n o t m e n tio n e d a n y sig n ific a n t p lace-n am es in c o n sid e rin g th e p o ssib le c o n n e c tio n b e tw e e n th e c e m e te rie s a n d th e e s ta te b o u n d a rie s re c o rd e d in c h a rte rs . T h e use o f th e p h ra s e ‘h e a th e n b u ria ls ’ in c h a r te r b o u n d a rie s , generallyta k e n to in d ic a te re c o g n iz a b le p re h is to ric re m a in s, c o u ld have th e m o re re s tric te d m e a n in g o f ‘p re -C h ristia n A n g lo -S ax o n c e m e te rie s ’, w h ic h h a d b e e n d isu se d a fte r th e c o n v e rsio n a n d re p la c e d b y m o re c e n tra lly s itu a te d c h u rc h y a rd s.
-INGAS AND ARCHAEOLOGY
3 9
remains that have survived for discovery and record in modern times have done so because, being situated in what is now' open country, they have been least subject to later disturbance. But for th a t very reason they are m ost likely to have belonged to com m unities which have either died out or were transferred elsewhere, so that their original names have been lost. On the other hand the communities that have retained their early names may have done so just because they have not been thus disturbed. In their case, however, the effect o f many centuries o f building and rebuilding on the same limited area may either have destroyed piecemeal their earliest remains or have made them inaccessible beneath a succession o f later structures. Thus, where early names of this kind are com m on one should not expect as much archaeological evidence to be available as in those districts where they have m ostly disappeared. It may well be relevant in this context th a t Essex, Sussex, and Kent, where -ingas names are m ost frequent, suffered on the whole much less from disturbance and devastation by neighbouring rulers in the later Saxon centuries than, for example, the Cambridge region, part o f which was for long a vulnerable frontier district o f the East Anglian kingdom defended against Mercian attack by massive earthworks. So to o the upper Thames Valley, which was continually in dispute between the rival powers o f Mercia and Wessex, has few -ingas names b u t a great wealth o f early Anglo-Saxon remains. Folk-names o f the -ingas type, which belonged to groups of people rather than specific places, m ust have been peculiarly susceptible to disappearance w ithout trace in the fluid conditions that prevailed before a more o r less stable pattern of settlem ent took perm anent shape. Opinions may well differ on the extent to which consider ations of this kind help to account for the fact that regions where early settlem ent is proved by copious archaeological remains do not everywhere correspond in detail w ith those in which -ingas names are most frequent. In recent years some scholars have emphasized the fact that -ingas names were certainly still being form ed after the spread o f Christianity in the seventh century was putting an end to the practice of burial in pagan cemeteries. They have also noted that these names are apt to be found in districts unlikely on geological
40
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
or other environmental grounds to have been attractive to the earliest settlers, such as the still heavily forested parts of Essex. They have accordingly concluded th at, so far from being indicative o f prim ary settlem ent, the -ingas names as a whole belong to secondary phases o f colonizing activity. They are thus thought o f as spreading out from those parts where the first settlers have left traces o f their presence cither in the pagan cemeteries o r by the use o f habitative names believed to be more prim itive.1 O f these names the most significant is certainly -ham.2 Names in -ham occur plentcously in the eastern and southern parts o f Britain as a whole. They also show a marked relation ship to Roman roads and settlem ents, and match closely in detail the archaeological distribution o f pagan Anglo-Saxon remains. -Ham is found com pounded not only with terms descriptive o f local features, in themselves undatable, but often with personal names that are of early m onothem atic type. It is commonly linked with the -inga- element itself, thus producing names in -ingaham. This has given rise to the suggestion th at, in place of the generally received sequence of these name-forms in which the -ingas were thought to come first followed by -ingaham and -ham, the reverse order should now be accepted with the -ham names as the most primitive, followed by -ingaham. Only later would then come the -ingas marking secondary settlem ents in the w ood land and waste peripheral to the first centres o f immigration. This theory first took shape on the basis o f a detailed study o f Kent and the adjacent south-eastern parts. It has been developed and in part modified following exam ination of the rather different pattern apparently revealed in some regions o f East Anglia and the eastern Midlands.3 Almost 1 E v id en ce fro m so m e p a r ts o f th e C o n tin e n t suggests th a t th e n o tio n th a t -ingas n a m e s o n ly re p re s e n t s e c o n d a ry s e ttle m e n ts s h o u ld b e v iew ed w ith sc e p ti c ism . T h u s, in th e B altic island o f Ö la n d , fin d s fro m th e R o m a n Iro n A ge have b e e n f o u n d in clo se a sso c ia tio n w ith n o less th a n n in e o f th e th ir te e n s e ttle m e n t sites w ith n a m e s in -inge, a n d fin d s o f th is p e rio d a p p e a r o n ly to o c c u r o n sites b e a rin g th is t y p e o f n a m e : see C . I. S tä h le . S tu d ie r o v e r d e sv e n sk a o r tn a m n e n pa -inge (U p p sala, 1946). : T ile fu lle st stu d ie s o f -ham n a m e s a re b y B. C o x in J E P -N S 5 (1 9 7 3 ), 1 5 7 3 , a n d b y J . M. D o d g so n in A n g lo -S a x o n E n g la n d 2 (1 9 7 3 ) , 1 -5 0 . 3 T h e a rg u m e n t c a n b e st b e s tu d ie d in th e a rtic le s b y D o d g so n in M e d . A rch . 10 ( 1 9 6 6 ), 1 -2 9 ; B. C o x , o p . c it.; a n d K. C a m e ro n in P B A 6 2 (1 9 7 6 ). 1 3 5 - 5 6 . Sec also M . G e llin g , S ig n p o s ts , 1 0 6 - 2 9 , a n d D o d g so n in P. B ra n d o n ( c d .) . The S o u th S a x o n s (1 9 7 8 ), 5 4 -8 8 .
DATE OF - I N G A S NAMES
41
certainly m ore factors are involved in the problem than have hitherto been generally appreciated. It is inherently im prob able that the com pound form -ingaham should precede in time the use o f the simpler -ingas, since this com pound is clearly made up o f tw o elements both o f which must have had a familiar meaning before they were combined. More over, while it is true that -ingas names might be expected to occur more frequently than they do in direct association with pagan cemeteries, there are some notable cases where this direct association is evident, such as Reading (Berks.), Kettering (N orthants.), and possibly Mucking (Essex).1 Some -ingas names were certainly formed, o r at least came to be attached to specific localities, at a later date than would be expected had this form been used exclusively in the earliest days o f the settlem ent. It clearly had a longer vogue than was at one time supposed, b u t this has no bearing on the question o f the date when these names first began to be used. More consideration should clearly be given to the factors governing the survival or loss o f early names in the confused conditions o f the age o f settlem ent. Names o f the -ingas type, which were essentially the names o f people rather than places, would be particularly susceptible to transfer from one place to another while the folk concerned retained any degree of m obility, o r indeed to total disappearance in the event of disaster overtaking them . Many o f the communities whose existence is now only known from pagan cemeteries of the fifth and sixth centuries may well have been called by -ingas names, which either no longer exist12*7 or have been transferred to villages elsewhere which only took shape when their people finally settled down or became Christian. 1 D o u b ts have re c e n tly b e e n raised a b o u t th e c la ssific a tio n o f M u c k in g as an -ingas n a m e : sin ce th e re a re n o spellings k n o w n b e fo re th e N o rm a n p e rio d , th e m a tte r m u st c le arly b e le ft o p e n . B u t o n e c a n n o t fail t o n o te th a t th e q u e s tio n w o u ld n e v e r have b e e n raised a t all h ad n o t t h e d isc o v ery o f th e v e ry im p o rta n t e a rly s e ttle m e n t a n d c e m e te rie s a t M ucking se e m e d to th re a te n th e n e w n o tio n th a t -ingas n a m e s w ere o n ly given t o s e c o n d a ry s e ttle m e n ts : sec G e llin g , o p . c it. 1 1 9 -2 2 . 7 T h u s th e lo s t N id in g h a m in G irto n p a ris h , C a m b s., p ro b a b ly p reserv es the n a m e o f th e fo lk w h ic h u se d th e w e ll-k n o w n G irto n pagan c e m e te ry n o w generally k n o w n b y t h e n a m e o f th e a d ja c e n t villag e, w h ic h is o f a ty p e u n lik e ly to b e as e a rly as th e p a g an p e rio d : B. C o x in K . C a m e ro n (c d .), P la ce-N a m e E v id en c e f o r th e A n g lo -S a x o n In v a sio n a n d S ca n d in a via n S e ttle m e n ts (N o ttin g h a m , 1 9 7 5 ), 5 8 .
