The Role of Art in the Late Anglo-Saxon Church 0198205414, 9780198205418

This study explores the role of early medieval religious art in its historical context, focusing on England from the rei

467 20 33MB

English Pages XLVI+312 [364] Year 1995

Report DMCA / Copyright


Polecaj historie

The Role of Art in the Late Anglo-Saxon Church
 0198205414, 9780198205418

  • Categories
  • Art

Table of contents :
List of Illustrations ix
Abbreviations xii
Introduction 1
1. The Illustration of Texts 6
Inherited Illustrations 9
The Physical Relationship of Text and Image 19
Frontispieces 20
Interspersed Imagery 30
Integrated Imagery 35
Modes of Illustration 41
Narrative Illustration 43
Complementary Illustration 45
Literal Illustration 48
The Effect of Illustration 53
2. Inscriptions 70
Written Content 76
The Appearance of Inscriptions 94
3. Visual Language: Setting the Scene 105
Introduction 105
Physical Form 107
4. Repetitions of Motif and Image and their Possible Implications 118
Repetition of Decorative Motif and the Co-ordinated Environment 118
Repetition of Imagery and Re-affirmation of Faith 126
5. Pictorial Narrative 135
6. Composition 150
Introduction 150
The Frame and Its Contents 152
Overlapping 160
Scale 166
Setting 176
Landscape 176
Architecture 180
Symbolic 185
Diagrammatic 187
7. Decoration 192
The Frame 192
Form 192
Function 208
Decorated Initials 213
Organization 216
Imagery 226
8. Broader Perspectives 235
The Reformed Monastic Houses: diversity in uniformity 235
North and South 237
The Monastic Church and the Secular Church 241
The Monasteries and the Secular World 248
Man and the Divine 260
Bibliography 266
Index of Art Objects 295
Index of Manuscripts 301
Index of People and Places 307

Citation preview








The Role o f A rt in the Late Anglo-Saxon Church




Oxford University Press, Walton Street, Oxford 0x2 6dp Oxford N ew York Athens Auckland Bangkok Bombay Calcutta Cape Town Dar es Salaam Delhi Florence Hong Kong Istanbul Karachi Kuala Lumpur Madras M adrid Melbourne Mexico City Nairobi Paris Singapore Taipei Tokyo Toronto and associated companies in Berlin Ibadan Oxford is a trade mark o f Oxford University Press Published in the United States by Oxford University Press Inc., New York

© Richard Gameson 1995 First published 1995 A ll rights reserved. N o part o f this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing o f Oxford University Press. Within the UK, exceptions are allowed in respect o f any fa ir dealing fo r the purpose o f research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, or in the case o f reprographic reproduction in accordance with the terms o f the licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside these terms and in other countries should be sent to the Rights Department, Oxford University Press, at the address above British Library Cataloguing in Publication Data Data available Library o f Congress Cataloging in Publication Data Gameson, Richard. The role o f art in the late Anglo-Saxon church / Richard Gameson. p. cm. — (Oxford historical monographs) Includes bibliographical references. 1. Christian art and symbolism —Medieval, 50 0 -1 5 0 0 — England. 2. Art, Anglo-Saxon — England. 3. Anglo-Saxons — England — History. I. Title. II. Series. N 7944.A1G 36 1995 704.9'482'094209021— dc20 94-36376 I S B N 0 -1 9 -8 2 0 5 4 1 -4

1 3 5 7 9 10 8 6 4 2 Typeset by Graphicraft Typesetters Ltd., Hong Kong Printed in Great Britain on acid-free paper by Biddles Ltd., Guildford & King's Lynn

To my parents

ACKNOWLEDGEMENTS It is a pleasure to begin by expressing my gratitude to those who have assisted in the preparation of this work. First and foremost I wish to thank Henry Mayr-Harting, whose friendship, support, and wise advice have been invaluable for the present book, just as they were for the doctoral thesis which lies behind it. Next, I would like to make grateful acknowledgement of the numerous libraries and institutions and their invariably helpful staff whose treasures I have studied and whose resources I have used. My debt to Corpus Christi College and Trinity College, Cambridge, the Conway Library of the Courtauld Institute of Art, the British Library, the Bodleian Library, the Bibliothèque Nationale, and the Bibliothèques Municipales of Boulogne and Rouen, is particularly great in this respect. No less deserving of special mention is the Church of England, which is still, happily, the owner and thus the curator of most of the extant material fabric of the late Anglo-Saxon church—not to mention of several splendid manuscript collections which I have used, notably at Durham, Exeter, Hereford, Salisbury, Worcester, and Winchester. I have had the good fortune to be associated with a number of fine academic institutions during the last decade. Trinity College, Oxford, my home for many years, provided the congenial milieu in which this work was first written. The Courtauld Institute of Art, University of London, afforded the ideal environment in which to revise it. The University of Kent at Canterbury was the supportive and inspiring setting for the final stages of production. For invaluable financial support which enabled me to undertake this and other work I am very grateful to Trinity College, Oxford (for the award of a Junior Research Fellowship); while I am doubly grateful to the British Academy, since the thesis was written with the support of one of its research studentships, and it was revised during the tenure of one of its Postdoctoral Fellowships. Numerous individuals have contributed in different ways to the development of my work over the years, and I would like especially to mention and thank here: Michelle Brown, Linda Brownrigg, Mildred Budny, Robin Cormack, Tilly de la Mare, George Henderson, Michael Kauffmann, John Lowden, Patrick McGurk, Rosamond McKitterick, Malcolm Parkes, and Bryan Ward-Perkins. I would also like to record my gratitude to Claire and Ian Nabney, for their repeated (and most congenial) hospitality in Cambridge, making my numerous visits there so pleasant. Ultimately and above all, I wish to thank my parents for their continuing interest, support and help; and Fiona, my dear wife, the least of whose manifold contributions to this work has been to listen patiently to each sentence— many times. R. G. Feast o f St Fridesmde 1993

CONTENTS List o f Illustrations Abbreviations

ix xii

Introduction 1. The Illustration of Texts Inherited Illustrations The Physical Relationship of Text and Image Frontispieces Interspersed Imagery Integrated Imagery Modes of Illustration Narrative Illustration Complementary Illustration Literal Illustration The Effect of Illustration

1 6 9 19 20 30 35 41 43 45 48 53

2. Inscriptions Written Content The Appearance of Inscriptions

70 76 94

3. Visual Language: Setting the Scene Introduction Physical Form

105 105 107

4. Repetitions of. M otif and Image and their Possible Implications Repetition of Decorative M otif and the Co-ordinated Environment Repetition of Imagery and Re-affirmation of Faith

118 118 126

5. Pictorial Narrative


6. Composition Introduction The Frame and Its Contents Overlapping Scale Setting Landscape Architecture

150 150 152 160 166 176 176 180



Symbolic Diagrammatic 7. Decoration The Frame Form Function Decorated Initials Organization Imagery 8. Broader Perspectives The Reformed Monastic Houses: diversity in uniformity North and South The Monastic Church and the Secular Church The Monasteries and the Secular World Man and the Divine Bibliography

185 187 192 192 192 208 213 216 226 235 235 237 241 248 260 266

Index o f Art Objects Index o f Manuscripts Index o f People and Places

295 301 307

LIST OF ILLUSTRATIONS [between pages 146 and 147]

NB: Where no source is given, the photographs are the author’s. 1a. St Mary’s Church, Breamore, Hants. I b. St Laurence’s Church, Bradford-upon-Avon. 2a. New Minster Charter, pictorial frontispiece. British Library, MS Cotton Vespasian A. viii, fo. 2V(photo: Courtauld Institute). 2b. New Minster Charter, descriptive verses. British Library, MS Cotton Vespasian A. viii, fo. 3r (photo: Courtauld Institute). 3. The Harley 603 Psalter; illustration and part of text for Psalm 13. British Library, MS Harley 603, fo. 7' (photo: Courtauld Institute). 4. Judith of Flanders’ Gospels, jewelled binding (front cover). New York, Pierpont Morgan Library, M 708, front cover (photo: Morgan Library). 5a. Ivory plaque with Christ enthroned. London, Victoria and Albert Museum, no. A32-1928 (photo: V. & A.). 5b. Ivory plaque depicting the Crucifixion. London, Victoria and Albert Museum, no. A80-1923 (photo: V. & A.). 6. The New Minster Liber Uitae, frontispiece. British Library, MS Stowe 944, fo. 6r (photo: Courtauld Institute). 7a. New Minster Liber Uitae. British Library, MS Stowe 944, fo. 6V(photo: Courtauld Institute). lb. New Minster Liber Uitae. British Library, MS Stowe 944, fo. 7r (photo: Courtauld Institute). 8. The Boulogne 11 Gospels, beginning of John’s Gospel. Boulogne, Bibliothèque Municipale, MS 11, fo. 107' (photo: C. R. Dodwell). 9. The Old English Hexateuch, the fall of the rebel angels. British Library, MS Cotton Claudius B. iv, fo. 2r (photo: Courtauld Institute). 10. The Benedictional of St Æthelwold, the Annunciation. British Library, Add. MS 49598, fo. 5V(photo: Warburg Institute). II a. The Benedictional of St Æthelwold, Dormition of the Virgin Mary. British Library, Add. MS 49598, fo. 102v (photo: Warburg Institute). 1\b. The Rouen Benedictional, Dormition of the Virgin Mary. Rouen, Bibliothèque Municipale, MS Y. 7, fo. 54' (photo: C. R. Dodwell). 12. The Harley 603 Psalter, illustration to Psalm 136. British Library, MS Harley 603, fo. 70r (photo: Courtauld Institute). 13a. The Benedictional of St Æthelwold, monastic bishop pronouncing a blessing. British Library, Add. MS 49598, fo. 118' (photo: Warburg Institute).

X 13b. 14.



16. 17a. Mb. 18a. \%b. 19. 20a. 20b.


2\b. 22.

23a. 23b.

24a. 24b. 25a.

List o f Illustrations

The Tiberius Psalter, the Harrowing of Hell. British Library, MS Cotton Tiberius C. vi, fo. 14r (photo: Courtauld Institute). The Old English Hexateuch, feast for the weaning of Isaac; Ishmael and Isaac playing, God addressing Abraham. British Library, Cotton Claudius B. iv, fo. 35' (photo: Courtauld Institute). The Bury St Edmunds Psalter, marginal illustration of the Adoration of the Magi at Psalm 71. Biblioteca Apostolica Vaticana, MS Reg. lat. 12, fo. 78v (detail) (photo: Courtauld Institute). The Bury St Edmunds Psalter, marginal illustration of the Crucifixion at Psalm 21. Biblioteca Apostolica Vaticana, MS Reg. lat. 12, fo. 35r (photo: Courtauld Institute). The Harley 603 Psalter. Illustration for and part of the text of Psalm 138. British Library, MS Harley 603, fo. 71r (photo: Courtauld Institute). The New Minster (Ælfwine) Prayer Book, the Crucifixion. British Library, MS Cotton Titus D. xxvii, fo. 65' (photo: Courtauld Institute). The Tiberius Psalter, St Michael and the dragon. British Library, MS Cotton Tiberius C. vi, fo. 16r (photo: Courtauld Institute). The Grimbald Gospels, portrait page of St John. British Library Add. MS 34890, fo. 114v (photo: Courtauld Institute). The Grimbald Gospels, initial page of John’s Gospel. British Library, Add. MS 34890, fo. 115r (photo: Courtauld Institute). The Arenberg Gospels, portrait page of St Mark. New York, Pierpont Morgan Library, M 869, fo. 57' (photo: Morgan Library). Potterne, Wilts., font with inscribed rim. Ivory plaque depicting the Last Judgement, with St Peter and the Virgin Mary. Cambridge, Museum of Archaeology and Anthropology, acc. no. 215154 (photo: Museum of Arch, and Anth.). The Caligula Troper, St Lawrence before Decius; the martyrdom of St Lawrence. British Library, MS Cotton Caligula, A. xiv, fo. 25r (photo: Courtauld Institute). The Caligula Troper, the Ascension. British Library, MS Cotton Caligula A. xiv, fo. 18r (photo: Courtauld Institute). The Cotton Cleopatra Psychomachia, illustration of Wisdom enthroned, with text of lines 868-81. British Library, MS Cotton Cleopatra C. viii, fo. 33r (photo: Courtauld Institute). Benedictional of St Æthelwold, St Etheldreda. British Library, Add. MS 49598, fo. 90' (photo: Warburg Institute). Benedictional of St Æthelwold, incipit to the blessing for the feast of St Etheldreda with historiated initial ‘O’. British Library, Add. MS 49598, fo. 91r (photo: Warburg Institute). Headbourne Worthy, Hants, sculptured Crucifixion (mutilated). Romsey, Hants, sculptured Crucifixion. St Lawrence’s Church, Bradford-upon-Avon, sculptured angels.

List o f Illustrations

25b. 26a. 26b. 26c. 21a. 21b. 28a. 28£. 29. 30. 31. 32.


Nether Wallop, Hants, painted angel, plus tip of mandorla. Bibury, Glos., impost of chancel arch (north side). Bibury, Glos., impost of chancel arch. Milborne Port, Som., south doorway: south face of inner capital on the west side. Canterbury Cathedral, crypt, Holy Innocents’ Chapel, capital 1, east side. Langford, Oxon., rood currently over south door (incorrectly reassembled). Alcester Tau Cross: Crucifixion. British Museum, MLA 1903, 3-23, 1 (photo: British Museum). Alcester Tau Cross: Christ trampling on the beasts (photo: British Museum). The Old English Hexateuch, the sacrifice of Isaac. British Library, MS Cotton Claudius B. iv, fo. 38r (photo: Courtauld Institute). The Benedictional of St Æthelwold, St Benedict enthroned. British Library, Add. MS 49598, fo. 99v (photo: Warburg Institute). The Tiberius Psalter, incipit to Psalm 51. British Library, MS Cotton Tiberius C. vi, fo. 12' (photo: Courtauld Institute). Reliquary Crucifix (gold sheet over cedar, with filigree decoration, enamels and walrus-ivory corpus). London, Victoria and Albert Museum, Department of Metalwork, inv. 7943, 1862 (photo: V. & A.).


AB ANS Arun. A SE BAACT BAR BAV BM BN BL Bodl. Brit. Mus. Cal. CCCC CCSL Claud. Cleo. col. pi. col. repro. Cott. CUL Dom. EEMF EETS Eg. EHR ep. HBS JB A A JW C I M GH Jul. Jun. Lans. Ost. Nat. PL PML Rawl.

Art Bulletin Proceedings o f the Battle Conference for Anglo-Norman Studies Arundel Anglo-Saxon England British Archaeological Association Conference Transactions British Archaeological Reports Biblioteca Apostolica Vaticana, Vatican City Bibliothèque Municipale Bibliothèque Nationale British Library, London Bodleian Library, Oxford British Museum, London Caligula Corpus Christi College, Cambridge Corpus Christianorum Series Latina Claudius Cleopatra colour plate colour reproduction Cotton Cambridge University Library Domitian Early English Manuscripts in Facsimile Early English Texts Society Egerton English Historical Review epistula Henry Bradshaw Society Journal o f the British Archaeological Society Journal o f the Warburg and Courtauld Institutes Monumenta Germaniae Historica Julius Junius Lansdowne Österreichische Nationalbibliothek, Vienna Cursus Patrologiae Series Latina (accur. J.-P. Migne) Pierpont Morgan Library, New York Rawlinson

Abbreviations RS SCH Tan. TCBS TRH S TCC Tib. Tit. V. & A. Vesp. Vit.

Rolls Series Studies in Church History Tanner Transactions o f the Cambridge Bibliographical Society Transactions o f the Royal Historical Society Trinity College, Cambridge Tiberius Titus Victoria and Albert Museum, London Vespasian Vitellius

Introduction This work explores the role of the visual arts in the English church (and especially in its monastic sector) from the reign of Alfred the Great (871-99) to the genera­ tion after the Norman Conquest. My principal concern is to elucidate the uses to which art was put, the ways in which it functioned, and the resources it had for doing so. Needless to say, an appreciation of its functions requires a clear under­ standing of its character and forms. Accordingly, I shall be considering the nature of the surviving artefacts and their visual language in order to assess what they could convey and how they did so. The decision to focus this study on a period of approximately two hundred years in the history of one country reflects two complementary considerations which the reader may be interested to know at the outset. It was dictated by the wish to have a field which was sufficiently large to allow us to perceive contours and changes within it, yet which was at the same time sufficiently confined for us to be able to make valid generalizations about it as a whole. Many of the issues discussed are equally relevant to other periods and countries, as are various of the findings. However, it seems wholly preferable to offer observations that are un­ doubtedly pertinent for their particular context (and which may well also have broader implications), rather than to offer generalizations which are only moder­ ately appropriate for much of the material to which they purport to relate. Nevertheless, the role of art is a broad, varied, and complex topic which does not admit any single set of answers or conclusions, and a work of this scale could not hope to do justice to it in a comprehensive way. Moreover, there is no reliable, established methodology for such a study, nor is there ever likely to be one. In these circumstances, my approach has been to address in detail a series of fairly specific issues. What is offered here, therefore, is a set of interrelated, quite sharply focused studies, which collectively define some of the parameters of the field as a whole, while casting a clear light on particular areas within it. The questions which we may profitably ask of the art of the late Anglo-Saxon church in our bid to elucidate the roles it fulfilled are to a large extent defined by the limitations of the available evidence. It is impossible to explore in any depth the nature and functions of Anglo-Saxon mural painting, for example, when the surviving physical evidence is limited in the extreme, and there are virtually no descriptive documentary sources.1 Decoration and ornament, on the other hand, survive in a fair quantity in a variety of media and can be examined in some detail. Note: works are cited by short title throughout the notes, the full details for each item being supplied in the bibliography. 1 The limitations o f the written sources are well expounded by Dodwell, Anglo-Saxon A rt, esp. 9 2 4 for mural decoration.



Moreover, despite (or perhaps because of) their ubiquitousness, they have re­ ceived considerably less attention than figurai imagery from modern commenta­ tors. Yet they were undoubtedly very important for contemporaries. On account of all these factors, they are given sustained treatment in this work. At the same time, it would be disingenuous not to state that the selection of themes and issues for discussion is a personal, albeit not unrepresentative, one. Manuscripts are un­ doubtedly the best preserved works of art of the period in question; they are also especially dear to the heart of the present writer: accordingly they receive the most attention here. Other questions could profitably be asked and answered, and it is hoped that the appearance of this work will encourage others to continue doing so. The loss of countless works of art of the highest quality and prominence is a serious impediment to the historian of the visual culture of late Anglo-Saxon England. O f the most cosdy, important, and public works, such as major programmes of sculptural or painted decoration, large-scale three-dimensional effigies, textiles, altars, precious-metal crosses and vessels, very little survives. Furthermore, for a study like this, the lacunae are even greater. No less important for assessing the role of art than the works themselves is their setting, and this too has largely vanished. Very few late Anglo-Saxon churches still convey a fair impression of their internal appearance as it was in the tenth and eleventh centuries, and those that do are invariably of secondary or tertiary, rather than primary, status. Among the most prestigious structures from the period that are substantially intact are St Mary’s Church and Odda’s Chapel, Deerhurst (Gloucestershire),2 St Mary’s Church, Breamore (Hampshire) (PI. la),34and St Laurence’s, Bradford-on-Avon (Wilt­ shire) (PI. lb),* and none of these is of more than second-rank importance. We cannot, by contrast, stand in Dunstan’s Canterbury, or Æthelwold’s Winchester.5 This is not to ignore the achievements of modern architectural historians and archaeologists; on the contrary, their painstaking work throughout the century and especially in the latter decades of it has revealed in great detail the design, the history, and the formal affiliations of many of the most important buildings. Plans and reconstructions convey a clear impression of these structures.6 Yet such work, admirable though it is, cannot fully convey what these buildings were like as the 2 Taylor and Taylor, Anglo-Saxon Architecture, i. 193-211, with ii, ills. 443-8; Taylor, Deerhurst Studies; Fernie, Architecture o f the Anglo-Saxons, 101-6; and Klukas, ‘Liturgy and Architecture’; also [C.] P. Wormald, ‘Anglo-Saxon Deerhurst’. 3 Green and Green, Saxon Architecture and Sculpture in Hampshire, 5-10; Taylor and Taylor, AngloSaxon Architecture i. 9 4 - 6 , with ii, ills. 4 0 5 -6 ; Fernie, Architecture o f the Anglo-Saxons, 112-14; and Rodwell and Rouse, ‘Anglo-Saxon Rood’. 4 Taylor and Taylor, Anglo-Saxon Architecture i. 8 6 -9 , with ii, ills. 401-2. Fernie, Architecture o f the Anglo-Saxons, 145-51. 5 For Canterbury see Gem, ‘Anglo-Saxon Cathedral Church at Canterbury’, and Woodman, Archi­ tectural History o f Canterbury Cathedral, 15-22; for Winchester see Biddle, 'Felix urbs Winthonia', esp. 136-9 with fig. 6. 6 The largest collection o f plans is in Taylor and Taylor, Anglo-Saxon Architecture. For general surveys of the period sec Taylor, ‘Tenth-Century Church Building’; and Fernie, Architecture o f the Anglo-Saxons, esp. 90-111. The problematic question o f the influence o f liturgical practices on the architecture o f the period is explored by Klukas, ‘Liturgy and Architecture’; and Spurrell, ‘Architec­ tural Interest o f the Regularis Concordia\



immediate setting for the art which is our concern here. For this setting consisted as much of liturgy, ceremony, music, and belief as of architecture. The perception of the most important works of art in the late Anglo-Saxon church was a synaesthetic experience which involved sound, light, and movement— not to mention faith and expectations. Art was part of worship and belief. The Good Friday observances centring around the cross that are detailed in the Regularis Concordia demonstrate the point quite clearly.7 The planned movements, the climactic unveiling of the cross, and the numerous prayers and the antiphons that accompanied and com­ mented on the ritual were as important as the cross itself in determining the aesthetic and devotional effect of this monastic rite. Correspondingly, the impact on the faithful beholder of, let us say, a large image or effigy of a saint or a divine figure when it was carried on an appropriate feast-day in solemn procession by serried ranks of clergy, perhaps in flickering candle-light, accompanied by chant­ ing and censing is difficult to recover. In the late Anglo-Saxon church, as else­ where, art, ceremony, and belief were intimately allied, and, indeed, interdependent. Alongside the grievous losses of works of art themselves, undoubtedly the great­ est handicap for one trying to study the role of the visual arts is the dearth of contemporary written comment on that subject itself. Despite a comparative wealth of early documentary references to tenth- and eleventh-century artefacts and decoration,8 there are relatively few extended descriptions of any of them, there is little comment about their makers, their patrons, and the nature of patronage, and there is virtual silence on their functions and the ways in which they were per­ ceived. Very occasionally written sources do offer direct comments on the sym­ bolic significance of a building or object. For instance, the biography of St Oswald informs us that because the saint ‘built under the venerable sign of the cross through which we believe ourselves to be saved, so he began to construct the buildings [of the church at Ramsey] in the form of a cross’;9 while Amalarius of Metz states in his De ecclesiasticis officiis that a censer ‘signifies the body of Christ, in which is fire namely the Holy Spirit’.10 However, concerning the role of art as 7 Regularis Concordia, ed. Symons, sects. 4 4 -5 (pp. 42-4). For contemporary illustrations o f other ceremonies see Temple, Anglo-Saxon Manuscripts, ills. 91 and 256; and Colchester (ed.), Wells Cathe­ dral, pi. 5. 8 The Latin texts are conveniently assembled in Lateinische Schriftquellen, ed. Lehmann-Brockhaus; to which Dodwell, Anglo-Saxon Art, provides an invaluable commentary. 9 ‘Quoniam igitur reuerendae crucis signo munierat, per quod nos credimus saluari, sicut et ipse ideo illius loci aedificia coepit construere in modum crucis'; Uita Oswaldi in Historians o f the Church o f York, ed. Raine, i. 434. For the design in question see Fernie, Architecture o f the Anglo-Saxons, 112-36. Further on the symbolic dimension o f Anglo-Saxon architecture see Gem, ‘Iconography o f AngloSaxon Architecture'. 10 ‘Thuribulum corpus Christi significat in quo est ignis scilicet Spiritus Sanctus', continuing, ‘ex quo bonus odor procedit, quem unusquisque electorum ad se uult rapere. Idem odor bonam operationem de Christo exire demonstrat, quam qui uiuere uult in suum cor trajicit' (PL 105, col. 1125a). The currency o f Amalarius' work in late Anglo-Saxon England is underlined by the surviving manuscripts (Boulogne, BM MS 82 o f the early 10th c. and the handsome T CC M S B. 11. 2 o f mid-10th-c. date; also BL M S Cott. Otho D. xv, fos. 102-21 o f r.1000 (excerpts)); and by Ælfric's use o f the text in his ‘Letter to the Monks o f Eynsham', ed. Bateson (for which see Gatch, ‘Old English Literature and the Liturgy', 240-1).



such, contemporary writers had nothing to say. The reason for this is fairly straightforward: unlike eighth- and ninth-century Byzantium,11 no one questioned the validity of art in the context of the late Anglo-Saxon church, and consequently no one saw any need to justify or explain it practically, theologically, or philosoph­ ically. The single passage which at first sight appears to be an important exception to this rule, transpires on closer investigation to be a chimera—albeit a very interesting one.12 Given these restrictions, exploring the role of art is not unlike trying to recon­ struct the effect of an opera in a particular theatre without direct access to the building, the scenery, the cast, and the orchestra—and sometimes without the score as well. We have a few of the props (some of which are broken) and a limited amount of information concerning the singers and the audience. The task is cer­ tainly daunting. Yet it is by no means a hopeless or an unrewarding one. In these circumstances, therefore, it has seemed best to proceed by addressing a series of specific questions to which answers may be obtained. The first subject treated (Chapter 1) is the illustration of texts: we consider the forms, the arrange­ ment, and the functions of illustrations, and the ways in which they may have been perceived. Next, in Chapter 2, we examine the role of the written inscriptions that appear on numerous works of art, assessing both their semantic contribution and that of their physical appearance. Then, as a preliminary to exploring various aspects of Anglo-Saxon visual language, Chapter 3 discusses the extent to which the physical form of an object could determine the choice and nature of motifs that were used to adorn it. In Chapter 4 we explore the possible implications of the repetition of decorative motifs and figurai imagery from one context to another and from medium to medium within a single environment. Chapter 5 examines the various methods used by Anglo-Saxon artists to convey narrative in pictorial form; while in Chapter 6 we study the way in which pictorial composition and depicted settings contributed to the meaning of visual imagery. In the following chapter we turn our attention to decorative motifs, looking in particular at the role of frames and initials. Finally in Chapter 8 we re-examine the corpus of material as a whole in a series of broader perspectives, exploring the importance of geographic loca­ tion, ecclesiastical status, and secular patronage for defining the nature and func­ tion of ecclesiastical art. We conclude by emphasizing the crucial role of art in the late Anglo-Saxon church in bridging the gulf between man and the divine. There is a lingering prejudice in certain circles that, whilst political, economic, ecclesiastical, and intellectual history is History, art history is not. It is hoped that the appearance of the present study— written by a history lecturer—in the Oxford Historical Monographs series will help to dispel the vestiges of this anachronistic " For an introduction to Iconoclasm and the voluminous modem literature on it sec Martin, History o f the Iconoclastic Controversy; Grabar, L'lconoclasme byzantin., P. Brown, ‘Dark Age Crisis’; Brycr and Herrin (eds.), Iconoclasm, and Cormack, Writing in Gold, 95-140. 12 The passage in question is Ælfric, Sermones Catholici, ed. Thorpe, i. 180-92 at 186; it is discussed by Gameson, ‘Ælfric and the Perception o f Script and Picture’.



view. Visual evidence is an important source for any period, but for the earlier Middle Ages, where the amount of other material is strictly limited, it is particu­ larly crucial. Moreover, tenth- and eleventh-century society expressed itself at least as much through visual means—ritual, ceremony, architecture, and art—as by documentary ones. Bishop Æthelwold of Winchester articulated his beliefs, ideals, and affiliations as forcefully in manuscript art as he did in writing; King Cnut expressed his aspirations no less through artistic patronage than by law codes and pronouncements. The indifference until recently of historical scholarship to this prominent medium of expression would undoubtedly have baffled and vexed both.