42
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
Considerations o f this kind are likely to prove significant in evaluating the historical evidence provided by these early names. It is inherently likely th a t many -ingas names now attached to particular villages distant from a pagan cemetery or situated on land unattractive to the earliest settlers, were at first applied to much wider areas which the folk in ques tion had made their own, perhaps as far back as the fifth century o r even earlier. A num ber o f such districts bearing names o f this kind survived into historic times, where they appear among the smaller units recorded in the Tribal Hidage of eighth-century Mercia1 or in land charters o f that and succeeding centuries. Many others must have lost their original identity by various forms o f disintegration, som e times so com pletely as to be no longer recognizable at all. It is not surprising th at, as the generations passed, such disintegration should have occurred. The original links that bound the members o f such com m unities together arose from real or imaginary kinship with the eponym ous founder whose name they bore, o r through original membership of a warband o r com itatus which had looked to him for leadership. They would inevitably become looser as the founding fathers passed out o f living memory and some o f their followers hived o ff to the exploitation o f fresh parts o f the district. Family connections would eventually be forgotten as the population grew and the degree o f cousinship became more distant. The process o f disintegration can be studied in its various stages by examining particular cases. Thus an early stage, which happened to become perm anent, may be indicated by the group of Roding villages in Essex, all o f which after becoming separate com m unities retained the original name ofthe folk from which they all claimed to spring.2 Two 1 Sec b e lo w , p . 141. * It has b e e n su g g e ste d w ith so m e p la u s ib ility th a t th e E ssex R o d in g a s m ay h a v e ta k e n th e ir n a m e fro m th e R o d in g as w h o a p p e a r as a c o n tin e n ta l tr ib e in W id sith , d e sc e n d e d p e rh a p s u ltim a te ly fro m th e R c u d in g i o f T a c itu s . T h e R e u d in g i a rc p r o b a b ly re p re s e n te d a rc h a c o lo g ic a lly b y th e p ro to -S a x o n F u h ls b ü tte l c u ltu re , d isc u sse d b y F . T is c h le r, F u h ls b ü tte l, e in B eitrag z u r S a ch sen fra g e (N c u m ü n s tc r, 1 9 3 7 ). G . O s te n (N ied ersä ch sisch es J a h r b u c h f ü r iM tid e sg e sc h ic h te , 51 (1 9 7 9 ), 102) is m is ta k e n in p la c in g th e E nglish R o d in g a s o n w h a t h e c alls ‘th e East A n g lian H e ig h ts ': th e ir s e ttle m e n ts c o v e r a n a re a o f so m e tw e n ty m ile s a c ro ss in th e h e a rt o f E ssex, n o rth w e s t o f C h e lm sfo rd , a n d a rc th u s m o re lik e ly t o d eriv e fro m a S a x o n th a n a n A ngle b a c k g ro u n d o n th e C o n tin e n t.
• I N G A S AS FOLK-NAMES
43
adjacent group settlem ents o f this kind in the middle Thames Valley, Sonning and Reading, which both retained their original extensive outlines by becoming enormous medieval manors, illustrate different ways in which development might take place. At Reading, where a group o f pagan cemeteries bears witness to the great antiquity o f the settlem ent at the confluence o f the Thames and the Kennet, the folk-name became localized at this conspicuous centre. It is not Found in any of the subsidiary villages th a t must have been formed by outlying groups o f Radingas established in rem oter parts of w hat became the later m anor. At Sonning, on the other hand, described as the provincia quae vocatur Sunninges in an eighth-century charter,1 there was no such obvious centre. The folk-name was eventually localized, perhaps not until Christian times, at a spot almost on the border o f its original territory, possibly as a deliberate indication that that was where the land which the Sunningas claimed began. But some detached members o f this folk left perm anent traces o f their presence in Berkshire far beyond the com pact limits of their main provincia at such spots to the east and west respectively as Sunninghili and Sunningwell,12 suggesting a foot-loose tendency that persisted much longer than was the case else where. So too it might well happen that a subsidiary settle m ent in the territory o f the Woccingas might be established at Wokingham earlier than the folk-name itself became localized at Woking, when the process of disintegration had eventually run its full course, and there was no longer any need to retain it for the wider area it had once covered. By no means all the early -ingas com m unities were o f the size to be remembered as provinciae in the charters, or to be assessed as separate taxable units by their Mercian overlords when the Tribal Hidage was drawn up. In Sussex, for example, while the Ilaestingas o f Hastings were certainly o f this sub stantial character, the numerous -ingas names on the coastal plain between the South Downs and the sea in West Sussex present a very different pattern. Here, where several such names may survive w ithin a single parish, it would seem that 1 C S 34. 2 P e rh a p s e v e n a t S u n b u ry , M id d lesex , w h e re a m an w ith t h e sam e n am e, S u n n a , a s th e fo u n d e r o f th e S u n n in g a s, m u s t have h a d a d e fe n sib le s e ttle m e n t in e a rly tim es.
44
II. A R C H A E O L O G Y A N D P L A C E - N A M E S
they can represent little more than the holdings of one family group, such as gave rise elsewhere to early names in -ham. -Ilam names arc in fact infrequent hereabouts and the fact that com pounds in -ingaham arc also uncom m on in Sussex suggests th a t there was little room here for the kind of expansion that led in other parts to the eventual disintegration o f the larger -ingas communities. Such would appear to be the most significant suggestions to be drawn from recent w ork on the earliest strata of English place-names. The subject is one which cannot be expected to produce many firm and generally acceptable conclusions to assist the historian in his work. Moreover, the present inconclusive condition o f these studies is aggravated by uncertainties that bedevil the interpretation o f several of w hat should be the most meaningful elements used in the earliest types o f names. Thus the greatest care has to be taken to distinguish among names now ending in -ing between those likely to be true folk-names o f the -ingas type and those containing the singular elem ent -ing in any of its less significant meanings. So too it may often be all b u t impossible to determ ine from the earliest surviving spellings w hether a modern name in -ham contains the early element -ham meaning a settlem ent, or the equally com m on -hamm meaning meadow or enclosed pasture. Hamm in these senses may have had a milch longer life as a descriptive term, and thus carries no such significant inform ation for the historian.1 These frustrating ambiguities, which the nature of the evidence in itself often makes it impossible to resolve, may sen e as final reminders o f the fundamental uncertainties inherent in all English place-name study. These uncertainties have always been recognized by the best scholars in this field, and it may perhaps not be o u t o f place to end this discussion with some wise words which one o f its foremost and most learned pioneers used to express his feelings on the m atter: ‘. . . i t is e s s e n t i a l t o r e m e m b e r t h a t i n t h e p r e s e n t s t a t e o f p l a c e - n a m e s tu d ie s th e s e re s u lts c a n o n ly b e te n ta tiv e , a n d t h a t e v e n w h e n th e p l a c e - n a m e s o f a ll E n g l a n d h a v e b e e n s u r v e y e d in t h e m i n u t e s t d e t a i l ,
1 F o r t h e e le m e n t -h a m m see th e a rtic le s b y M. G e llin g a n d K. I. S a n d re d in N a m n o c h B y g d , 4 8 (1 9 6 0 ), 1 4 0 -6 2 , a n d 6 4 (1 9 7 6 ), 7 1 - 8 7 .
VALUE OF PLACE-NAMES
45
t h e co n clu sio n s w hich m a y b e d ra w n f r o m th e m will fall far sh o rt of scientific p re c is io n .’1
In spite of the great increase in the volume o f available material and in the ever-expanding knowledge and expertise that can be brought to bear on its interpretation, those words are as true now as they were when Sir Frank S tenton wrote them more than forty years ago. 1 F . M. S te n to n , T R H S 4 t h sc r. 2 2 (1 9 4 0 ) , 21.
3 T H E C O N T IN E N T A L BACKGROUND O discussion of the continental background to the English settlem ents can start from any other point than Bede’s statem ent that the invaders came from three o f the most powerful peoples o f G erm any, the Angles, the Saxons, and the ju te s .1 That this should still be so more than twelve hundred years after Bede’s day, is a tribute to his remarkable capacity as a historian to express in simple but fundam entally accurate terms the essential facts of a process whose com plexity was probably at least as obvious to him as it is to the m odern student with a much wider range of evidence to evaluate. And when he goes on to locate the continental Saxons in the lands known in his day as those still inhabited by the Old Saxons, the Angles in a region known to him as Angulus which had been left em pty by their migration, and the Ju te s in a less well defined area situated somewhere beyond the Angles, he was also providing the essentials for a map o f the N orth Sea coastlands with all the political detail that was strictly necessary for his purpose. In deriving the Saxons from the districts occupied in his time by the Old Saxons, Bede is certainly pointing to the coastlands betw een the Elbe and the Weser valleys, where, as J . M. Kemble12 first dem onstrated more than a century ago, the cemeteries have produced much material directly parallel to that found in England. By Angulus, the region th at lay between the Saxons and the Ju tes, Bede certainly meant Schleswig, where the name is still found on m odern maps as the district o f Angeln. There has been much difference of opinion over the location o f Bede's Ju tes. But if, as he implies, they lay somewhere beyond the Angles, he must have thought o f them either to the north o f Schleswig, in Ju tlan d , o r to the east in m odern Holstein, wrhere again there
N
1 H E i. 15. 2 J . M . K e m b le , ‘O n m o rtu a ry u rn s a t S ta d e -o n -th e -E lb e , a n d o th e r p a rts o f n o r th G e rm a n y ’, A rc h a e o lo g ia , 3 6 (1 8 5 5 ), 2 7 0 -8 3 .
ANGLES, SAXONS, JUTES, AND FRISIANS
47
is a suggestive district name, Eutin, which may indicate the home of some part o f the tribe at one time. Archaeologically speaking, as will be seen later, Bede’s Ju tes need present no great problem. South Jutland and Fyn have close cultural links in the fifth century w ith Kent, and throughout eastern England at this time ceramic fashions derived directly or indirectly from East Holstein are con spicuously evident, though w ithout marked concentration in any particular areas. Both Jutland and East Holstein were closely related to the main south Scandinavian Kulturkreis that included the Angles o f Schleswig, though neither may have developed such recognizably distinctive fashions. It is not impossible that this rather imprecise relationship with the Angles was exactly the impression that Bede intended to convey by the vagueness o f his reference to the exact location of the Jutes. Bede does not in this passage include Frisians among the major peoples settling in Britain, though he does refer to them in this connection elsewhere.1 This may therefore be the right point to m ention that the literary evidence for their presence among the settlers goes back much earlier than Bede. The Byzantine historian Procopius2 writing in the sixth century, and apparently using inform ation provided by Angles who accompanied a Frankish mission to Ju stin ian ’s court, understood that in his day the population o f Britain, apart from the native Britons, was divided between the Angles and the Frisians. It may well be that the Franks, as their near neighbours on the lower Rhine, took an exaggerated view of the p art played by Frisians in the movement to Britain, but their participation was certainly far from negli gible. What makes it difficult to evaluate archaeologically is that, as will be seen later, the Anglo-Saxon peoples were them selves infiltrating the Frisian seaboard between the Weser and the Rhine m ouths in force at this time, many o f them as a preliminary to the short sea passage to south-eastern England. Thus the culture o f the Frisians was itself becoming very mixed, and it is scarcely possible to be certain from the archaeological evidence how substantial a contribution the Frisians themselves may have made to the movement in 1 H E V. 9 . * P ro c o p iu s, G o th ic W ar, iv . 20.
4 8
THE CONTINENTAL BACKGROUND
terms of manpower. Though it may well have been con siderable, especially in south-eastern Britain where w hat has been term ed ‘Anglo-Frisian’ pottery is especially common in the cem eteries and settlem ents,1 it did not result in the establishment of any specifically Frisian kingdom o r recog nizable Frisian institutions. This is no doubt the reason why Bede did not emphasize Frisian participation in the settle m ent. But the linguistic similarities that long linked the speech and dialects of south-eastern England to the Low Countries suggest th a t the physical influx o f Frisian folk may have been im portant.2 They certainly played a major part in the recovery o f English trade with the C ontinent, o f which the revival o f London, already recognizable in Bede's day as a focal centre of Frisian m erchants, was an early and conspicuous symbol. This, however, is to anticipate by two centuries or more the distant consequences o f possible Frisian collaboration in the early stages o f the movement. We need to enquire how it came about that the Angles, the Saxons, and the Jutes mentioned by Bede,' with or w ithout the substantial Frisian participation for which Procopius provides near-contem porary testim ony, were in a position to bring about the transfor mation o f w hat had been the civil provinces o f the Roman diocese o f Britannia into a loosely-knit group o f hereditary barbarian kingdoms. It was no part o f Bede’s purpose in writing his Ecclesiastical History o f the English Nation to describe the detail o f this transform ation, which was all but com pleted while the newcomers were still pagan. He must have known a great deal about the process, especially as it affected his native N orthum bria, but this knowledge was ’ See m y a rtic le , ‘S o m e E nglish p a ra lle ls to th e A n g lo -S ax o n p o tte r y o f H o lla n d a n d B elg iu m ’, in L 'A n t i q u it é c la ssiq u e, 17 (1 9 4 8 ), 4 5 3 - 7 2 . 7 C e rta in sp e c ific a lly F risian w o rd s fo u n d th e ir w ay in to E nglish p lace-n am es as n o te d b y S te n to n , P rep a ra to ry to A n g lo -S a x o n E n g la n d (O x fo rd , 1 9 7 0 ), 2 6 9 . T h e m o s t in te re s tin g is p e rh a p s r o th , a c le arin g , w h ic h o c c u rs in R o th w ell an d R o th le y , N o rth a n ts . E kw all h a s s h o w n th is to b e a n e x a c t p a ra lle l to O ld Frisian r o th e : see E P-N S N o r th a n ts . (1 9 3 3 ), 119. Its o c c u rre n c e in R o th w e ll e x a c tly m a tc h e s th e fin d in g th e re (a n d a t n e a rb y D e sb o ro u g h ) o f tw o sm all p o ts c le arly fro m th e sam e w o rk sh o p as th a t w h ic h p ro d u c e d th re e sim ila r vessels f o u n d at S t. G illes-les-T crm o n d c in B elgium : see m y C o rp u s, i. 2 7 a n d ii. fig. 143. A F risian c o n n e c tio n b e tw e e n N o rth a n ts . a n d th is p a r t o f B elgium in th e six th c e n tu r y is th u s im p lied b o th b y a rc h a e o lo g y a n d b y p lac e-n a m e s. 3 H E i. 15.
EVIDENCE OF TACITUS AND PTOLEMY
49
not relevant to the story he wanted to tell, and so is lost. So, too, it is impossible to guess how much or how little he may have known about the continental background o f the Angles, the Saxons, and the Ju tes, beyond the bare facts o f their political geography as he understood it. There are two main sources from which further inform ation on these m atters can be gathered. Roman historians and geographers, notably Tacitus and Ptolem y, were interested in the tribal situation in free Germany beyond the Imperial frontiers in the first and second centuries AD. They had occasion to refer to the peoples m entioned by Bede among many other tribes who had been their neighbours several hundred years before his time. There is also a wide range of archaeological material, extending in date from the earlier phases o f the pre-Roman Iron Age in north Germany and southern Scandinavia to the later stages o f the Migration period, which can be used to illustrate the cultural changes which preceded, accompanied, and followed the settlement of some o f these peoples in Britain. Much labour and learning has been devoted to attem pts to relate the archaeological evidence for the changes, develop ments, and movements in this cultural background to the activities o f particular tribes and confederations o f folk recorded in the historical sources. While some o f these equations are plausible enough, and can be used with some confidence as pointers to particular historical situations and relationships betw een the tribes in question, care has to be taken not to overstrain the evidence in identifying any given cultural assemblage o f material as necessarily the work of any tribe or group o f tribes whose names happen to be known. Writing towards the end o f the first century AD, Tacitus, in his account o f the Ingaevones, the group o f peoples wrho then occupied the north-western seaboard o f Germany, m entions, among many others less relevant to the present purpose, the Frisians, the Angles, and the Eudoses.1 The Frisians already occupied the coastland in northern Holland which still bears their name. The Angles and Eudoses, whose precise position at that time was probably no clearer to 1 T a c itu s , G erm ania, 40.