1 The Illustration of Texts The New Minster Charter was written in 966 to commemorate and perpetuate the replacement of secular clergy with Benedictine monks at the New Minster, Win­ chester, which had been achieved through the agency of King Edgar (959-75).1 This was a controversial affair,2 and the document is correspondingly unusual (PI. 2). The lengthy text in hermeneutic Latin, for which Bishop Æthelwold of Win­ chester himself (963-84) may well have been responsible, describes the spiritual implications and the practical details of the new settlement, and attempts to sanc­ tify and hence legitimize the arrangement in the most powerful way possible—by casting the affair into a cosmic perspective of the fall and redemption of mankind. The designer of the document realized precisely the same aim by preparing it according to the format of a supremely de luxe liturgical book. This is the ultimate development of the well-attested practice of adding sacred authority to a record by copying it into a gospel book.3 Most unusually, the New Minster Charter takes the form of a small codex. It is an exceptionally beautiful one. The parchment is uniformly clean and creamy in tone with very little contrast between the hair and flesh sides of the membrane. The text, which is the work of a single scribe, is written throughout in gold ink in an elegant early English Caroline minuscule with chapter headings in uncials. Needless to say it is spaciously laid out, and the text is surrounded by generous margins of blank parchment which, when the codex was in its original binding, must have been greater still. In point of fact, no other late Anglo-Saxon manu­ script, not even the finest, is entirely written in gold, and one must look to the products of certain late Antique, Byzantine, Carolingian, and Ottoman scriptoria for books of comparable luxury.4 It is quite understandable, therefore, that this 1 BL MS Cott. Vesp. A. viii. See F. Wormald, ‘Late Anglo-Saxon Art’ (Collected Writings, i. 10810). Its text is printed in Councils and Synods, i, ed. Whitelock et al., 119-33. Further on the text and its authorship see John, Orbis Britanniae, 271-5, and Lapidge, ‘Æthelwold as Scholar and Teacher’, 96. For a reproduction o f the script sec Watson, Dated and Datable British Library, ii, ill. 20. 2 See C. P. Wormald, ‘Æthelwold and his Continental Counterparts’, esp. 3 7 -4 2 . 3 e.g. the note o f Cnut’s confraternity with Christ Church, Canterbury, and his confirmation o f the privileges o f that house which were added to BL M S Royal 1 D. ix, fos. 43' and 44' (reproduced in Hicks (ed.), England in the Eleventh Century, ill. 11; Ranger (ed.), Prisca Munimenta, pi. 2). See further Ker, Catalogue, no. 247; Warner and Gilson, Catalogue, i. 17-18. 4 See E. M. Thompson, Greek and Latin Palaeography, 3 2 -4 . For an Ottonian manuscript which is written entirely in gold on purple parchment, PM L M 23, see Lowe, ‘The Morgan Golden Gospels’; with col. pi., Berkowitz, In Remembrance o f Creation, frontispiece. However, by this date the use o f single purple leaves in key positions was the norm.

The Illustration o f Texts


feature should have attracted the attention of a later Winchester chronicler.5 The proem, which is written against a striking bright blue ground, and the start of the charter itself are set within matching gold and colour frames; and it is highly likely that the volume was originally encased in a precious binding. The first page of the original document contains not text but a picture. Before the beholder can read the proem on fo. 3V (‘King Edgar granted this privilege to the New Minster and conceded it to Almighty God and his mother Mary, whilst praising his great works’),6 this idea is forcibly presented to him by the famous frontispiece image which occurs on fo. 2V(PI. 2a).1 Here, against a purple ground and within a frame of gold bars and foliage— the height of opulence and of contemporary fashion— King Edgar is depicted between the Virgin Mary and St Peter, genuflecting in adoration while holding up a codex (the charter) to the cross-nimbed Deity above; the latter is seated on an arc within a mandorla, attended by four angels. The composition as a whole is reminiscent of the eastern iconography for the Ascension which was certainly known in late Anglo-Saxon England.8 Be that as it may, the highly formal, symmetrical arrangement of the figures undoubtedly helps to ex­ press and underline the divine order behind the event it portrays. The image is complemented by a couplet in golden uncials on the facing recto (PI. 2b). Level with Christ is written: ‘Thus he who established the stars sits on a lofty throne’, and it continues below: ‘King Edgar, prostrate, venerates and adores him.’9 It is symptomatic of the importance of visual criteria in late Anglo-Saxon eccle­ siastical culture that the physical appearance of this document was felt to be at least as important as its textual content—it is not particularly easy actually to read the latter, written as it is in all-gold script—and that the one was carefully designed to support and augment the other. The general presentation of the charter and, most explicitly, the pictorial frontispiece, underlines its sacrality and demonstrates the authority on which it rests. Christ prefaces and presides over the text, precisely as he does over that of some gospel books; while the King, an important agent of his will on earth, is shown in a suitably subordinate yet nevertheless privileged and direct relationship with him. God and the company of heaven sanction, and are directly involved in, the remarkable, imposed monastic reform of Bishop Æthelwold and King Edgar. In this chapter we will examine the sources, the arrangement, the nature, and some of the functions of the pictorial matter that occurs in extant late Anglo-Saxon s Annales Wintoniensis, in Ungedruckte Anglo-Norman Geschichtequellen, ed. Liebermann, 69; also Lateinische Schriftquellen, ed. Lehmann-Brockhaus, no. 6405. * ‘Eadgar Rex hoc priuilegium nouo edidit monasterio ac omnipotenti domino eiusque genitrici mariae eius laudens magnalia concessit’ (repro.: F. Wormald, Collected Writings i, ill. 96). 7 Col. ills.: Campbell (ed.), The Anglo-Saxons, ill. 164; Wilson, Anglo-Saxon A rt, ill. 261. 8 As in the late 6th-c. Rabbula Gospels (Florence, Biblioteca Medicea Laurenziana, M S Plut. 1. 56, fo. 13v: see Cecchelli et al., Rabbula Gospels). The design is repeated in 10th-c. England in the ‘Athelstan’ Psalter (BL M S Cott. Galba A. XVIII, fo. 120': Temple, Anglo-Saxon Manuscripts, ill. 31). 9 ‘Sic celso residet solio qui condidit astra | Rex uenerans Eadgar pronus adorat eum .’ For brief comments on the language see Lapidge, ‘Æthelwold as Scholar and Teacher’, 96 with n. 53.


The Illustration o f Texts

books. Although the majority of the surviving manuscripts do not have illustrative matter, a considerable number do.10 A large proportion of these are, predictably, liturgical volumes and other ‘altar’ books (among which we may class the New Minster Charter), with a preponderance of gospel books and gospel lectionaries, closely followed by psalters.11 However, pictorial matter is also found in a few nonliturgical texts. Aldhelm’s De laude uirginitatis (both prose and poetic versions) and Boethius’ De consolatione philosophiae are cases in point.12 If the provision of illus­ tration here is to some extent a reflection of the revered status of the texts in question within the canon of Anglo-Saxon monastic reading and hence also of the frequency with which they were copied, one cannot, it must be stressed, argue the reverse from the absence of illustration. Outstanding among non-liturgical books, judged both by the number of illus­ trated copies and by the number of pictures within each, was Prudentius’ Psychomachia, of which no fewer than three fully or near-fully illustrated versions and a fragment of a fourth remain.13The universal applicability of the theme of the poem (the battle for the soul), the highly dramatic presentation of the subject matter in terms of a series of physical combats between virtues and vices, and the existence of a firm tradition of illustration for the text all help to account for this phenomenon. Interestingly, these four illustrated copies are not all books of the same grade, which may imply that they were created to fulfil different roles. British Library, MS Cotton Cleopatra C. viii, which is the smallest of the four and whose condition attests to extensive handling, would seem to have been designed and used as a communal ‘library’ book (PI. 22).14 Cambridge, Corpus Christi College, MS 23 (part I), by contrast, the largest of the four, was clearly conceived as a de luxe volume. Its special status is underlined by the fact that it is one of the comparatively few Anglo-Saxon manuscripts which has a pre-Conquest anathema and dedicatory inscription.15 The striking contrast between these two illustrated copies of the same text usefully reminds us that the hand-produced book was a very flexible commodity. The reasons why certain copies of non-liturgical texts received illustration but others did not are now generally unclear. It sometimes, no doubt, reflects impon­ derable factors such as the availability of models and artists at a given time, not to mention the changing nature of the scriptorium, its aims, its work-load, and its economics. That the involvement of a particular patron or the status of a projected 10 For extant manuscripts as a whole see Gneuss, ‘Preliminary List’ (in need o f revision, especially for the period c. 1050-1100). For illustrated manuscripts sec Temple, Anglo-Saxon Manuscripts; Ohlgren, Insular and Anglo-Saxon Manuscripts; and Kauffmann, Romanesque Manuscripts, cats. 1-15. 11 Anglo-Saxon liturgical books are conveniently listed by category in Gneuss, ‘Liturgical Books in Anglo-Saxon England’. 12 See Temple, Anglo-Saxon Manuscripts, cats. 20, 39, and 57. 13 CCCC M S 23; BL Add. M S 24199, and Cott. Cleo. C. VIII; Munich, Staatsbibliothek, Clm. 29031 b. (Temple, Anglo-Saxon Manuscripts, cats. 48-51; see further Wieland, ‘Anglo-Saxon Manu­ scripts o f Prudentius’ Psychomachia'). None seems to be directly dependent on any o f the others. M For a facsimile pi. o f fo. 10' see Schools o f Illumination i, pi. 12a. 15 The inscription is written in ungainly monumental capitals on fo. iiv. For the text see James, Catalogue o f the Manuscripts o f Corpus Christi College, Cambridge, i. 44.

The Illustration o f Texts


recipient could influence the grade, the appearance, and the decoration of the finished product is not in doubt—one need only recall that one of the finest and most extensively illustrated service books of the age was specifically commissioned by the flamboyant Bishop Æthelwold, the probable inspiration behind the New Minster Charter, for his personal use (Pis. 10, 11a, 13a, 23, 30). On the other hand, wishes could not transcend resources, and one may legitimately wonder, for instance, whether the magnificent benedictional to which we have just alluded16 could have been made in the 960s or 970s anywhere other than at Winchester. The great majority of the illustrated books that survive from late Anglo-Saxon England, including all those mentioned so far, have Latin texts. Two of the most interesting, however, are in Old English. These are Bodleian Library, MS Junius 11, the so-called ‘Caedmon’ manuscript of Old English biblical poetry;17 and Brit­ ish Library, MS Cotton Claudius B. iv, the illustrated Old English Hexateuch (Pis. 9, 14, 29).18 With more than 400 pictures, the latter is, in fact, by far the most copiously illustrated volume of the period. If the reason for the predominance of Latin texts is not difficult to understand, these two remarkable Old English books reflect the considerable importance which was afforded to the vernacular in late Anglo-Saxon literary culture.19 It is greatly to be regretted that we have no evid­ ence, and hence can only speculate, concerning the original context for which these two volumes were designed, not to mention how they were used.

INHERITED ILLUSTRATIONS Pre-existing images and traditions of illustration were of considerable importance in determining the Anglo-Saxon illustrator’s approach to his task and in defining his visual vocabulary. Accordingly it will be useful to examine some of the inher­ ited cycles of illustrations that were available to him and consider his relationship with them. The Psychomachia is a case in point. Aurelius Prudentius Clemens lived from 348 to r.410, and wrote the Psychomachia, along with his other Christian poetry, 16 BL Add. M S 49598: Temple, Anglo-Saxon Manuscripts, cat. 23; colour plates and brief discussion in F. Wormald, Benedictional o f S t Ethelwold; monochrome facsimile: Warner and Wilson, Benedictional o f S t Æthelwold. 17 Ker, Catalogue, no. 334; Temple, Anglo-Saxon Manuscripts, cat. 58; facsimile: Gollancz, Cœdmon Manuscript; for the text see Anglo-Saxon Poetic Records i. The Junius Manuscript, ed. Krapp. The possible implications for the illustrations o f the mixed nature o f the underlying Latin text o f Genesis A (Remley, ‘Latin Textual Basis o f Genesis A ’) have yet to be assessed. 18 Ker, Catalogue, no. 142; Temple, Anglo-Saxon Manuscripts, cat. 86; facsimile: Dodwell and Clemoes, Illustrated Hexateuch; for the text see Old English Heptateuch, ed. Crawford. 19 On which see Bullough, ‘Educational Tradition’, and Kelly, ‘Anglo-Saxon Lay Society and the Written Word’. The role o f Old English in the late Anglo-Saxon church has yet to receive the extended discussion it deserves.


The Illustration o f Texts

towards the end of his life, perhaps r.405.20 The success of this first Christian, allegorical epic was phenomenal, as the numerous surviving manuscripts attest; it was his most frequently copied work in Anglo-Saxon England. The text was first supplied with illustrations in the fifth or sixth century, and twenty pre-thirteenthcentury manuscripts preserve some or all of an original set of about ninety pic­ tures.21 It is clear, therefore, that copyists of the volume in late Anglo-Saxon England had a well-established tradition to guide them in their work. We know that neither the Anglo-Saxon scribe or book-designer, if the two were separate in a given case, nor the artist had a particularly problematic task making this picturebook. The model showed both how to integrate the pictures with the text—they immediately precede the verses to which they relate—and what the pictures should contain to function as effective illustrations of the textual content. The de luxe Anglo-Saxon copy, Cambridge, Corpus Christi College, MS 23, is particularly instructive in this connection. Spaces remaining in the text suggest that the de­ signer envisaged the inclusion of extra illustrations which are not attested in any other version of the cycle and were presumably not present in the model. How­ ever, whereas the spaces which correspond with pre-existing illustrations were duly filled by the artist, those that did not were left blank. Hrabanus Maurus’ De laudibus sanctae crucis, the Marvels o f the East, Herbals, Cicero’s Aratea, the Occupational Calendar, and the Utrecht Psalter are the other extensively illustrated texts which are known to have been copied in late AngloSaxon England in this way.22 In all these cases, although individual artists and designers might introduce changes and variations, Anglo-Saxon scriptoria could profit similarly from inherited organization and iconography. The task of provid­ ing imagery for the liturgical volumes and library texts that required less extensive illustration was far less complicated in practical terms. Here too, earlier books provided an invaluable guide to the content and arrangement of the pictorial matter that could and should be supplied in a given case. The basic decorative programme in a late Anglo-Saxon gospel book, for instance, can generally be seen to follow that of Carolingian manuscripts. 20 T he most accessible edition is Prudentius, ed. Thomson, i. 274-343. For a summary review o f his poetry as a whole see Raby, Christian Latin Poetry, 4 4-71. For a detailed critical study o f the Psychomachia in relation to its late Antique context see M. Smith, Prudentius' Psychomachia. 21 They are all reproduced (greatly reduced) in Stettiner, Illustrierten Prudentius Handschriften, which should be supplemented by Woodruff, Illustrated Manuscripts o f Prudentius. W oodruff’s stemma is un­ helpful and requires reconsideration; however the fact that there are two basic families o f illustrations, and that the Anglo-Saxon manuscripts belong within the same family (ibid. 35) seems clear. T he nature o f the relationship between the pictures in the Anglo-Saxon copies would repay closer examination. 22 See Temple, Anglo-Saxon Manuscripts, cats. 14, 52, 87, 63, 42, 62, and 64; with Ohlgren, Insular and Anglo-Saxon Illuminated Manuscripts, cat. 224. One might also include Isidore’s De natura rerum (BL M S Gott. Dom. i) and Etymologiae (BL M S Royal 6 C. i; and Oxford, Queen’s College, M S 320) in which simple diagrams are integrated into the text. Opinions remain divided concerning the extent to which the illustrated Hexateuch, Cott. Claud. B. IV, and the ‘Caedmon’ manuscript o f Old English biblical poetry, Jun. 11, were indebted to earlier models. For the former, contrast Henderson, ‘Late Antique Influence’ and Dodwell and Clemoes, Illustrated Hexateuch; for the latter, contrast Raw, ‘Probable Derivation’ and Broderick, ‘Method o f Illustration’; see further in general Weitzmann and Kessler, Cotton Genesis, passim. In both cases the matter deserves a thorough re-examination.

The Illustration o f Texts


Yet, even when following exemplars which provided complete models for both text and illustration, Anglo-Saxons, like other artists, can be seen altering the images they had inherited, and this is exactly what we should expect.23 Some of the alterations were made deliberately for a clearly perceived effect; others were essen­ tially incidental, and it is important to be sensitive to the distinction if we are to interpret the material correctly. The belief that medieval copyists inevitably de­ based their material is clearly mistaken; however the reverse assumption, that all alterations were deliberate and meaningful, is equally untenable. Drawing a parallel between illustration and textual transmission can be misleading, for whilst absolute fidelity to the exemplar was the aim of the scribe, it was generally not that of the artist: at most his task was more like that of producing a paraphrase— transmitting the essence of a source in a new language. Nevertheless, the parallel is helpful to the extent that it underlines that even when complete fidelity to the source was the aim, genuine errors regularly crept in. At the most basic level discrepancies of detail and in overall effect arose through the differing abilities (occasionally one should say inability) of the copyists. The two Anglo-Saxon versions of the Marvels o f the East cycle demonstrate the dra­ matic contrast between a good and a second-rate artist’s interpretation of related archetypes: while one re-creates the inherited images, the other mars them.24 The same point is made to a lesser degree by the two surviving illustrated versions of Hrabanus Maurus’ carmina figurata, De laudibus sanctae crucis, the one exceedingly fine, the other rudimentary. In this case the contrast also highlights another, more subtle dimension of the issue, namely that the projected context of a volume influenced the calibre of the artist chosen to decorate it, and hence helped to determine the quality of the pictorial matter supplied. The set of painted diagrams that are by far the finest are found in Cambridge, Trinity College, MS B. 16. 3, an imposing volume which is handsomely written by a single scribe, measures a generous 418 x 341mm., and contains only this one text.25 It was clearly designed to be an elegant book and was surely prepared for a prestigious purpose or person. The less impressive set appear in Cambridge University Library, MS Gg. 5. 35, a sprawling Canterbury ‘classbook’ measuring 213 X 145mm., written by five scribes, in which De laudibus sanctae crucis is but one text amongst many.26 The striking 23 Compare Lowden, Octateuchs, 79-104. 24 BL M SS Cott. Tib. B. V, and Cott. Vit. A. XV. Comparison may conveniently be made both in James, Marvels o f the East and in McGurk, Illustrated Miscellany. See further Friedman, ‘Marvels of the East Tradition’. 25 The poems are written in rustic capitals, the commentaries in Anglo-Saxon square minuscule. Temple, Anglo-Saxon Manuscripts, cat. 14; Keynes, Anglo-Saxon Manuscripts, no. 4. For col. pi. of fo. 30' see Ramsay and Sparks, Image o f S t Dunstan, 11. For discussion o f the date see Dumville, ‘English Square Minuscule Script’, 1 75-6. As Linda Brownrigg first pointed out (unpublished lecture) the scribe is to be identified with the second scribe o f the respectably written but modestly conceived copy o f Amalarius’ De ecclesiasticis officus, Boulogne, BM M S 82 (Temple, Anglo-Saxon Manuscripts, cat. 29). 26 It occupies fos. 209v-225r. For illustrations o f fos. 211', 218', and 225r see Ohlgren, Insular and Anglo-Saxon Illuminated Manuscripts, pis. 48-50. Further on the manuscript as a whole see Rigg and Wieland, ‘Canterbury Classbook’. The Illustrated Miscellany, BL M S Cott. Tib. B. V originally included a copy of the text with figures (see McGurk, Illustrated Miscellany, 25).


The Illustration o f Texts

contrast in quality and effect between the two sets of paintings reflects the fact that a much higher premium was put on the appearance of the former book as a whole. None the less, with one possible exception, these differences are the by-products of discrepancies in natural talent and, perhaps, in the resources available at the time; they are not deliberate changes for a calculated effect. The exception is the omission from the Cambridge University Library manuscript of the full-page frontispiece which shows Hrabanus Maurus presenting his work to Pope Gregory IV.27 One might reasonably speculate that, whilst this seemed highly appropriate at the start of a de luxe volume consisting of this text alone, it was judged unneces­ sary in the context of a copy embedded within a ‘class’ or study book. Then there is the point that contemporary artistic styles and conceits tend to influence the artist’s treatment of his model to a greater or lesser extent. Here, by contrast, many of the changes involved were deliberately engineered for a clearly envisaged effect. A particularly striking example of the phenomenon is provided by the de luxe psalter, British Library, MS Harley 603 (Pis. 3, 12, 16),28 which was modelled on, and many of whose illustrations were directly copied from, the famous Carolingian psalter which is now MS 32 in the University Library, Utrecht.29 Although in the first phase of work (represented by the present fos. 1-27 and 507: Psalms 1-47 and 100-11) the creators of Harley 603 studiously preserved, page by page, the layout of the Utrecht Psalter, both the text and the script were altered to accord with Christ Church usage of the early eleventh century. The Gallicanum text was replaced by the Romanum,30 while Caroline minuscule was substituted for the rustic capitals of the earlier book (PI. 3). Precisely the same happened to the illustrations. The four Anglo-Saxon artists who were responsible for the work in question copied the Carolingian images with remarkable skill, even altering their manner of drawing to emulate the changes of hand and style there. In attention to pictorial content and layout much of their work is more aptly described as a facsimile than a copy; however they had resolved to modernize the ninth-century 27 Neither manuscript has the imagery for Louis the Pious, as in the 9th-c. Fulda copy, BAV M S Reg. lat. 124 (Miitherich and Gaehde, Carolingian Painting, pi. 12). 28 See Backhouse, ‘Making o f the Harley Psalter’; Gameson, ‘Anglo-Saxon Artists’ and ‘Roman­ esque Artist’; and Pfaff, ‘Eadui Basan’, 271-3. All the folios with illustrations are reproduced (reduced) in Ohlgren, Anglo-Saxon Textual Illustration, 147-248. For a monochrome facsimile plate o f fo. 54' see Schools o f Illumination, i, pi. 11; for representative colour plates o f work by each o f the earlier artists see Ramsay et al., S t Dunstan, pis. 7-14. 29 Facsimile: Van der Horst and Engelbrecht, Utrecht Psalter. Corresponding pages from the two books may conveniently be compared in Swarzenski, Monuments o f Romanesque Art, pi. 2; and Alexan­ der, Medieval Illuminators, ills. 120-1. 30 The Romanum is traditionally identified with St Jerome’s first revision o f the Old Latin text, although it is more probably the version which he was in fact correcting. T he Gallicanum was Jerome’s second revision, based on the Greek Septuagint and Origen’s Hexapla; introduced into Gaul it enjoyed considerable popularity and became the Psalter o f the Vulgate and the Roman Breviary. The Hebraicum, Jerome’s third translation, was based directly on the Hebrew. See further Kelly, Jerome, 158-63. Harley 603’s text is mainly Romanum; however Pss. 100 to 105: 25 (the work o f one particular scribe) are Gallicanum, while the section written by Eadui Basan (fos. 2 8 -4 9 ) is Romanum with occasional Gallicanum readings. For preliminary comments on the perception of the different Psalter texts at Christ Church at this time see Brooks, Church o f Canterbury, 261-5; and Pfaff, ‘Eadui Basan’, 27 2 -4 .

The Illustration o f Texts


drawings and all but one of them substituted the coloured-line technique that was currently fashionable at Christ Church, Canterbury, for the monochrome of the model. Needless to say the resulting alteration to the overall visual effect of the pictures is very dramatic. Most of the landscapes which provide the framework for the illustrations have in consequence a much greater surface pattern and decorative effect but a lesser illusion of recessional depth than those in the Utrecht Psalter. Occasionally the change also affected the clarity of individual details.31 The use of coloured lines also required the Anglo-Saxon artists to approach their task in a different way from their ninth-century predecessors. Instead of working system­ atically through the composition, drawing one area after another, they had to pro­ ceed piecemeal, drawing one small part of a figure or building in one colour and subsequently returning to draw the next part of the figure when they had changed to a different colour ink. One result of this working method was that they some­ times omitted small elements of the pictures. Another ‘fashion’ of late Anglo-Saxon art was the opulent foliage-adorned frame, and its effect on the appearance of an image was no less radical.32 An Anglo-Saxon evangelist figure may be a descendant of a Carolingian one, but when the tenth- or eleventh-century picture is dominated, and indeed the visual impact of the whole opening is determined by a lavish and distinctively Anglo-Saxon foliate frame, its appearance is very different, as a comparison between the portrait pages in, for instance, the ninth-century Abbeville Gospels and the eleventh-century Trinity College, Cambridge, Gospels demonstrates.33 To eleventh-century eyes the latter presumably looked more handsome, and this reminds us that, whilst the debt of Anglo-Saxon draughtsmen to Carolingian and other sources was undoubtedly great, none the less after the mid-tenth century they reformulated the images that influenced them with increasing confidence in order to make them their own. If some Carolingian artists tried to revive Classical and late Antique modes of pic­ torial representation, many Anglo-Saxon artists (and their Ottoman counterparts) set out to ‘modernize’ them. In this context it is worth reflecting that the criticism which is levelled at the stylized illustrations in early-medieval herbals by some botanically minded com­ mentators (or by commentators who assume that naturalism is a primary aim of all artists) is not entirely just.34 The stylization was, no doubt, partly the accidental 31 See further Gameson, ‘Anglo-Saxon Artists’. The change from Gallicanum to Romanum does not seem to be reflected in this phase o f work: I have not noticed elements in the Anglo-Saxon illustrations which suggest a response to the Romanum as opposed to the Gallicanum. However as these illustra­ tions appear to have been added to Harley before rather than after the text was written» this is hardly surprising. 32 The sources behind» and formal development of» these frames are discussed in Ch. 7. 33 Respectively Abbeville» BM M S 4 (Hubert et al., Carolingian A rt» ill. 77; see further Rosenbaum, ‘Evangelist Portraits o f the Ada School’); and T CC M S B. 10. 4 (Temple, Anglo-Saxon Manuscripts, cat. 65; Keynes, Anglo-Saxon Manuscripts, no. 20; and Ohlgren, Anglo-Saxon Textual Illustration, 348-71). 34 See Singer, ‘Herbal in Antiquity’, and Blunt and Raphael, Illustrated Herbal. More sympathetic is P. M. Jones, Medieval Medical Miniatures, 7 8 -8 1 , who points out that a degree o f stylization might in fact aid identification.


The Illustration o f Texts

result of repeatedly copying from models rather than from nature, and judged as aids to the identification of species, the pictures in Anglo-Saxon and Romanesque herbals are indeed less helpful than those in the earliest known examples of the genre such as the Vienna Dioscorides.15 Nevertheless, bearing in mind that some of these plants would have been unobtainable in the North, we should also recog­ nize that the changes could be partly a positive reinterpretation in the light of contemporary taste. These later depictions are significantly more decorative than their Antique precursors; they thus accord more nearly with contemporary canons of manuscript art, and were no doubt appreciated as such.3536 In contrast to such general stylistic alterations, some artists occasionally substi­ tuted or added details inspired by the real world of their own time or by alternative artistic sources to replace the antiquated fashions preserved in a given model. It should be stressed that this happens erratically (and certainly less than in Roman­ esque art), and that in general Anglo-Saxon artists were quite happy to repeat culturally obsolete images. Thus the aqueduct of the Utrecht Psalter (fo. 14v) reappears in Harley 603. On the other hand, the Anglo-Saxon artists of the latter book added weathercocks to the buildings they copied or redesigned, and some­ times they replaced the flat-hilted swords of the model with an Anglo-Saxon trilobate pommel akin to that of the ninth- or tenth-century Abingdon sword.3738 The same substitution of an Anglo-Saxon sword— this time replacing a round pommelled model—was also made by the Anglo-Saxon copyist of Cicero’s Aratea?* Such details rarely have any effect on the import of the illustration in question and it is often difficult to decide whether a given instance was an intentional modern­ ization or a subconscious reversion to a more familiar form. It can be equally difficult, as we have noted, to decide whether the influence of art or the real world lies behind some changes of this nature. An interesting case to consider in this context is the late eleventh-century herbal from Bury St Edmunds, Bodleian Library, MS Bodley 130.39 While most of the illustrations in this book are as stylized as those of other broadly contemporary herbals, a few undoubtedly reflect 35 Vienna, Öst. Nat., M S med. gr. 1: Weitzmann, Late Antique and Early Christian Book Illumina­ tion, pis. 15-20; see further his Ancient Book Illumination, 11-15. 36 This is especially true o f Romanesque copies such as BL M S Sloane 1975, and Bodl. M S Ashmole 1462 (for colour plates see Blunt and Raphael, Illustrated Herbal, 42-3, and 4 6 -7 ), but also applies to the m id -llth -c. Old English volume, BL M S Cott. Vit. C. Ill, and the late llth -c . manuscript, Bodl. M S Ashmole 1431 (col. pis.: ibid. 30-1). 37 T he Abingdon Sword is Hinton, Ornamental Metalwork 7 0 0 -II0 0 , cat. I, pi. 1. On the substitu­ tions in Harley 603 and the possibility that they reflect real-world items see Carver, ‘Contemporary Artefacts’. T he individual artists differed in their reaction to Antique artefacts according to their degree o f fidelity to the 9th-c. model as a whole. Thus B was more faithful than A, who was more prone to modernize. An approach to the question o f clothing may be made through Owen-Crocker, Dress in Anglo-Saxon England, chs. 7-10. This work does not specifically address the problem o f differentiating between the roles o f model and reality in late Anglo-Saxon illuminations (except superficially, 132-6), but see her comments in Owen, ‘Wynflæd’s Wardrobe’, 196, and also Dodwell, Anglo-Saxon Art, 171. 38 The Aratea is in BL M S Cott. Tib. B. V, its probable model being the Carolingian book in the same library, M S Harley 647 (see McGurk, Illustrated Miscellany, 67-78). 39 Facsimile: Gunther, Herbal o f Apuleius Barbarus', col. pis.: Blunt and Raphael, Illustrated Herbal, 3 4-5 and 38; see further Pacht, ‘Early Italian Nature Studies’.