50
THE CONTINENTAL BACKGROUND
Tacitus than he makes it to us, could a t least be described as ‘defended by rivers and forests', in an area which made natural their com m on participation with other tribes in the rites of a divinity called N erthus, centred on an island of the Ocean. By the Ocean Tacitus probably m eant the N orth Sea rather than the Baltic. It is thus likely from his account that the Angles and Eudoses were occupying before AD 100 approxim ately the regions which Bede many centuries later seems to have thought appropriate for his Angles and Jutes. Tacitus, however, makes no m ention, in this passage or anywhere else, o f the Saxons. They first appear in the Geo graphy o f Ptolem y,1 a w riter who lived in the middle of the second century, though it is generally believed that he derived much o f his inform ation from a source about a hundred and fifty years earlier. Ptolemy placed the Saxons ‘on the neck o f the Cimbric peninsula’, by w’hich he probably meant somewhere in the m odern Holstein. On the coast between the Saxons and the Frisians — from the Elbe, that is, as far as the Ems — Ptolemy and Tacitus agree in locating the Chauci, an im portant tribe which, in the time o f Tacitus, could be described as ‘the noblest o f the German race'. They seem to have occupied much the same region in the first and second centuries as Bede’s Old Saxons were occupying in his time. Between these dates it would appear that the Saxons, moving south-west across the lower Elbe from the base of the Danish peninsula in Holstein, had somehow replaced, ejected, or absorbed the Chauci. They seem to have taken over both their lands and their role as pirates and sea-robbers, soon making themselves a major menace on the coasts o f Roman Gaul and Britain. These im portant changes, o f which no details of the political history are know n, seem to have occurred during the third century. After this the Chauci are no longer men tioned in these coastal regions, though they arc referred to in later Roman sources as participating in disturbances near the Imperial frontier on the middle Rhine. The archaeological evidence does not suggest th a t the changeover from the dominance o f the Chauci to th a t o f the Saxons in these parts was accompanied by any drastic o r sudden cultural change. 1 P to le m y , G eog. II. x i. 7.
CHAUCI AND SAXONS
51
The two peoples had long been neighbours in the lower Elbe valley exhibiting related styles o f decoration in pottery and metal-work, and these fashions continued to develop after the third-century’ movements w ithout any marked evidence for serious dislocation. It would thus seem likely that the westward extension o f Saxon power in the third century, while it broke up the old political hegemony o f the Chauci, did n o t displace the bulk o f the population, though some irreconcilable elements m ay have retained their name and independence by migrating south-westwards towards the middle Rhine. Much o f the archaeological evidence for the culture of the Chauci in these coastal regions before the third century, as for their successors the Saxons later, comes from the num erous Terpen o r m ound-settlem ents, thickly scattered on the low-lying lands and among the marshes th a t fringed the sea m ost o f the way from the Elbe estuary in the neigh bourhood o f Cuxhaven to the Frisian settlem ents in north Holland where much the same conditions prevailed.1 Roman writers, such as the elder Pliny,12 were aware o f the curious life half-way between the land and the sea which the Chauci had based on these artificial mounds. Pliny’s description of them is almost echoed later in the fourth century b y Orosius who writes o f the Saxons3 as ‘a people o f the Ocean settled in pathless swamps and on the sea shore’. Many of these artificial m ounds still carry m odern villages; others, which have been excavated by Dutch and German archaeologists, have throw n much light on the social evolution of the communities th a t occupied them in the centuries be fore and during the Migration period. Nearly everywhere the evidence points to a growing population, greater elaboration of buildings, and increasing pressure on the habitable area as the m ounds were raised and enlarged to accommodate more people and to counter the encroaching sea. At Feddersen Wierde, on the Weser estuary north o f Bremerhaven, for example, several small mounds with individual farm-houses 1 A v e ry u se fu l g en eral s tu d y o f th e s e m o u n d s w ith p la n s a n d illu stra tio n s is H . H a lb e rts m a , T e rp e n tu s s e n Vlie e n E e m s , 2 vols. (G ro n in g e n , 1 9 6 3 ). 2 P lin y , N a t. H ist. X V I. i. 2 - 5 . 3 O ro siu s, H ist, a d versu s P aganos, e d . C . Z a n g e m e iste r (L eip zig , 1 8 8 9 ), vii. 32.
52
THE CONTINENTAL BACKGROUND
were linked together in the first century AD to form one large Terp which soon came to carry as many as thirty houses radially arranged. In the second century a chieftain’s house in a ditched and fenced enclosure surrounded by workshops became the focus of some fifty houses, and later still granaries and a large hall were added w ith industrial installations for bronze- and iron-working. O ccupation ceased about the middle o f the fifth century, and it is tempting to connect this desertion with the possibility o f movement overseas to Britain.1 In any case the pottery and other objects from the final phase at Feddersen Wierde correspond very closely with those found in such early English settlem ents as that at Mucking (Essex), a site on the Thames estuary very similar in geographical terms to that o f Feddersen Wierde on the Weser. A village site recently excavated at Wijster (Looveen) in north Holland seems to show a somewhat similar history.2 It apparently began with a single farmstead in the second century AD which was several times rebuilt. By the third century there were at least three houses and the population was increasing. After a brief decline in the early fourth century there was soon a rapid expansion: many houses were built in rectangular blocks with lanes betw een, and there was evidence for great activity; b u t early in the fifth century occupation came to an end. Here too it is very tem pting not only to link the final desertion o f Wijster with a movem ent o f its people to Britain, but to see in the preceding phase o f increased activity the signs of Saxon pressure on the older Frisian population which proved too heavy for the local means o f subsistence to support. It will be necessary to say something shortly about the forces in free G erm any and Scandinavia which stimulated this movement westward o f the Saxons into the lands o f the Chauci and the Frisians. But it was hardlv accidental that it should have coincided in time with the increasingly disturbed conditions that troubled th e Roman world during the third century. There was great political confusion in Gaul at that *
1 In te rim re p o rts b y W . H aarn ag el a rc in G e rm a n ia , 3 4 (1 S 5 6 ), 3 5 (1 9 5 7 ), 3 9 ( 1 9 6 1 ) , a n d 4 1 (1 9 6 3 ). : W. A . v an E s, W ijster, a N a tiv e Village b e y o n d th e Im p e ria l F ro n tie r 1 5 0 4 2 5 A .D . (G ro n in g e n , 1 9 6 7 ).
CARAUSIUS AND THE SAXON SHORE
53
time, accompanied by severe inflation and economic weakness. The continuing uncertainties o f political control in the Empire were reflected in increasing irresponsibility and local independence among the generals on whom the burden of frontier defence lay. Such conditions proved an irresistible tem ptation to the restless tribes in movement beyond the frontier on the N orth Sea coast. It would seem th a t the Roman authorities attem pted to control these movements through the establishment o f a virtually independent naval com m and, based on fortified ports and harbour installations on both sides o f the Channel. This policy, however, only increased the political confusion, for the com m ander in charge, Carausius, soon took advantage o f his position in 287 to set up an independent empire o f his own. Carausius was himself a Menapian with a family tradition of piracy based on the Rhine m ouths, where his tribe belonged. It is no accident that the coastal installations and fortresses on which his pow er rested in Britain should have come later on to be known as those o f the ‘Saxon Shore’. For, as has been seen, it was precisely at this time, in the later part of the third century, that Saxon pressure on the Chauci and Frisians in the coastlands between the Elbe and the Rhine had created an unstable situation, which stimulated the more adventurous elements in all these peoples to seek fame and fortune through plunder, piracy, and even perhaps through settlem ent overseas. The British side o f this situation will require further discussion later on. It is enough here to point o u t th a t this is the earliest m om ent at which conditions on the Continent might have made possible a Saxon settle m ent on the coast o f Britain, sufficient in scale to give rise to the notion o f a ‘Saxon Shore’. As the fourth century wore on it would seem that the forces of nature took a hand in increasing the pressure of the Saxon population on the available means o f subsistence. It is probable that the whole north German coastline had been slowly sinking for some centuries, and that the widespread practice o f living on artificial mounds had arisen as m an’s natural reaction to the slow advance ol the sea. It would certainly appear that the ever-increasing size and height of the m ounds, which all the excavations have revealed, is to be explained not merely by the increase o f the population but
5 4
THE CONTINENTAL BACKGROUND
also by the rising level o f high water. There are reasons for believing that this sinkage, which eventually led in the Carolingian age to the form ation o f the Zuider Zee, the Dollart, and the Jad e Bay, was proceeding at an accelerated pace from the fourth century onw ards.1 While all the German tribes bordering on the Roman frontiers were watching with increased restlessness the weakening of its defences, and were themselves beginning to feel the westward pressure o f nomad peoples from Asia, the Saxons and Frisians crowded on the German coast were thus faced in addition with local troubles o f their own. Behind these local troubles, however, were the major movements southwards and westwards o f Scandinavian tribes, which had forced the Saxons from their earlier homes in Holstein across the Elbe into the lands o f the Chauci. They had also stim ulated the equally m om entous migration o f the Lombards up the Elbe valley tow ards the Alpine passes and their eventual settlem ent in north Italy. It was this pressure from southern Scandinavia which m ust have unsettled the Angles from their homeland around Angeln in eastern Schleswig, and their neighbours the Jutes from Ju tlan d and Fyn, perhaps in part from East Holstein also. German archaeologists have done much in recent years to sort o u t the various cultural strains which can be identified as contributing, from the pre-Roman Iron Age onwards, to w hat may be term ed the proto-Angie, proto-Saxon, or proto-Jutish phases o f Bede’s three familiar tribes.12 It is thus difficult to avoid the use o f such literary labels in distinguishing the m ore significant, for our purposes, of these related cultural groups. Prom inent among them in the first and second centuries AD had been the Fuhlsbüttel folk of Holstein, in whom can probably be detected the roots of the 1 A sim ila r sin k a g e o n th e e a s t c o a st o f B rita in led t o t h e flo o d in g an d a b a n d o n m e n t o f th e p ro s p e ro u s a g ric u ltu ra l e s ta te s e sta b lis h e d in e a rlie r R o m a n tim e s in th e F e n s. M o st o f th is re g io n h a d re v e rte d in th e fif th c e n tu r y , if n o t b e fo re , t o w a ste an d m a rsh . L o w -ly in g c o asta l s e ttle m e n ts in E ssex a n d in th e T h a m e s e s tu a ry also b e c a m e u n in h a b ita b le d u rin g th e fo u rth c e n tu ry fo r th e sam e re a so n . 2 S ee e sp e c ia lly : M . B. M a c k e p ra n g , K u ltu r b e z ie h u n g e n im n o rd isc h e n R a u m (L e ip z ig , 1 9 4 3 ); F. T is c h le r, 'D e r S ta n d d e r S a c h se n fo rs c h u n g arch äo lo g isch g e se h e n ’ in B e r ic h t d e r R ö m isc h -G e rm a n isc h e n K o m m is s io n , 3 5 (1 9 5 6 ). 2 1 2 1 5 ; A. G e n ric h , F o rm e n k re ise u n d S ta m m e s g r u p p e n in S c h le sw ig -H o lstein (N c u m ü n s tc r, 1 9 5 4 ).
COASTAL SINKAGE AND T R IB A L MOVEMENTS
55
culture later characteristic o f the Saxons.1 To the north, in eastern Schleswig and Fyn, the Angles were already recog nizable, and closely associated with them was the Obcrjersdal culture of southern Ju tlan d .2 This was soon to extend all down the west coast o f the Danish peninsula, including the N orth Frisian islands, to form a complex known to German scholars as the Westgruppe. Its northern elements probably included the ancestors o f the later Ju tes, while its southern boundary m arched w ith the Fuhlsbüttel people north o f the Elbe estuary. Increasing pressure from the north and east during the third and early fourth centuries seems to have had the effect of forcing many o f the smaller tribes to coalesce into larger com binations whose individual sections became less distinctive archaeologically as they merged. By the m id-fourth century, if not earlier, the western parts o f Ju tlan d and Schleswig were occupied by a broadly based Nordseekustengruppe. Its characteristic fashions in pottery and metal-work were beginning to spread across the estuary o f the Elbe, to mingle with th e Saxon com plex now occupying in force the lands between the lower Elbe and the Weser and the old territories of the Chauci beyond. N orth and east o f the lower Elbe, southward pressure from the Angles and possibly some neighbouring Ju tish folk in east Schleswig was mingling with and perhaps dom inating the remaining tribes in East Holstein, to produce what the Germans appropriately term a Mischgruppe incorporating some cultural features from all these elements. O ther Angle and Ju tish folk evidently joined the Saxon advance beyond the Ems into Frisia. Their presence there can be dem onstrated not only archaeologically, in the widespread use o f cruciform brooches and pottery o f so-called ‘Anglo-Frisian' types, but b y the literary traditions that link the Jutes with the Frisians in the tales about Hengist and other heroes o f which echoes survive in B eo w u lf and the Finnsburh fragment. It is not possible in this context (and w ithout more numerous illustrations than the plan o f this History can include) to describe in any detail the characteristic features 1 F. T is c h le r, F u h ls b ü tte l. 3 F . T is c h le r, Das G rä b e rfe ld O b erjersd a l (H a m b u rg , 1955).
56
THE CONTINENTAL BACKGROUND
of the equipm ent, in metal goods and pottery, which enable archaeologists to distinguish the various groups o f Germanic folk concerned in these movements which took the main part in the settlem ent of England. There are, however, certain broad differences o f fashion which can be m entioned as useful aids in recognizing the presence o f one or m ore of these peoples, or in estimating their relative im portance in the various sites o f occupation o r burial which provide the material evidence for them. More w ork has probably been done on the typology of brooches, on both their form and their decoration, than on any other group o f metal objects o f this age. As the principal items combining both functional use and decorative desir ability among the personal possessions o f wom en, brooches were not only indispensable parts o f their dress and equip m ent, b u t were peculiarly liable to ceaseless developm ent and evolution according to the changing dictates o f taste and fashion. M inor differences in form and ornam entation are thus likely to be especially significant, both for dating and for localizing individual styles and even individual pieces. This is n o t the place for entering into such m inutiae,' but some appreciation o f their existence is necessary to explain the im portance attached to the typology o f brooches, and to their value as pointers to the relative dates o f other less informative objects often found in association with them. The different types o f brooches certainly help to dis tinguish the cultures and pinpoint the early distribution of the peoples whom Bede knew as the Angles, the Saxons, and the Ju tes. Throughout southern Scandinavia, Ju tlan d , the Danish islands, and Schleswig the main fashion at this period was for various forms o f long brooch. This was a type whose coiled spring was concealed behind a head-plate at one end, while the pin was held when closed by a catch-plate at the back of the other. Between the two was a bow, varying in size and shape, which lent itself to decorative treatm ent. Two main types o f long brooches developed in these northern regions. In one the head-plate was square or rectangular in shape entirely covering the spring, while the foot, which similarly covered the catch-plate, was often diamond- or lozenge-shaped and sometimes divided down the centre by a ridge into tw o parts on slightly different planes.