The Illustration o f Texts


a fresh response to plants which would have been familiar to an English artist; yet it is uncertain whether these were introduced in this particular copy, or in an earlier late Anglo-Saxon version lying behind it. Certain changes to illustrations were induced by circumstances which were external to the imagery, and were sometimes, no doubt, beyond the control of the artist. If the new book differed appreciably in size from its exemplar, for example, it gave the artist correspondingly more or less space to fill with the picture. He could, of course, respond to the challenge by reducing or enlarging the scale of the image as a whole; alternatively—and especially if the shape of the available space were different— he might prefer to adjust its content. The point may be illustrated by comparing the two Anglo-Saxon copies of Prudentius’ Psychomachia which are now in the British Library, one of which is more than half as big again as the other.40 The two copies do sometimes differ in iconography, but even when they preserve essentially the same design, in the larger book the compositions are spread out while the artist of the smaller book often had to reduce the number of figures in order not to exceed the space available for the illustration. To cite a specific instance, whereas twenty figures menace Patientia on folio 8r of Additional MS 24199, only six could be fitted into the corresponding picture in Cotton Cleopatra C. viii,41 with the side-effect that the predicament of the Virtue here appears marginally less alarming. On the other hand, only two figures are required in the scene where Patientia promises Job peace, and when the artist of Additonal 24199 separated them so as best to fill his large picture space, he weakened the sense of contact between them. In Cotton Cleopatra C. viii the two figures are necessarily in close proximity and as a result are more clearly seen to be in communication.42 As we have indicated, the discrepancy in size can be seen to have had an indirect influence on subsidiary aspects of the visual narrative. The depiction of the Baptism of Christ in the Benedictional of St Æthelwold reveals another artist’s response to a related circumstance.43 If, as seems probable, his composition for the scene was based on a model similar to that on the Carolingian ivory casket from Metz which is now in Brunswick, the artist had to adapt a predominantly horizontal design to the upright-oblong frame shape that was used throughout the Benedictional.44 Interestingly his response was to compress the figurai elements into a square which occupies the central two-thirds of the frame and to fill the additional upper and lower areas with highly decorative turquoise and 40 MS Cott. Cleo. C. viii: 215 X 135mm. Add. M S 24199: 320 x 240mm. All the pages with illustrations in Cleo. C. viii are reproduced (enlarged) in Ohlgren, Anglo-Saxon Textual Illustration, 473-525; for a facsimile pi. see Schools o f Illumination, i, pi. 12a. 41 Cf. Stettiner, Illustrierten Prudentius Handschriften, pis. 51-2 (11) and (16). 42 Respectively fos. l l v and 13r. Ibid. 5 3 -4 (11) and (16). 43 BL Add. M S 49598, fo. 25r (Temple, Anglo-Saxon Manuscripts, ill. 85; reproduced in colour in Grabar and Nordenfalk, Early Medieval Painting, 180). 44 Brunswick, Herzog Anton Ulrich-Museum, MA 59: A. Goldschmidt, Elfenbeinskulpturen, i, no. 96. The relevant side is reproduced in colour in Brandt and Eggebrecht (eds.), Bemward von Hildesheim, ii. 402. For conveniently juxtaposed pictures of the casket and the page in question see Swarzenski, Monuments o f Romanesque Art, ills. 121 and 124; and Alexander, ‘Benedictional o f St Æthelwold1, pis. 8 and 9, with comment at p. 177.


The Illustration o f Texts

pink bands. If the modifications to the Psychomachia illustrations were incidental in origin and subtle in their effect, this change in the Benedictional reflects a deliberate decision and the result is visually dramatic. The coloured bands echo the predominant colours used within the picture area proper and contribute to the surface pattern of the page. The artist of the Benedictional of St Æthelwold here chose to indulge a taste for colourful abstract decoration rather than to rearrange his figures to occupy the available space. Whether this was his own decision, that of his patron, or the result of discussion between the two of them is impossible to say. The circumstance that the pictures in Anglo-Saxon herbals sometimes bear little relation to the specimens they purport to depict is less the result of inaccurate draughtsmanship than a combination of criteria of design along with the fact that the immediate models for these illustrations had themselves been altered through successive copying. We can, however, point to certain cases where tenth- and eleventh-century English artists do seem to have misunderstood aspects of their pictorial models. Some such misunderstandings distort the appearance of a detail but do not affect the iconography of the image as a whole—the inconsistent, contorted pose of St Luke in a gospel book which is now in New York,45 and the misalignment of Christ’s torso and waist in the image of the Crucifixion in the Sacramentary of Robert of Jumièges preserved at Rouen46 are relevant examples. On other occasions the iconography was affected. Attention may be drawn here to two contrasting examples from the Harley 603 Psalter. The illustrations in the Utrecht Psalter are highly complex and consist of a multitude of little details which depict individual phrases within the text. Accordingly, the task of reproducing each illustration with complete accuracy was a very daunting one, particularly given the piecemeal working method that was imposed upon the Anglo-Saxon artists by their use of the coloured line technique. Thus skilful and able though the original group of Anglo-Saxons generally were, certain points occasionally escaped them. For instance, in the bottom left-hand corner of the illustration to Psalm 13 in the Utrecht Psalter (fo. 7: PI. 3) we see two men fighting over a woman: as they struggle one stabs the other with a spear. But the Anglo-Saxon copyist, misinter­ preting the spear and misunderstanding the event, drew instead in his version the bizarre image of two men attempting to saw a woman in half!47 As it happens, this works just as well as an illustration of the relevant phrase in the text, ‘destruction and unhappiness are in their ways’ (v. 3), as did the original motif. However the same artist also made alterations and slips (and the fact that he was in all respects the freest of the first artists justifies our regarding some of them as such48) which distance his image slightly from the text. One occurs in the illustrations to Psalm 45 PM L M 708, fo. 42v: Ohlgren, Insular and Anglo-Saxon Illuminated Manuscripts, pi. 36. 46 Rouen, BM M S Y. 6, fo. 7 lv: Raw, Crucifixion Iconography, pi. 15. 47 Fo. 7V: Ohlgren, Anglo-Saxon Textual Illustration, 159; also reproduced (enlarged) in Beckwith, Ivory Carvings, ill. 48. 48 Pace Henderson, ‘Late Anglo-Saxon Religious Imagery’. See further Gameson, ‘Anglo-Saxon Artists’, 3 6 -4 0 .

The Illustration o f Texts


2.49 At the centre of the illustration in the Utrecht Psalter we see Christ with a rod in his hand and a broken pot at his feet, but in Harley 603 the pot is unbroken: consequently the motif is no longer quite so effective as an illustration of the relevant passage of the text (v. 9): ‘You shall rule them with a rod of iron and shall break them in pieces like a potter’s vessel.’ In this case the relationship between text and image altered slightly because the image itself was accidentally changed; altogether more significant alterations oc­ curred when details from inherited images or pictorial cycles were deliberately redeployed in new contexts.50 Cycles such as those mentioned above were not regarded as illustrations indissolubly joined to their source texts, but rather as an ‘image pool’, an encyclopedia of designs to which reference might be made when a need arose. No single cycle was more frequently referred to by late Anglo-Saxon artists than the Utrecht Psalter. It provided the inspiration for certain figure types, most notably that of the imploring suppliant with large hands, outstretched fingers, and head thrust forward that occurs in several contexts with various identities. Small relevant details were extracted from its compositions and inserted into vacant spaces in the early pages of an Anglo-Saxon psalter which is now in Paris;51 while its imagery provided the ultimate inspiration for many motifs in the Bury St Edmunds Psalter, and for at least one picture in the Tiberius Psalter.52 An image from the illustration to Psalm 60 in the Utrecht Psalter probably inspired the unusual iconography of the angel on a pedestal which occurs in the portrait of St Matthew in the Grimbald Gospels.53 It has been suggested that the frontispiece to the Junius 11 Codex of Old English poetry, which, incidentally, includes details that are not mentioned in the text it prefaces, is a pastiche of Utrecht Psalter motifs.54 The innovative Christological cycle which unfolds above the canon tables of the Arenberg Gospels depends on its imagery;55 while the architectural elements 49 Fo. 2r: Temple, Anglo-Saxon Manuscripts, ill. 200; Ohlgren, Anglo-Saxon Textual Illustration, 149. Utrecht and Harley may conveniently be compared at this point in Alexander, Medieval Illuminators, ills. 120-1. 50 The fundamental study o f this phenomenon, though concerned with the Antique ‘originals’ rather than with the subsequent contexts, is Weitzmann, Roll and Codex. Weitzmann’s pictorial stemma methodology has rightly been criticized as misleading, most recently by Lowden, Octateuchs, esp. 6 - 8 and 37-8. 51 Paris, BN MS lat, 8824: Temple, Anglo-Saxon Manuscripts, cat. 83; Avril and Stirnemann, Manuscrits enluminés, no. 25; facsimile: Colgrave, Paris Psalter. 52 Respectively BAV MS Reg. lat. 12 (Temple, Anglo-Saxon Manuscripts cat. 84; Ohlgren, AngloSaxon Textual Illustration, 249-97); and BL MS Cott. Tib. C. VI. (Ker, Catalogue, no. 199; Temple, Anglo-Saxon Manuscripts, cat. 98; F. Wormald, ‘Eleventh-Century Psalter’. For the Psalter text see Tiberius Psalter, ed. Campbell. For the suggested connection between its depiction o f the Harrowing of Hell and the Utrecht Psalter sec Openshaw, ‘Christ and Satan’, 19.) 53 BL Add. MS 34890, fo. 10v (Temple, Anglo-Saxon Manuscripts, cat. 68, ill. 215). >4 Broderick, ‘Method o f Illustration’, 165-9. 55 PML M 869: Temple, Anglo-Saxon Manuscripts, cat. 56; von Euw, Vor dem Jahr WOO, no. 45; Ohlgren, Anglo-Saxon Textual Illustration, 3 3 1-47. For a colour plate o f fo. IT (end o f Canon IICanon III) see Brandt and Eggebrecht (eds.), Bemward von Hildesheim, ii. 228. For a detailed explica­ tion o f the cycle and its relation to the Utrecht Psalter see J. Rosenthal, ‘Arenberg Gospels’, esp. 188-99, 211, and 219-24. It is possible that the artist o f the book is identical with Artist A o f Harley 603: see Gameson, ‘Anglo-Saxon Artists’, 40-1.


The Illustratton o f Texts

in the highly unusual frames which surround the evangelists in the same book are probably derived from its buildings (PI. 19),’6 as is a sketch on the flyleaf of British Library, MS Royal 15 A. xvi.’7 In addition, there is the famous and highly unusual drawing of the ‘Quinity’ in a prayer-book from the New Minster, Winchester, which was devised from elements in the Utrecht Psalter’s illustrations to Psalm 109, the Gloria, and the Creed.58 The wealth of this material provides a striking testimony to the extent of AngloSaxon interest in this particular cycle of inherited illustrations—and not merely at Canterbury where the Utrecht Psalter is known to have been kept, but also at Winchester. Equally striking is the versatility with which individual elements from it were reused and adapted. Visual imagery of all sorts from the Utrecht Psalter was sensitively redeployed in a wide variety of new contexts to supply new illustrational needs: it appears to have served as a visual thesaurus which greatly enriched the vocabulary of certain Anglo-Saxon illustrators, at least some of whom had evidently perused it thoroughly. The fact that six artists were employed in copying the illustrations of the Utrecht Psalter into Harley 603 at various times between c.1000 and £.1080, and the circumstance of its location at Christ Church, an important centre with a thriving scriptorium, provide a partial explanation for the extent of its influence. However this by no means fully elucidates the phenom­ enon. The extraordinary richness of the pictorial cycle, which includes literally hundreds of scenes and thousands of dramatic figures, the dynamism of the style in which they are drawn, the importance of the Psalter per se, the difficulties of illustrating its text in detail which are here triumphantly surmounted, and the presumably treasured status of this book itself should undoubtedly also be taken into account. It is notable that the pictures in both the Utrecht Psalter and Prudentius’ Psychomachia, the most influential and the most popular inherited cycle of illustra­ tions respectively, are dominated by violence, and this alerts us to another factor behind late Anglo-Saxon fascination with the art of this Carolingian Psalter. Whether or not the physical drama per se was part of their attraction, a key point is surely that the general message of the text and imagery in both books accorded well with contemporary notions of the earth and man’s soul as the battle-ground between God and the Devil, and of the monastic way of life as playing a crucial part in the struggle. The fusion between spiritual and physical combat that characterized late ,6 See further Gameson, ‘Anglo-Saxon Artists’, 41-2; and, more generally, J. Rosenthal, ‘Architec­ tural Settings o f the Arenberg Evangelists’. >7 Temple, Anglo-Saxon Manuscripts, cat. 85, ill. 211; see further F. Wormald, English Drawings, 4 4 5. The main manuscript, whose medieval provenance was St Augustine’s, Canterbury, is o f 9th-c. date and Continental origin and contains Juvencus, Evangelia, and other poems; Aldhelm’s Aenigmata; an extract from Bede’s De arte metrica, to which a glossary was appended in the late Anglo-Saxon period. See further Warner and Gilson, Catalogue, ii. 146. 58 BL M S Cott. Tit. D. XXVII, fo. 75‘: Temple, Anglo-Saxon Manuscripts, cat. 77, ill. 245. On the manuscript see further Ker, Catalogue, no. 202. On the imagery sec further Kantorowicz, ‘Quinity o f Winchester’, and Kidd, ‘Quinity Reconsidered'.