CRUCIFORM AND CIRCULAR BROOCHES
57
All three sections, head, bow, and foot, lent themselves to decoration in ways that were fashionable at the date of m anufacture. A few early examples show scroll patterns carried out in the chip-carving technique. But the type developed mainly in the sixth century with ever more elaborate zoom orphic decoration. This came to include marginal excrescences around the edges o f the head-plate and lobes projecting from the sides and bottom o f the foot. These generally carried Style 1 animal ornam ent o f increasing com plexity as time w ent o n .1 The other main type o f these long brooches has a more significant early history. It is known as the cruciform, because the ends o f the spring extend beyond the edges of the head-plate and are covered with protruding lateral knobs, whose form is repeated in a third knob on the top edge (Fig. 1). These knobs in the early examples were separately attached and o f simple rounded form, but they gradually became enlarged and flattened to form decorative features o f the head-plate itself, no longer having any functional purpose. The feet o f these cruciform brooches also show a characteristic development. At first narrow and almost featureless, they are soon provided with a small eye on each side and nostrils at the end to become em bryo horses’ heads. Inevitably the Germanic fondness for animal ornam ent leads to these simple zoom orphic features being developed and exaggerated in an increasingly elaborate and even grotesque manner to m atch the exuberant taste o f the sixth century. This very simplified outline o f what is in fact a quite complex typological evolution m ay help to show how these brooches can provide clues o f considerable value as historical evidence. Whenever square-headed brooches o f the kind with divided foot or any sort o f cruciform brooches appear among the grave goods o r settlem ent debris o f this period, whether on the C ontinent or in Britain, one can be reasonably certain that the folk to whom they belonged had links, direct or indirect, personal o r ancestral, with one o r other o f the ! T h e m o s t c o m p le te s tu d y o f th e sq u a re -h e a d e d b ro o c h e s is E. T . L eed s, A C o rp u s o f E arly A n g lo -S a x o n G reat S q u a re -h e a d e d B ro o c h e s (O x fo rd , 1949} w h ic h , th o u g h n o w in n e e d o f so m e re v isio n , re m a in s th e e sse n tia l to o l req u ire d fo r th e ir u n d e rs ta n d in g . B esid es th e ty p e s o f n o r th e r n o rig in , it co v ers a lso th o se w h o se c o n n e c tio n s a re r a th e r w ith t h e F ra n k ish R h in elan d .
58
THE CONTINENTAL BACKGROUND
Fig. 1. D e v e lo p m e n t o f t h e cru c ifo rm b r o o c h a Mildenhall, S u ffo lk , b Ix w o r th , S u ffolk, c East S h e ffo r d , Berks. d West S t o w H e a th , Suffolk, e Little W ilbraham , Cam bs. / Barrington, Cam bs. g B rooke, N o r f o lk , h Kcnninghall, N o r f o lk , i L a k e n h e a th , S u ffolk, j S leafo rd , Lines, k S leafo rd , Lines. G r o u p L atb ,c . V./.A.
G r o u p II. d ,e .
F o r so u rc e s o f fig u re s se c p . xiv.
G r o u p III. f g .
G r o u p IV. h,i.
G ro u p
CRUCIFORM AND CIRCULAR BROOCHES
5 9
northern peoples comprised in the Scandinavian Kulturkreis of this age. The objects themselves may often enable one to go further, to localize their origin o r their antecedents in one or other part o f this com plex, and to date them m ore o r less exactly. When, for example, a num ber o f simple cruciform brooches o f early-fifth-century type, similar to others that come from the Frisian Terpen, o r from cemeteries in Jutland or Schleswig, are found in Kent, they can be used quite properly by historians to set beside Bede’s statem ent that Kent was first settled by Jutes. So too they may be linked with the scraps o f tradition that picture Hengist, their leader, as an exile from his Ju tish homeland who had since had some n o t altogether creditable adventures among the Frisians. Similarly the general distribution o f cruciform brooches in the rest o f England can be treated as a relevant factor in assessing Bede’s judgem ent that the settlers who came to people East and Middle Anglia, Mercia, and Northum bria were Angles.1 Quite different, but equally instructive, arc the principal brooch types in vogue at this time among the Saxons and their close neighbours in the lands between the Elbe and the Weser. Here also two main types predom inate in the cemeteries and settlem ents o f the fifth century. These are the round brooches and the equal-armed brooches, each group appearing in two main forms. Of the round brooches one group comprises those which are cast in one piece, with the circular decorated surface surrounded by a more o r less deeply dished upstanding rim, so that they are normally know n in England as saucer brooches. In a less common variety o f this type, sometimes termed button brooches, the overall size is norm ally smaller, the rim reduced or om itted, and the whole decorated surface cast to represent a highlystylized human face (Fig. 2). The other main group o f circular form is that o f the socalled applied brooches. In these the decoration is carried on a separate thin sheet o f metal which is subsequently attached 1 T h e c ru c ifo rm b ro o c h e s o f E n g la n d w ere firs t a sse m b le d a n d c la ssifie d in N . A b erg , T h e A n g lo -S a x o n s in E n g la n d (U p p sala, 1 9 2 6 ). T h o u g h m a n y m ore h a v e b e e n f o u n d sin ce h e w r o te , a n d so m e m o d ific a tio n is n o w re q u ire d in his g ro u p in g a n d d a tin g o f th e m , h is c la ssific a tio n a n d c h ro n o lo g y a re still b a sic to a c o rre c t a p p re c ia tio n o f th e ir sig n ifican ce.
60
F ig . 2 .
T H E C O N T I N F. N T A L B A C K G R O U X D
C irc u la r b r o o c h e s o f v a rio u s ty p e s
a H a s tin g s , S u s se x , b K in g s to n - b y - L e w e s , S u s s e x , c H ig h d o w n , S u s se x . d A l f r i s t o n , S u s s e x , e S e l m e s t o n , S u s s e x . / A l f r i s t o n , S u s s e x , g F a irf o r d , G Io u c s . h F a ir f o r d , G lo u c s . F o r so u rc e s o f fig u re s see p . xiv.
E Q U A L - A R M ED B R O O CIIE S
61
by some adhesive to a flat disc forming the back-plate o f the brooch. Applied brooches are obviously more fragile than cast saucer brooches, as the sheet carrying the decoration is itself somewhat flimsy and can easily come adrift from the back-plate after some degree o f wear o r rough handling. Blank back-plates which have lost their decoration, or detached applied sheets, generally in fragments, are often all that may survive o f applied brooches. This fragility is no doubt the main reason why the type w ent out o f use comparatively early in this period, whereas the sturdy saucer brooches continued in production throughout these centuries, showing a long sequence o f decorative styles more o r less parallel with those displayed by the various types o f long brooches. Thus the geometrical and scroll m otifs in chip carving technique characteristic o f the fifth century, gradually give way to a range o f zoom orphic designs becoming ever more elaborate and com plex, as Style 1 animal ornament develops throughout the sixth. The equal-armed brooches form the most characteristic type of bow brooch in the Saxon regions. Unlike the vertical shape taken by the long brooches o f the north these have a squarer appearance. This is caused by the head and foot being both extended horizontally above and below the bow to form wide triangular panels, more or less equal in size. The type, which grew out o f a simple fourth-century form , the socalled Stützarm fibel, o f which English examples are known, for example, from Dorchester and Abingdon, O xon., Luton and Kem pston, Beds., and Mucking, Essex, was fully developed by the beginning o f the fifth century, b u t went out of use before 500. It can therefore be an im portant pointer to the date o f deposits in which it is found.1 Less than tw enty examples of the fully developed form of equal-armed brooch have so far been recorded in Britain, all of them probably im ported by the womenfolk o f early Saxon immigrants. While they are normally decorated with the scrolls and other motifs in chip-carving characteristic o f their date, it is o f some interest th a t they fall into two distinct T h e m o s t r e c e n t a c c o u n t o f th e s e b r o o c h ty p e s in E ngland is b y V . I. Evison in H .-J. H ässler (e d .), S tu d ie n z u r S a c h se n fo r sc h u n g ( 1 9 7 7 ) , 1 2 7 - 4 1 . T h eir d is tr ib u tio n in E n g la n d is a lm o s t e n ti r e l y n o r t h o f t h e T h a m e s , m o s t ly in a broad b a n d r u n n in g f r o m E ast a n d M iddle A nglia t o t h e A b in g d o n a re a: ib id ., fig. 6.