The Illustration o f Texts


Anglo-Saxon conceptions of this ultimately Pauline59 view of Christian life is forcibly (though not particularly lucidly) articulated in the New Minster Charter. In the chapter entitled, ‘How the abbot and the monks snatch away the king from the trial of devils’, we read: Moreover the abbot, girded with spiritual weapons, reinforced all about by a battalion of monks, drenched in the dew of heavenly grace, fighting the airy wiles of the devils, and with Christ’s assistance, with whose virtue they fight, this soldier, thoroughly unafraid, defend­ ing expertly with the sword of the spirit, from the rabid persecution of invisible enemies, protecting with the fine shield of faith as a means of defence, engaging in battle with robust triumph, snatches away the king and all the clergy of the kingdom.60

Both the Psychomachia and the Utrecht Psalter in their different ways provide compelling visual as well as textual equivalents for such modes of thought. They underline the reality of this struggle, showing the strength of the opposition, but, importantly, they also stress the ultimate triumph of virtue and the predominance of God respectively. They thereby provide highly relevant models and justification for the spiritual life of the monk.61

TH E PHYSICAL RELATIONSHIP OF TEX T AND IMAGE Late Anglo-Saxon books display a variety of approaches to the problem of physically attaching an image to a text, and of course Anglo-Saxon scriptoria had earlier exemplars to inspire and guide them in this.62 It should be emphasized at once that the nature of the physical relationship is to a significant extent dependent on the character of the text, and it is inseparable from the content of the imagery. The 59 Eph. 6: 11-17: lPut on the armour o f God, so you may be able to withstand the deceits o f the devil. For our struggle is not against flesh and blood, but against principalities and powers, against rulers o f the world o f darkness, against spirits o f wickedness in high places . . . Stand therefore, your loins girded with truth, wearing the breastplate o f justice, your feet shod in the readiness o f the gospel o f peace, always using the shield o f faith, so you may be able to resist all the fiery darts o f the evil one. Wear the helmet o f salvation, and the sword o f the spirit (which is the word o f G o d ). . .’ 60 ‘Abbas autem armis succinctus, spiritalibus, monachorum cuneo hinc inde uallatus. carismatum celestium rore perfusus, aerias demonum expugnans uersutias. regem omnemque sui regminis clerum. Christo cuius uirtute dimicant iuuante. a rabida hostium persecutione inuisibilium sollerter spiritus gladio defendens, fidei scuto subtili protegens tutamine, robusto prelians triumpho miles eripiat inperterritus.’ Ch. 15 (Councils and Synods, i, ed. Whitelock et al., 128). Cf. inter alia, the prayer Dominus Deus omnipotens rex regum (Pre-Conquest English Prayer Book, ed. Muir, 21 and 29-30), and Eph. 6: 11-17 (see n. 59 above). 61 Monks are depicted as warriors on fo. 69r o f Harley 603 (Ohlgren, Anglo-Saxon Textual Illustra­ tion, 239; KifT, ‘Images o f War’, pi. 3); Prudentius is depicted as a monk in Cott. Cleo. C. VIII, fo. 6' (Ohlgren, Anglo-Saxon Textual Illustration, 477). For analysis of the militaristic nature of the Psychomachia and its reflexes in Anglo-Saxon literature see Hermann, Allegories o f War, 7-36. The theme o f spiritual warfare in the illustration o f early medieval psalters as a whole is explored by Openshaw, ‘Weapons in the Daily Battle’ and for the Tiberius Psalter in particular see her ‘Christ and Satan’. 62 For a summary o f the Antique background see Weitzmann, Roll and Codex, 47-129.


The Illustration o f Texts

issue, ‘Where should the picture be placed?’ is intimately allied to the questions, ‘How can this text be illustrated?’ and ‘What will the visual matter express?’. A narrative cycle of illustrations cannot, for instance, be easily integrated into a non­ narrative text. If grade of book and projected function were factors which influ­ enced the amount of pictorial matter that was included in a given volume, the nature of the text and the character of the imagery were the crucial variables in determining how the one was attached to the other. In this section we shall consider frontispieces, interspersed imagery, and inte­ grated imagery in turn. A general point about the physical relationship between text and image which is worth stressing at the outset, however, is that the visual matter commonly precedes the text to which it relates. Consequently, the beholder sees the picture before he reads the writing. The text may well predate the illus­ tration, but in the finished book the illustration often precedes the text. Frontispieces A logical place to start an examination of the different methods that were used to join image and text is at the beginning of the book itself—with the frontispiece. Although a phenomenon of Antiquity and the early Christian period, pictorial frontispieces are not a feature of surviving non-liturgical Insular books, and the earliest examples found in Anglo-Saxon volumes of this nature date from the second quarter of the tenth century. There are two from this period: the famous image of King Athelstan and St Cuthbert which prefaces a handsome copy of Bede’s Uitae of that saint,6’ and the depiction of Hrabanus Maurus presenting his work to Pope Gregory IV which introduces the Trinity College, Cambridge, copy of De laudibus sanctae crucis.M It is difficult to know how much significance to attach to this coincidence in date. What is clear, however, is that both volumes, and particularly the latter, underline the importance of a new availablility of, and receptiveness to, Continental influences. Moreover, both, and especially the former, hint that the stimulus of Athelstan personally and the opportunities provided by his reign (924—39) were important catalysts in this as in other matters.65 In the case of de luxe liturgical volumes the most important frontispiece imagery may well have been supplied not within the book itself but on its cover. Certain Carolingian books, most notably the Dagulf Psalter, provide suggestive parallels here. The Dagulf Psalter was made between 784 and 795 for presentation to Pope Hadrian I.66 Although a splendid manuscript, it contains no pictorial matter M CCCC MS 183, fo. r . (Temple, Anglo-Saxon Manuscripts, cat. 6, ill. 29.) 64 T CC MS B. 16. 3, fo. V (ibid., ill. 48). 4> Further on the cultural dimension o f Athelstan’s reign see Keynes, ‘King Athelstan’s Books’; Dumville, ‘English Square Minuscule Script’, esp. 173-8; and Lapidge, ‘Schools, Learning, and Literature in Tenth-Century England', esp. 966-74. “ Vienna, Öst. Nat., M S 1891 (Charlemagne, no. 413; Irblich, K arl der Grosse, nos. 9-11; facsimile: Holter, Dagulf Psalter). It docs not seem to have reached the Pope, but rather to have remained at Charlemagne’s court.

The Illustration o f Texts


whatsoever; however the binding was decorated front and back with relevant imagery—a four-part cycle showing David and Saint Jerome, divided between two ivory plaques.67 Consideration of this and other Carolingian and Ottoman decorated bindings alerts us to the fact that such covers could be both visually magnificent and iconographically complex. Only a couple of Anglo-Saxon gospel books of the period preserve approximately contemporary precious-metal bindings, both prob­ ably of Flemish workmanship (PI. 4).68 Be that as it may, their subject matter— one shows a standing, blessing Christ in a mandorla surrounded by the four evangelist symbols and the other has Christ enthroned between seraphim in an upper register with the Crucifixion below it— is eminently likely to have been represented on the comparable bindings that were undoubtedly once made in England. Documentary evidence underlines the point. The precious pre-Conquest book bindings that were at Ely and are inventoried in the Liber Eliensis, for instance, included two with an image of Christ enthroned and nine with the Crucifixion, not to mention the one reputed to have been presented by King Edgar which was adorned with Christ, angels, and apostles on the front, and virgins on the back.69*The themes represented on the various ivory panels that survive from the period, at least some of which are likely to have been originally attached to book covers, include the Virgin and Child attended by angels, the Baptism of Christ, Christ enthroned (PI. 5a) and, by far the most common, the Crucifixion (PI. 5b).10 These panels are of such a scale that they may well have been arranged in groups to achieve more complex configurations. The most impressive ivories of Anglo-Saxon workman­ ship which are likely to have adorned a binding are the free-standing figures of the Virgin and St John in Saint-Omer.71 They clearly belonged to a crucifixion group, and in this case the size of the individual figures (125mm. high) indicates that, when complete, the group must have dominated the binding of its book. This was the image which first greeted the user of the volume in question, and it was probably the one that was most frequently seen. The extant frontispieces that occur within books are characterized by great variety. In addition to the inevitable differences of subject matter, the images take a wide range of forms and the amount of space they occupy varies from a few lines to several pages. For convenience of discussion the material can be loosely divided into two classes: frontispieces which are the only pictorial matter in their books, and those that supplement other imagery which follows. The artist of the frontispiece that was the only pictorial decoration of the 67 See A. Goldschmidt, Elfenbeinskulpturen, i, no. 3; Charlemagne, no. 518; Lasko, Ars Sacra, 29; and Gaborit-Chopin, Ivôtres, 4 6 -9 and 185. “ PM L M 709 (Campbell (ed.), The Anglo-Saxons, ill. 192), and M 708 (Ohlgren, Anglo-Saxon and Insular Manuscripts, ill. 34). Concerning the origin o f the latter see Hinkle, ‘Anglo-Saxon Gospel Book’, and Dodwell, Anglo-Saxon A rt, 201-3 (where Hinkle’s views are countered at 311 n. 136). 6" Liber Eliensis, ed. Blake, ii, ch. 50 (pp. 290-1). See further Dodwell, Anglo-Saxon A rt, 201-3. 7(1 See Backhouse et al., Golden Age, cats. 117, 119, 122, 123, and 126-8. 71 Musée Sandelin, no. 2822: Beckwith, Ivory Carvings, no. 25; Gaborit-Chopin, Ivoires, 8 8 -9 0 (with col. pi.); and Backhouse et al.. Golden Age, cat. 119, with col. pi. 27.


The Illustration o f Texts

volume had the task of providing the imagery that could best introduce a whole book. In the case of the Liber Uitae of the New Minster, Winchester,72 since the volume is a register rather than a conventional text, the artist was responsible for providing the principal subject-matter in the book. The draughtsman entrusted with the task rose admirably to the challenge and supplied two striking images arranged on three pages. The first page (fo. 6r: PI. 6) depicts a contemporary benefaction: King Cnut and Queen Ælfgifu presenting a gigantic cross to Christ at the altar of the New Minster, watched by the community of monks on earth and by the Virgin Mary and St Peter, patron saints of the foundation, in heaven. This was probably designed as a counterpart to the frontispiece of the New Minster Charter which it echoes both visually and in its message. Cnut, like Edgar before him, is shown as a model Christian king and a benefactor of monasticism. Further­ more, as was the case in the New Minster Charter, the imagery includes a depic­ tion of the book in which it occurs. The central monk at the bottom of the composition is holding an open codex which is presumably the New Minster’s Liber Uitae, and whose spiritual significance is underlined by the fact that it corresponds to the cosmic Liber Uitae that is held by Christ at the top of the page. The following opening in the manuscript (fos. 6 '-7 : PI. la and b) boasts a dra­ matic composition that summarizes the possible fates of the soul.73 Although not necessarily devised to be read in sequence, the images introducing this book serve to juxtapose time present with time eternal, and both in their different ways stress the interpenetration of mortal life and afterlife. The second composition, a com­ pelling vision which is simultaneously narrative and hierarchical, is a very apt image to preface the community’s ‘book of life’ and expound the virtues of being included in it; and a heavenly book of good deeds is shown playing a crucial role in the fate of a soul for whom St Peter is battling in the middle register on folio T. The first image physically incorporates King Cnut and his Queen in the ‘book of life’— including them in its benefits—in a particularly prominent way. A comparably extended frontispiece introduces the Sherborne Pontifical, a vol­ ume which was probably produced at Christ Church, Canterbury, for St Dunstan (d. 988).74 Here an image of the Crucifixion (fo. 4V) is followed by three pages which each show a similar standing figure type with a cross-nimbed halo: the first is crowned, bearded, and holds a cross-staff and a book, the second has cross-staff 72 BL, M S Stowe 944: Temple, Anglo-Saxon Manuscripts, cat. 78, ills. 244, and 247-8; for the text see Liber Vitae, ed. Birch; for specimens of the script see Watson,, Dated and Datable British Library, i. 162; with ii, pi. 36. For further discussion see Gerchow, ‘Prayers for King C nut\ 73 Facsimile pis.: Schools o f Illumination, i, pi. 13. The iconographie and conceptual framework for this image has yet to be explored in detail. For an interesting parallel in the 9th-c. triumphal arch mosaic at Santa Prassede, Rome, see Oakeshott, Mosaics o f Rome, ills. 121-2. Vercelli, Homily 15, on Judgement Day, describes St Peter locking the gates o f Hell and throwing away the key ( Vercelli Homilies, ed. Scragg, 260-1); here an angel is depicted performing this task (as is also the case in the 12th-c. Winchester Psalter, Cott. Nero C. IV, fo. 39r: F. Wormald, Winchester Psalter, pi. 42 and col. frontispiece). 74 Paris, BN MS lat. 943: Temple, Anglo-Saxon Manuscripts, cat. 35, ills. 134-8; see further J. Rosenthal, ‘St Dunstan’s Pontifical'.

The Illustration o f Texts


and book but no crown, while the third is beardless and holds a thin branch (fos. 5V-6 V). These three intimately related yet iconographically distinct figures may depict the Trinity. Alternatively, it has been suggested that they represent three aspects of the person of Christ, functioning as a visual counterpart to the presen­ tation of Christ in the Ordo for the dedication of the church which is the first text in the Pontifical itself.75 Whatever the truth of this particular case, the phenom­ enon of a frontispiece which relates especially closely to the beginning of the text is readily intelligible and has many parallels. One Anglo-Saxon example is found in the copy of Boethius’ De consolatione philosophiae probably dating from the third quarter of the tenth century which is now in Trinity College, Cambridge.76 This is prefaced by a monumental drawing of Lady Philosophy who appears much (though not exactly) as she is described at the beginning of Boethius’ text.77 Other instances are mentioned below. A highly respectable, equally long-established, and very common alternative to illustrating some aspect of the content of the text was to depict its author.78 One late Anglo-Saxon copy of Aldhelm’s prose De laude uirginitatis and one of his Carmen de uirginitate have pictorial frontispieces.79 In the former this consists of a drawing inserted between the proem and the first chapter of the text; in the latter it takes the form of two full-page drawings on the recto and verso of the first folio. Neither is dependent on the other, but in both cases it is the author himself and his presentation of the book to the nuns of Barking that is shown. The existence of well-established traditions of writing-author and book-presentation iconography encouraged this approach. The ubiquitous evangelist portraits not only provided con­ venient exemplars for writing figures, they acted simultaneously as positive, venerable recommendation for the practice of prefacing a respected text with such an image of its author. Moreover the image of a saint at the head of the text was a visual commendation of the quality of the spiritual writing to be found therein—akin to the succinct enumeration of the author’s qualifications and honours that appears on the covers of most modem academic tomes. The frontispiece to the Regularis concordia, the monastic reformers’ supplement to the Regula Benedicti, presents an interesting variation on this theme.80 The upper register of the composition shows 75 J. Rosenthal, ‘Three Drawings’; contested by Henderson, ‘Late Anglo-Saxon Religious Imagery’, 242-5. '6 TC C MS O. 3. 7: Temple, Anglo-Saxon Manuscripts, cat. 20, ill. 44; Keynes, Anglo-Saxon Manu­ scripts, no. 11. 77 Book 1, ch. 1, lines 1-25. 78 For the practice in Antiquity see Weitzmann, Ancient Book Illumination, 116-27. 79 Respectively, London, Lambeth Palace Library, M S 200; and Bodl. M S Bodley 577: Temple, Anglo-Saxon Manuscripts, cats. 39 and 57. In addition, a sketch o f a writing author appears on fo. 85’ o f BL M S Royal 7 D. xxiv, a copy o f the prose version dating from the second quarter o f the 10th c. (ibid. cat. 4, ill. 27; Warner and Gilson, Catalogue, i. 192). This, however, is unfinished, its date is uncertain, and it has been partly redrawn. BL, M S Cott. T ib., A. Ill: Temple, Anglo-Saxon Manuscripts, cat. 100, ill. 313; for a better photo see Backhouse et al., Golden Age, 49. On the manuscript as a whole, which is a very complex one, see Ker, Catalogue, no. 186. For this text see Regularis Concordia, ed. Symons. For a recent study o f the imagery see Deshman, ‘Benedictus'.


The Illustration o f Texts

Bishop Æthelwold, King Edgar, and Archbishop Dunstan side by side under arches, two haloed saints and one crowned ruler, a Trinity of God’s agents on earth, holding between them a long scroll— the text which they collectively en­ dorse. The Bishop and Archbishop hold thin staffs and look towards King Edgar, who, holding a palm, stares out at the beholder. Although the surviving manu­ script is of mid-eleventh-century date, it is probable that the basic composition was first devised around the time of the original promulgation of the text in the late tenth century;81 it was clearly designed to underline the dual authority, secular and religious, that lay behind the text, and to stress their intimacy and accord. Prefatory figures of the Deity present an analogous case: the artist was deploying very familiar images that had their own powerful devotional resonance which were also universally suitable ones— God can most appropriately introduce and preside over the content of any Christian book. The use of one such image as a frontispiece may also have been encouraged by, and have evoked, extra-textual homiletic asso­ ciations. In the typological linking of the church year and cosmic history, Lent was equated with earthly life, Easter was associated with the Second Coming, and the time after Easter with the afterlife. More specifically, the anonymous ‘Bückling Homilist’ warns that the Last Judgement will occur on Easter Day, and he begins his homily for Easter with a description of God in judgement. These associations may have influenced the artist of an Ælfric homiliary in which the first text is the homily for Easter Sunday and the frontispiece (which may originally have faced it) shows Christ the Just Judge.82 The large de luxe psalter, British Library, MS Harley 2904, provides a complimentary case to consider. The only image in this exceed­ ingly handsome book is the beautiful Crucifixion that prefaces and faces the begin­ ning of Psalm l,83 a place more commonly occupied by an image of King David. Although the arrangement is not without precedent—a depiction of the Crucifixion appears at Psalm 1 in the ninth-century Stuttgart Psalter, for instance84—the 81 See further Deshman, ‘Benedictus’. On its reception in the m id -llth c. see Gameson, ‘M idEleventh Century’, 77-9. 82 T CC M S B. 15. 34, fo. 1'. (Temple, Anglo-Saxon Manuscripts, cat. 74, ill. 241). Ælfric him self alludes only very briefly to the Judgement in his homily for Easter Sunday (Sermones Catholici, ed. Thorpe, i. 2 2 0 -8 , at 222-4). On the typological association see Gatch, ‘Anonymous Old English Homilies’, 129 ff. T he possibility o f a connection with the Bückling homily is suggested by Haney, Winchester Psalter, 58. T he homily in question is Bückling Homilies, 1, ed. Morris, 82-97. It is largely based on the Apocalypse o f Thomas (Apocryphal New Testament, ed. James, 556-62), although not in the association o f the Judgement and Easter Day. It should be noted that this homily is unique, except for the adaptation o f part o f it in an llth -c . Worcester collection (see Scragg, ‘Corpus o f Vernacular Homilies’, 233-5 and 253-7). If there is a connection between the illustration and the homiletic tradition represented in Bückling, 7, it is ironic that a collection o f Ælfric texts should be prefaced by a picture inspired by the tradition o f preaching he hoped his works would supersede. 83 Fo. .V. See Temple, Anglo-Saxon Manuscripts, cat. 41, with ill. 142; F. Wormald, English Draw­ ings, colour frontispiece. Contrary to Tem ple and Backhouse et al.. Golden Age, cat. 41, the book does not measure 285 x 242mm., but rather 335 x 250mm. For facsimiles o f fos. 3‘ and 144’ see Schools o f Illumination, i, pis. 9 and 10; for col. pi. o f fos. 36’ (text) and 125’ see Gullick, Calligraphy, pis. 11 and 12. 84 Stuttgart, Württembergische Landesbibliothek, M S Bibl. Fol. 23. Facsimile: Hoffmann, Miitherich, et al.. Der Stuttgarter Bilderpsalter.

The Illustration o f Texts


tenth-century Anglo-Saxon artist’s choice of picture can be seen as part of a developing Insular tradition for making specific reference to Christ in the prefatory matter to the Psalter.85 This image, like the various prayers that stress the connec­ tion between Christ’s life and the Psalms, defines a Christological frame of mind in which to perceive the Psalter text. More particularly, being the Crucifixion, it focuses attention on the Redemption (complementing the sinfulness of man which is alluded to at the start of Psalm 1). At the same time, as a simple yet highly emotive portrayal of the dead Christ and his grieving mother, it helps to foster in the reader an appropriately compunctive disposition. Sometimes a frontispiece referred to an owner or donor—such an individual is rarely depicted elsewhere in an Anglo-Saxon book86—and this is the visual coun­ terpart to the fact that, then as now, it was generally at the front of a volume that important names and commemorative inscriptions (as opposed to colophons) were written. The compositions prefacing the New Minster Charter, the New Minster Liber Uitae, and Bede’s Uitae Cuthberti which include Edgar, Cnut, and Athelstan respectively have already been mentioned. The circumstance that these examples all feature kings is a just reflection of the fact that only the very important were thus immortalized. Precisely the same applied on the Continent and in the East. What exactly these three images reveal about Anglo-Saxon kingship, contempo­ rary perceptions of it, and the relationship between Anglo-Saxon monarchs and monastic scriptoria, depends on whether or not it is believed that these are typical of late Anglo-Saxon ‘royal’ books and royal images—a difficult question to an­ swer.87 Yet what these Anglo-Saxon images undoubtedly all share is a concern to express the personal spirituality of the monarch, his persona as ‘brother of monks’,88 and his special relationship with the divine as channelled through both the text in question and a particular religious community. Pierpont Morgan Library, M 709, is one of the gospel books associated with Judith of Flanders (1032-94).89 The daughter of Adela of France and Count 85 See Haney, Winchester Psalter, 47-52. 86 An exception is Ælfwine who appears at fo. 19' o f the second portion o f the New Minster Prayer Book, BL MS Cott. Tit. D. XXVI (Temple, Anglo-Saxon Manuscripts, cat. 77; ill. 243); and who is also named in an inscription on fo. 65' o f the first portion, Cott. Tit. D. XXVII (Temple, Anglo-Saxon Manuscripts, ill. 246; facsimile: Schools o f Illumination, i, pi. 12c). Further on the manuscript see Ker, Catalogue, no. 202; and Watson, Dated and Datable British Library, i, no. 561; ii, pi. 35. An edition of the text is now available: Ælfwine 's Prayerbook, ed. Giinzel. 87 Heslop, 'De Luxe Manuscripts’ presents a strong case for associating a large number o f the extant de luxe gospel books with the patronage o f Cnut and Emma. T he flaw in this aspect o f his invaluable study is the complete lack o f any dedication image or inscription within the volumes in question. He acknowledges the problem (pp. 180-1); however his suggestion, namely that the appropriate words or images could have been contained on the lost, presumably decorated, bindings, lacks conviction in the face o f strong parallels for including such matter within rather than (or as well as) on the outside o f such books. The issue o f the relationship between royalty and ecclesiastical art is discussed further in Ch. 8 below. 88 See further Wollasch, ‘Kaiser und Könige als Bruder der Mönche’. m PM L M 709, fo. 1'. (Temple, Anglo-Saxon Manuscripts, cat. 93, ill. 289; see further Harrsen, ‘Judith o f Flanders’, and, on the frontispiece, Raw, Crucifixion Iconography, 158-61; Gameson, ‘MidEleventh Century’, 89-90.)


The Illustration o f Texts

Baldwin V of Flanders, Judith married Tostig Godwinson and came to England in 1051. After Tostig’s death (1066) she married Welf, Duke of Bavaria, ending her life a widow in Weingarten Abbey, Bavaria. The frontispiece to M 709 is a fullpage miniature of the Crucifixion, and in it there appears a diminutive female figure clutching the bottom of the cross. If, as seems likely, this is meant to be Judith herself, then here we have a prefatory image which includes a depiction of a contemporary laywoman. The appearance of such a figure in this context is a just reflection of the fact that grand women are documented as important patrons and owners of fine books in the eleventh century.90 This frontispiece is a personal, devotional icon which intimately associates Judith (assuming it is meant to be she) with the historic event, and perpetuates her in a direct relationship with the Redemption. At the same time, it stresses the woman’s self-abasement. Now Judith was a powerful lady—indeed only a very eminent individual could have commissioned and owned such a book—and this reminds us that humility before Christ was a highly potent form of self-negation. Judith’s imagery is an example of a mode of depiction which was sensitively and quite widely used in early medieval art in association with important contemporary figures. Of the various comparable compositions, perhaps the most closely similar to that in Judith’s Gospel Book (though with a male suppliant) appears in the slightly earlier Gundold Gospels from Cologne.91 The theme of humility before Christ was, as we have noted, a very powerful one, and it was used with different overtones in another, earlier Anglo-Saxon frontispiece— that to St Dunstan’s ‘Classbook’.92 Dunstan is here shown as a very small figure prostrate and seemingly shielding his eyes beside a towering Christ. His exemplary humility functions as a highly potent indication of his spirituality, and this, along with his intimate, personal relationship with Christ, underscores his entitlement to special authority on earth.93 It is interesting to compare this image, which highlights Dunstan’s humility before Christ, with the frontispiece to the Regularis concordia which stresses his authority among men. At one level, the contrast reflects the difference between a personal devotional image, perhaps drawn by the saint himself, and a ‘published’ icon for general monastic consumption, which was probably designed and drawn by someone else. This contrast is paral­ leled in writing—while Dunstan describes himself as indignum,94 the Regularis 90 See Gameson, ‘Mid-Eleventh Century’, 70 -1 . 91 Stuttgart, Württembergische Landesbibliothek, M S 402, fo. 9' (A. Goldschmidt, German Illu­ mination, pi. 88; von Euw, Vor dem Jahr 1000, no. 4, with ill. on p. 39). 92 Bodl. M S Auct. F. 4. 32, fo. 1': Temple, Anglo-Saxon Manuscripts, cat. 11, ill. 41; facsimile: Hunt, S t Dunstan’s Classbook. See further Higgitt, ‘Glastonbury, Dunstan’, and Budny, “ ‘St Dunstan’s Classbook” and its Frontispiece’. 93 See further Deshman, ‘Exalted Servant’, discussing the point in relation to the Prayer-book o f Charles the Bald. 94 ‘Indignum abbatem Dunstanum XPE respectes’— the phrase spelled out by the final letter o f each line in Dunstan’s acrostic poem, O Pater Omnipotens (printed in Lapidge, ‘Hermeneutic Style’, 108-11; see further his ‘Dunstan’s Latin Poetry’). A particularly interesting use o f this conceit is that o f Ramwold, Abbot o f Saint-Emmeram (975-1000) in the miniature he had added to the Codex Aureus

The Illustration o f Texts


concordia styles him egregius.95 At another level, however, the force of the Regularis concordia image is underlined by the implications of the drawing in the ‘Classbook’: it is Dunstan’s outstanding personal spirituality and his humility before Christ that gives him great authority among men. Moreover Christ was not only the point of reference but also the model for this philosophy. St Paul stated: Yours is to be the same mind which Christ Jesus showed. His nature is, from the first, divine, and yet he did not see in the rank of Godhead a prize to be coveted; he dispossessed himself, and took the nature of a slave, fashioned in the likeness of men, and presented himself to us in human form; and then he lowered his own dignity, accepted an obedience which brought him to death, death on a cross. That is why God has raised him to such a height, given him that name which is greater than any other name; so that everything in heaven and on earth and under the earth must bend the knee before the name of Jesus, and every tongue must confess Jesus Christ as the Lord, dwelling in the glory of God the Father. (Phil. 2: 5-11.)

Not every owner or patron who was depicted in the frontispiece of an AngloSaxon book was shown in a deferential mode. Altogether different is the anony­ mous ecclesiastic who dominates the full-page image that heads the Herbal, British Library, Cotton Vitellius C. iii (a text, incidentally, that had a frontispiece al­ ready).96 By far the largest figure and centrally placed on the page, he is shown triumphantly treading on a beast which he transfixes with his pastoral staff (in imitation of Christ triumphant), flanked on one side by a small soldier, on the other by a humble monk who presents him with a book. The depiction was clearly designed to flatter, and, as such, it is broadly reminiscent of other presentation images—that of Archbishop Egbert of Trier (977-93) receiving his book of pericopes from the diminutive monks Keraldus and Heribertus which prefaces the Codex Egberti for example.97 This Ottonian picture is visually far more magnificent than that in the Vitellius Herbal (as indeed is the book in which it appears). In terms of iconography, however, it is the prefatory image in the Herbal that goes the furthest in exalting its recipient. It is a great pity that the identity of the Anglo-Saxon ecclesiastic who was thus depicted, and the circumstances behind the commission, are now unknown. o f Saint-Emmeram (Munich, Bayerische Staatsbibliothek, Clm. 14000, fo. l r: Mütherich and Dachs, Regensburger Buchmalerei, no. 13, pi. 91). Ramwold stands at the centre o f an elaborate decorated page; around him are the four cardinal virtues and the four evangelist symbols. An inscription records his restoration o f the manuscript. At the top o f the page in display capitals is his name; towards the bottom in matching capitals is Indign[us]abbas. A powerful mixture o f hubris and humility. 95 Regulans Concordia, ed. Symons, sect. 7 (p. 4). % Apuleius’ Herbarium: BL M S Cott. Vit. C. iii, fo. l l v (Temple, Anglo-Saxon Manuscripts, cat. 63, ill. 187). This prefaces the table o f contents (fos. 12-18). The ‘inherited’ pictorial frontispiece to the Herbal occurs on fo. 19r, followed on fo. 19v by a full-page presentation o f the title o f the book surrounded by an ornamental fillet. The imagery o f both frontispieces is discussed by Voigts, ‘Her­ barium Apulei’. For a full description o f the manuscript see Ker, Catalogue, no. 219; and Old English Herbanum and Medicina de Quadrupedibus, ed. de Vriend, pp. xi-xx. 9/ Trier, Stadtbibliothek, M S 24, fo. 2r (facsimile: Schiel, Codex Egberti).