62
THE CONTINENTAL BACKGROUND
groups. There is a simpler form in which the decoration is confined within the wide triangular panels provided by the head* and foot-plates, and a m ore elaborate style in which the edges o f these plates also carry projecting animal figures and other decorative features giving them a much richer and more sum ptuous appearance. In the lands between the Elbe and the Weser, where the equal-armed brooches reached their developed form , there appears to be a local difference in the distribution o f the two varieties. The simpler type is mainly found west o f the river Oste which roughly divides this region in half; it is especially at home in the cemeteries east of the Weser estuary such as Wcsterwanna near Bremerhaven. The more elaborate ones come chiefly from sites east o f the Oste, such as Perlberg in the Elbe valley near Stade. It is significant that a similar pattern seems to be made by the distribution of the tw o forms of Saxon round brooch, the cast saucer brooches occurring mostly with the more elaborate equal armed brooches to the east in the Elbe valley, while the applied brooches belong rather with the simpler type in the Weser region.1 The English examples o f these brooches can thus be used not merely as pointers to the homeland o f these Saxon folk in general who took part in the English settlem ents; they can also provide clues to the local origin o f some o f them in one or other o f the tw o main centres o f Saxon population in the Elbe and Weser valleys at this time. The prevalence in England o f the fashion for cast saucer brooches is one of the reasons for thinking that a large part, o f our Saxon population came rather from the Elbe valley settlem ents, named after the cem etery at Perlberg near Stade, than from the Weser estuary where the type site is Westerwanna. These are some o f the ways in which a study o f broochtypes m ay be useful to historians. It is also w orth noting that the differences such personal articles show in form and use may indicate corresponding differences in the style of dress fashionable among the various tribes and cultural groups.2 Where, for example, it is usual to find on wom en’s bodies 1 T h e d i s tr ib u tio n s a re w ell sh o w n in T is c h le r, ‘D e r S ta n d d e r S a ch sen fo rs c h u n ß a rc h ä o lo g isc h g e s e h e n ’, A b b . 3 3 a n d d isc u sse d ib id ., 9 9 - 1 0 1 . 7 S ee several articles b y H . V ic rc k in C. A h re n s ( c d .) , S a c h sen u n d A n g elsa ch sen (H a m b u rg , 1 9 7 8 ), 2 3 1 - 7 0 .
BROOCHES AND POTTERY
6 3
in the cemeteries a pair o f similar brooches, one on each shoulder, sometimes linked by a string o f beads o r a chain, it is evident th a t the dress was different in style, and required a different m ethod o f fastening, from those cases in which only a single brooch is found centrally placed on the chest, and where the accompanying beads, if present, were worn as a necklace. So to o , the use o f wrist-clasps, a very com mon fashion among the Angles, suggests that sleeves o f a shape which required fastening in this way were more in vogue among them than among those tribes which rarely seem to have needed such fastenings. The other main category o f household goods that supplies useful clues to the origin o f different groups o f the invaders is of course their pottery. This is n o t the place to enter into an exhaustive discussion o f the various forms and styles of decoration popular at this time in north Germany and Scandinavia. N or is there any need to trace the styles prevalent at the tu rn o f the fourth and fifth centuries back into earlier times before the invasions began. But it may be useful to indicate some o f the m ore obvious differences in ceramic tradition which help to distinguish the products prevalent among the Angle, Saxon, and Jutish folk on the C ontinent, and so to make some o f these peoples recognizable among the settlers in this country. The known development of p o ttery styles on the C ontinent may also be helpful in suggesting the dates at which they became firmly established in Britain.1 Fashions in pottery followed a t this time roughly the same cultural divisions th a t have already been noted in considering the brooches. There is a similar broad distinction between the styles in vogue among the northern group o f peoples in eastern Schleswig and Ju tlan d , the Danish islands, and southern Scandinavia on the one hand, and those favoured by the Saxons and their neighbours between the Elbe and the Weser, with outliers eastwards in Holstein and westward in Frisia, on the other. The Nordseekustengruppe in western Schleswig and Ju tlan d , with southward extensions across the 1 See m y C orpus f o r a r o u g h c la ssifica tio n a n d full illu s tr a tio n o f t h e m ain f o rm s w h ic h th is p o t t e r y m a y t a k e , esp ecially t h e H istorical S u m m a r y (i. 1 1 4 - 2 7 ) , w h e re t h e significance o f the r e la tio n s h ip b e tw e e n t h e v a rio u s g ro u p s o f the E nglish a n d c o n ti n e n t a l m a te ria l is d iscussed.
64
THE CONTINENTAL BACKGROUND
estuary o f the Elbe, is less clearly distinctive as a ceramic region, being influenced by Anglian fashions on the east and by Saxon on the south; it has, however, some characteristic features o f its own. The whole northern K ulturkreis, including for the present purpose Bede’s continental Angles and Ju tes, favoured what may be termed a rectangular style in pottery decoration. Whatever the forms taken by their pottery, whether lall jars, shouldered o r globular vessels, o r shallow or carinated bowls, there is a tendency for the decoration to follow much the same p attern, with massed or grouped lines or grooves, horizontal on the neck and vertical on the shoulder or body. This linear o r grooved decoration is often continuous, giving a ribbed or corrugated appearance to the surface it covers. Ehe pottery is all hand-made, b u t is often o f high quality technically, sm ooth in fabric, regular in shape, with wellmoulded rims, and finished with a dark grey o r black surface bürnish, reminiscent o f polished metal. Curvilinear motifs in the decoration arc rarely used, and mostly confined to finger-tip rosettes, small swags, and roundels o f various kinds. Raised slashed collars, and the occasional appearance of vertical and horizontal strips overlaid on corrugated surfaces, lead on to a more extensive em ploym ent o f plastic ornam ent in the fifth century. This mostly takes the form o f shoulder bosses or long vertical bosses which fit easily into the rec tangular designs popular in these northern parts.1 Stamped decoration is almost unknown, but linc-and-dot patterns occur (Fig. 3). Variants o f this kind o f pottery, characteristic of the continental Angles, occur also in Jutland and through the territo ry o f the Nordseekustengruppe down the west side of the Danish peninsula. So far as Ju tlan d is concerned, particular interest attaches in the present context to the occurrence there o f a range o f bowls and jars, m ostly small and of rather squat appearance, with hollow necks and wellmoulded rims. They are decorated with simple grooved designs o f Anglian type, but including some chevron and 1 A . G e n r ic h , o p . c it . A b b . 2 , illu strates t h ir ty s t a n d a r d f o r m s w h ic h this p o t t e r y o f t h e c o n ti n e n t a l A ngles m a y t a k e . See also E. A lb re c tse n , F y n s k e je n ia ld e rg ra v e , iii (C o p e n h a g e n , 1 9 6 8 ), f o r m a n y o th e r s . A full range o f the E nglish m a te r ia l is in m y C o rp u s, ii. especially figs. 2 0 6 - 7 , 2 1 9 - 2 2 .
ANGLIAN POTTERY
F ig . 3 . a c e g
E n g lis h a n d c o n t i n e n t a l p o t t e r y o f A n g l i a n t y p e s
B o rd e s h o lm . B o rg s te d t. H a m m o o r. B o rg s te d t.
F o r so u rc e s o f figures see p . xiv.
b d / h
L o v e d e n H ill, L in e s . C a is to r , N o rfo lk . C a is to r , N o rfo lk . N e w a rk , N o tts .
65
66
THE C O N TIN EN TA L BACKGROUND
curvilinear m otifs accompanied by dots, dimples, and raised slashed collars. These can be m atched w ith similar pots from several sites in K ent, including tw o pieces from the early Grubenhäuser found w ithin the walls o f Roman C anterbury.1 They can moreover be linked culturally with the group of early-fifth-century cruciform brooches from Kent, which also have parallels in Ju tlan d . T hey greatly strengthen the case for believing that Bede's Ju te s came, directly or in directly, from th a t part o f the C ontinent (Fig. 4 and Map 5). As already noted, the pottery of the ATordseekustengruppe can also be associated with the Anglian and Ju tish K ultur kreis. Similar linear decoration predom inates here, although the corrugated technique is less com m on and diagonal or chevron designs are more prevalent, giving less rigidly rectangular effects. Curvilinear m otifs also are not unknown and, while stamps are not em ployed, line-and-dot designs are common, especially in com bination w ith chevrons to produce chcvron-and-dot patterns. These patterns were especially popular both in free G erm any and in the Late Antique repertoire o f Roman art in the fourth and fifth centuries. In Britain this chevron-and-dot decoration is characteristic o f the period o f overlap between Roman and early AngloSaxon times, occurring on pottery,12 metal, and bone objects on both sides of the cultural divide between them. The Saxon lands betw een the Elbe and the Weser show marked differences in ceramic fashions from those prevalent further north where Anglian styles predom inate. This is partly because o f the persistent influence from the earlier culture o f the Chauci and their neighbours long established there from pre-Roman times, and partly because of the intrusive forces from the proto-Saxon folk represented by the Fuhlsbüttel culture o f Holstein now extending south and west across the Elbe. Two pottery forms that arise from these mixed origins seem especially characteristic o f the developing Saxon fashions in the years before and after AD 400. One, derived in part from the Schalenurne culture o f East Holstein, comprises 1 T h e English scries fro m K e n t is illu stra te d in m y C o rp u s, ii. fig. 2 7 9 . 2 e.g. o n G r o u p F o f t h e so-called ‘R o m an o -S ax o n * p o t t e r y : D . B. H a rd en (c d .), D a rk-A g e B rita in ( 1 9 5 6 ) , 3 0 , fig. 5, n o s . 3 - 6 . A n g lo -S a x o n e x a m p le s a re illu s tr a te d in m y C o rp u s , ii. figs. 2 8 5 - 6 .
J UTI SH POTTERY
F ig . 4 .
67
J u tis h p o tte ry
a B e a k e sb o u rn e , K e n t, b W e s tb e re , K e n t, c W e s tb e re , K e n t. d O rp in g to n , K e n t. e W e s tb e re , K e n t. / H a n w e ll, M d d x . g S a rre , K e n t, h F a v e rs h a m , K e n t, i H o w le tts , K e n t. F o r so u rc e s o f Figures see p . xiv.
a wide range o f shouldered and carinated bowls, decorated with linear patterns including curvilinear swags. Some of these, the so-called Stanclfussschalen, may be m ounted on elegant footstands or pedestal feet, a teature continuing a tradition that had been com m on among the Chauci, but in its earlier form did not last far into the fifth century. Others are round-bottom ed, and this type had a longer period of developm ent, extending at least to 450. Both sorts are
68
THE CONTIN ENTAL BACKGROUND
M a p 5 . D i s t r i b u t i o n o f c o m p a r a b l e p o t t e r y in J u t l a n d a n d s o u th - e a s t E n g la n d
occasionally decorated with a faceted carinatiori, an early feature which develops later into a continuous row o f small bosses, thus providing one point o f origin for the shoulderboss bowls com m on on b o th sides o f the N orth Sea in the later fifth century. The other characteristic ceramic forms com m on among the Saxons comprise large urns and jars of rounded o r shouldered
MIXTURE OF POTTERY STYLES
69
contour, with narrow necks and well-moulded rims. They may be decorated either with linear or linc-and-dot chevrons, or w ith curvilinear m otifs, especially standing arches, stehende Bogen, interspersed with finger-tip rosettes and similar devices (Fig. 5). This style gives rise to looser and more varied effects than those produced by the massed horizontal and vertical lines and grooves which the Angle potters pre ferred. The contrast between them was accentuated when the fashion for plastic ornam ent became more pronounced as the fifth century wore on. Curvilinear features lend themselves more readily to the free development o f bossed ornam ent than does the Angle style, whose rigidity leaves little room for anything b u t straightforward shoulder-bosses. The Saxon fashion, on the other hand, made all sorts o f elaboration possible, with designs built up o f stehende Bogen bosses, split-oval and diagonal bosses, all enriched with finger-tip rosettes, chevron-and-dot patterns, and line-and-groove orna m ent, to culminate in the extravagant Buckelurnen o f the mid- and later fifth century (Fig. 6). Before long, moreover, stamped ornam ent became very popular among the Saxon potters, and gave them yet another means o f decorating plain surfaces and of emphasizing the structure o f designs built up by the use o f lines and bosses. The whole range o f these ceramic developments that took place in the continental lands from which the Angles, the Saxons, and the Ju te s sprang in the fourth and fifth centuries is reflected in the pottery o f the earlier English settlements. It constitutes a rich and varied source o f inform ation on all aspects of the m ovem ent, supporting and enlarging in many ways that supplied by the tools, weapons, jewellery, and other personal equipm ent of the invaders in materials such as metal and bone. Much o f the English pottery inevitably reflects, in the mixture o f its styles, the tribal confusion already prevalent in the lands from which the invaders came, as well as the dislocation of traditional techniques o f p o t making which the movement overseas involved. But it can on occasion identify with unexpected precision the exact continental localities with which some o f the English com munities were connected. Such links have been noted among the Angles, the Saxons, and the Ju te s alike. Among the crem ation urns from the Anglian cemetery at Sancton in
THE CONTINENTAL BACKGROUND
70
F ig . 5 . a c e
S a x o n S te h e n d e B o g e n p o tte ry
Z u id la rc n . W e s te m a n n a . W e s te rw a n n a .
F o r so u rc e s o f figures see p . xiv.
b d /
C a is to r, N o rfo lk . L ittle W ilb ra h a m , C a m b s . E l k i n g t o n , L in e s .
SAXON P O T T E R Y TYPES
F ig . 6 . a c e
S a x o n B u c k e lu m e n
W e s te rw a n n a . B lu m e n ta l. P e rlb e rg b e i S ta d e .
F o r so u rc e s o f figures see p . xiv.
b d /
S a n d y , B ed s. L u to n , B ed s. R u s h f o r d , N o rfo lk .
71
72
THE CONTINENTAL BACKGROUND
East Yorkshire, for exam ple, there are too many close parallels with the corresponding p o ttery from Bergstedt in Angeln to be explicable by pure coincidence.' Then again it is virtually certain that one o f the Anglian urns from Caistorby-Norwich, Norfolk, came from the same workshop as that which made tw o others, showing the same stylistic idiosyncracics, for cemeteries at Hammoor in Holstein and S^rup in Schleswig.2 So far as the Ju te s arc concerned, tw o tall narrow vases, whose form and elaborate decoration are closely similar, but unparalleled elsewhere, came respectively from Drengsted near the southern border o f Jutland and from Bifrons in K ent.3 Several such links exist between the English pottery and particular cemeteries or settlem ents in the Saxon lands between the Elbe and the Weser. O f these the m ost rem ark able are perhaps with Wehden near the Weser estuary. Here, in addition to a num ber o f other pieces markedly similar to English pots, was found an elaborately decorated Buckelurne w ith bosses modelled as human faces. The only known parallel to this—and a very close parallel indeed—came from the Markshall cemetery in Norfolk: they must be the work of one potter.4 A nother Markshall Buckelurne is one of only three known examples o f a highly specialized ceramic technique, in which the decorated sections o f an urn were apparently made separately and subsequently luted together and joined to the rest o f the p o t in a second firing. The other two known vessels in this unusual technique came from Feddersen Wierde and G udcndorf. All three must be products of one workshop, presum ably somewhere among the conti nental Saxons.5 It is w orth noting that while Markshall is in 1 J . N . L. M y re s a n d W. H . S o u t h e r n , T h e A n g lo -S a x o n C re m a tio n C e m e te ry a t S a n c to n ( 1 9 7 3 ) , 13. : M yres a n d B. G r e e n , T h e A n g lo -S a x o n C e m e te rie s o f C a isto r b y N o rw ic h a n d M a rk sh a ll, S o c . A n t. R e se a rc h C o m m . R e p o r t 3 0 ( 1 9 7 3 ) , 4 7 a n d fig. 3 . U rn B2 a n d the p iec es f r o m S