The Illustration o f Texts

The Vitellius Herbal and Judith of Flanders’ Gospel Book have other illustra­ tions in addition to the frontispieces which we have discussed. The former in­ cludes an ‘author’ portrait and a series of ‘scientific’ illustrations; while, as we would expect, there are portraits of the four evangelists in the latter. All the frontispieces considered in the remainder of this section are in books which, like these, contain other imagery. Pictures of Christ enthroned were sometimes used to preface gospel books, and we noted earlier that the subject appeared, along with the Crucifixion, on their bindings (Pis. 4, 5a). The most striking surviving Anglo-Saxon depictions of Christ in this context appear after the canon tables, before the text of Matthew’s gospel— presiding in effect over all four accounts of Christ’s life— in the Trinity Gospels98 and in Boulogne, Bibliothèque Municipale, MS 11, a gospel book which was made at St Bertin but was decorated by an Anglo-Saxon artist.99 This is a variation on the tradition of placing a maiestas domini image before the gospels which is found in the Codex Amiatinus , \lb). These present elaborate visual meditations on the nature of Christ, and on David and Christ respectively, which the beholder encounters before he reaches the core texts of the volumes. It is interesting to remark that the gospel book, that is the volume which contains the narrative accounts of Christ’s life, is prefaced by a thematic Christological cycle, whereas the Psalter, which of course has no explicit reference to Christ, is prefaced by a typological narrative cycle. In both cases the imagery is carefully calculated to complement and expand the nature of the writings that follow, while also defining their significance. The exact nature of the complementary images which adorn the text of the Benedictional of St Æthelwold varies quite widely, as do the literary and icono­ graphie sources behind them. A few miniatures, notably those of Saints Etheldreda 171 T he relevant comparative material is conveniently set out in Weitzmann and Kessler, Cotton Genesis, 5 4 -6 , with ills. 3 9 -4 9 , and is discussed by Raw, 'Probable Derivation’, esp. 139- 46. For a recent thought-provoking review o f the status o f the Cotton Genesis considered in relation to the other extant illustrated ‘Genesis’ manuscripts, see Lowden, 'Cotton Genesis’. 172 2 Cor. 11: 14. 1,3 Apocalypses Apocryphae, ed. Tischendorf, 1-23. For the influence o f the Uitae Adae et Euae on Genesis imagery see Kessler, Illustrated Bibles from Tours, 28-32. On the Old English text see further Woolf, 'Fall o f Man’, 17-19; and, suggesting that the shift reflects Eve’s state o f mind, Saxon Genesis, ed. Doane, 141-53.


The Illustration o f Texts

(PI. 23a) and Swithun,174 show individual, semi-iconic figures; others depict events for the most part of biblical origin (PI. 10); all are replete with symbolism. The illustrations of biblical events are not simply visual representations of the relevant scriptural texts. Orthodox teaching required the Christian to look for the inner spiritual meaning of each recorded happening; correspondingly the aim of the artist was to present in his depiction both the historic event and its Christian significance.175 We noted above how the juxtaposition of depictions of the Adora­ tion of the Magi and the Baptism of Christ before the blessings for Epiphany in the Benedictional of St Æthelwold reflected traditional exegesis of that feast and its general import.176 Christian iconography had from the earliest times been con­ cerned with conveying the spiritual significance of biblical events as well as the mere occurrences themselves, and many of the iconographies which the AngloSaxon artist had inherited were accordingly designed to do this. We can sometimes see tenth- and eleventh-century artists refining the traditions they had inherited. Individual details in both parts of the Epiphany diptych in the Benedictional, for example, place a novel stress on the kingship of Christ, a theme which is taken up elsewhere in the imagery of the book.177 To demonstrate the phenomenon in more detail we may usefully consider the very first image in the body of Æthelwold’s Benedictional, namely the Annuncia­ tion (PI. 10).178 Under the title, ‘The messenger from heaven stands here to announce to Mary: “behold, blessed one, you will give birth at one and the same time to God and man” ’,179 we see the angel Gabriel on the left, his right hand raised in blessing, approaching Mary who is seated under a canopy holding a golden oval in one hand (which has reasonably been interpreted as a shuttle), her other hand resting on an open book which is supported on a draped lectern. A swirling cloud extends from around the figure of Gabriel to embrace Mary’s head—an evocation of the Holy Spirit coming upon her. The basic model for the composition seems to have come from Metz, and it is paralleled fairly closely in two ninth-century ivories from that centre.180 As well as depicting the biblical account of the Annunciation (Luke 1: 26-38), certain details of the design, notably the book and the shuttle, reflect the apocryphal account of the Protoevangelium of James and the derivative Gospel of Pseudo-Matthew which are, in effect, semiinterpretative.181 Pseudo-Matthew explains that Mary was supremely well-versed in the Psalms—a point which, as later commentators stressed, made her the fitting choice for the Mother of God— while she is credited with weaving a purple cloth 174 Respectively fos. 90v (F. Wormald, Benedictional o f S t Ethelwold, pi. 6) and 97v (Deshman, ‘Anglo-Saxon Art after Alfred’, ill. 51). ,7S This is discussed further in Ch. 5 below. 176 See nn. 125-6 above. 177 See Deshman, 4Christus Rex'. 178 Fo. 5V: Rickert, Painting in Britain, pi. 26; see further Clayton, Virgin M ary, 160-1, and pi. 5. 179 ‘Nuntius e caelo hic stat praedicando Mariae: | Ecce Deum paries hominemque simul benedicta.’ 180 On the Brunswick and Louvre caskets (A. Goldschmidt, Elfenbeinskulpturen, i, no. 96; Schiller, Iconography, i, ill. 74). 181 Apocryphal New Testament, ed. James, 3 8 -4 9 at 43; Evangelia Apocrypha, ed. Tischendorf, 4 9 112 .

The Illustration o f Texts


for the veil of the Temple, a very suitable activity for a ‘type’ of the church. Mary’s nobility and her role as the pure temple, the appropriate vehicle for the incarnate Lord, are further conveyed by the baldachino under which she sits, with its resonance of containing the reserved sacrament. Moreover, the additional complex of buildings beyond this structure, to which it thus serves as the entrance, simul­ taneously alludes to her position as the doorway between heaven and earth. This was a dual role. On the one hand she was the intermediary through which God passed to mankind; on the other she was man’s way back to paradise—as Ælfric said, ‘our old mother Eve shut the gate of the kingdom of heaven to us, and the holy Mary afterwards opened it to us’.182 The theme of the door of heaven is echoed in the second blessing for the Feast of the Annunciation that follows the miniature in the Benedictional which begins: ‘Open the doors of heaven, O Lord, and visit your people in peace, and send your spirit from on high and water our land so that spiritual fruit may grow in us.’183184 It is quite clear that careful thought went into the creation of the miniatures in such books. This was not just directed at the iconography but also at the overall visual effect, and it is possible on occasions to watch the interplay of aesthetic considerations, iconography, and meaning. A comparison of the pictures for the Dormition of the Virgin Mary in the two illustrated Winchester Benedictionals is instructive in this respect (PI. 1la and b).m The relevant De transitu beatae Mariae texts and Old English versions of it describe a dramatic death-bed scene: Mary rests, attended by three virgins; all the apostles are in attendance and Christ enters with a great company of angels.185 The illustration in the Benedictional of St Æthelwold (which does not, of course, contain this text) includes all these ele­ ments, with the additional touch that the hand of God holds a crown above the Virgin. This is the earliest-known depiction of the death of the Virgin in the West;186 moreover because of the presence of the crown, indicating victory over death, it represents an important preliminary stage in the history of the iconogra­ phy of the Coronation of the Virgin.187 The novelty of the image reflects and 182 ‘Ure ealde moder Eva us beleac heofenan rices geat. and seo halige Maria hit eft to us geopenode’ (‘D e Natale Dom ini’: JElfric's Catholic Homilies, ed. Godden, 11). Cf. Blickling Homilies, ed. Morris, 9 (‘On the Annunciation’): ‘Iu geara heofonrices duru . . . belocen standeþ þurh þa ærestan men, nu heo sceal þonne þurh þe ontened beon.’ 183 Fo. T . ‘Aperi d[omi]ne ianuas caeli, et visita plebem tuam in pace, et mitte sp[iritu]m tuum de alto, et irriga terram n[ost]ram. ut germinet nobis sp[irit]ualem fructum. Arnen.’ 184 Add. M S 49598, fo. 102v (Rickert, Painting in Britain, pl. 27), and Rouen Y. 7, fo. 54v (Temple, Anglo-Saxon Manuscripts, ill. 87). T he two are conveniently juxtaposed in Clayton, Virgin M ary, pis. 6 -7 . 185 Apocalypses Apocryphae, ed. Tischendorf, 124-36; Apocryphal New Testament, ed. James, 194— 227; Blickling Homilies, ed. Morris, 137-59 (on which see Willard, ‘Blickling Homily XIII’, and Scragg, ‘Corpus o f Vernacular Homilies’, 234 and 241); and Three Homilies, ed. Grant, 13-41. The popularity o f the story in Anglo-Saxon England is discussed by Grant at 14-15; see further Clayton, Virgin Mary, 2 3 2 -4 4 . The question o f which version o f the Transitus text (A, B2, or C) may have inspired the artist is reviewed ibid. 162-3. 186 For a summary o f the early iconography o f the Assumption see Boase, York Psalter, 8-14. ,R/ See further Verdier, Le Couronnement de la Vierge, esp. 19-21.


The Illustration o f Texts

further underlines the emphasis placed on the Virgin in the visual articulation of Æthelwold’s book. The iconography is without close parallel and a case has been made for its being the personal creation of the artist of the Benedictional, who used motifs which are found elsewhere in his work to create an illustration of the De transitu text.188 Certainly his composition is quite distinct from the Eastern type which informed the Ottoman versions of the subject that were produced at a slightly later date.189 Yet, highly significant and very interesting though the com­ position is, it is also undeniably overcrowded. The artist of the later but intimately related Benedictional that is now in Rouen evidently seems to have felt so. An image like that in Æthelwold’s manuscript provided the starting point for his design, but he greatly simplified it. It is interesting to see what he did. He excluded both angels and apostles, and also removed the hand of God from the picture area, placing it instead in the arch of the frame. The other notable changes are that one extra female mourner has appeared, and Mary’s orans gesture is more pronounced. The resulting picture is bolder, clearer, and visually more successful; at the same time the modifications have affected the precise connotations of the image, pre­ sumably deliberately so. No longer a detailed illustration of the account in the De transitu, the refined image is more an icon of the Virgin which stresses both her special relationship with God, and her role as intermediary between heaven and earth (the latter represented by the mourners at her side and the attendant at her head).190 Instead of a wealth of anecdotal detail, the new image places greater emphasis on the Virgin herself, on her serenity and prayerfulness at death, and on the heavenly crown above her; equally it highlights the relationship between Mary and the mourning women, symbols of devoted humanity.

Literal Illustration The impetus and example for literal illustration was principally provided by the Utrecht Psalter; and the work of the first group of artists in the Harley 603 Psalter, who generally reproduced its pictures with remarkable iconographie accuracy, is the major Anglo-Saxon monument in this technique (PI. 3). The artists of the Utrecht Psalter (and their Anglo-Saxon followers) successfully translated the poetic, non-narrative text into an elaborate series of literal word illustrations. So, for example, when Psalm 1 likens the blessed man to ‘a tree which is planted near the running waters, which shall bring forth its fruit in due season, and its leaf will not fall o ff. . .’ (v. 3), this is rendered by an image of a leafy tree, laden with fruit, standing beside a stream. The vignettes which depict the individual phrases of 188 Alexander, ‘Benedictional o f St Æthelwold’, 178-9. 189 See Demus, Byzantine Art and the West, ills. 10, 25, 86, and 174; Mayr-Harting, Ottoman Book Illumination, i. 139-55; and A. Goldschmidt, German Illumination, pi. 67. 190 See further Deshman, ‘Servants o f the Mother o f God’, esp. 59.

The Illustration o f Texts


each psalm are placed within a common landscape setting to form a single picture; the setting makes an important contribution to the visual success of the whole, since it gives the compositions a spatial unity which compensates for their lack of thematic unity and narrative flow. This remarkable Carolingian manuscript evidently fascinated the community at Christ Church, Canterbury, and exerted an important influence on artists who were active there and elsewhere: clear echoes of its style and iconography reappear in a variety of contexts, as we have seen. The technique of literal illustration dominates the work of the artists of the Paris and the Bury Psalters, much of whose imagery is related— whether directly or indirectly is unclear—to the Utrecht Psalter cycle. For example, the king standing on a mound inscribed ‘mons syon’ and holding a scroll which bears the words, ‘But I am appointed king by him over Sion his holy mountain’ (Ps. 2: 6), who appears beside Psalm 2 in the Bury Psalter,191 and the figure of Christ who breaks a vessel with a rod, for verse 9 of the same psalm (‘you shall rule them with a rod of iron and shall break them in pieces like a potter’s vessel’) in the Paris Psalter192 are both literal illustrations of the texts in question, and both seem to derive from the figure of Christ who appears at the centre of the illustration for this psalm in the Utrecht Psalter (fo. 2r). It is interesting to note that although the image has been used differently in the two manuscripts and illustrates discrete phrases—the Paris Psalter follows Utrecht very closely, Bury diverges from it—nevertheless in both cases the actual mode of illustration remains the same. Both the Paris Psalter and the Bury Psalter include details that are not present in the Utrecht cycle but which remain within the tradition of literal word illustra­ tion, and they thereby attest to the wish and the ability of eleventh-century English artists to refine pictures of this sort and create them for themselves, making use of other sources, visual traditions, and their own inventiveness to do so. The Paris Psalter contains a parallel Latin and Old English text. The drawing of a devil who shoots arrows at a man and woman embracing within a simple building, which occurs in Psalm 7 immediately after verse 14,193 does not come from the Utrecht cycle, and it relates not to the Latin text (in which it is set) but to the divergent Old English translation of verses 13-14. This reads as follows: ‘And [the devil] bends his bow which is now ready to shoot; he intends to shoot the vessel of death, that are the unrighteous. He takes his fiery arrows that he may shoot with them, and burn those who are here on fire with lust and vice.’194 In the circumstances it seems likely that the motif is the Anglo-Saxon artist’s personal rendering of this Old English text. 1.1 Fo. 22': Rickert, Painting in Britain, pi. 39b; Ohlgren, Anglo-Saxon Textual Illustration, 250. 1.2 Fo. 2': Ohlgren, Anglo-Saxon Textual Illustration, 299. 1.3 Fo. 6': ibid. 302; see further Harris, ‘Anglo-Saxon Psalter in Paris’. 1.4 ‘And he bende his bogan, se is nu gearo to sceotanne; he teohað þæt he scyle sceotan þæt deaðes fact, þæt synt þa unrihtwisan; he gcdeð his flan fyrena þæt he mæge mid sceotan and bærnan þa þe her bymað on wrænnesse and on unðeawum.’ For the text see West Saxon Psalms, ed. Bright and Ramsay, 12.


The Illustration o f Texts

As another example we may consider the illustration for Psalm 12, verse 4 (‘Consider and hear me, O Lord, my God. Enlighten my eyes that I never sleep in death’) in the Bury Psalter.195 In the Utrecht Psalter’s illustration for this psalm (fo. T) we see the Lord in the heavens directing a flaming horn towards the Psalmist who, sitting under a tree, receives its beam of light at a distance. The Bury Psalter uses the same two figures but the relationship between them is much more direct: a youthful, beardless Christ with a cross-nimbed halo now stands on the shoulders of the Psalmist, his horn immediately above his head, its flames pouring straight into the latter’s upturned face. Among other modifications, the Psalmist now has his hands outstretched and a scroll is looped between them; on this is written the second half of the verse in question: ‘Illumina oculos meos nequando obdormiam in morte ne dicat inimicus preualui aduersus eum.’ An ‘intermediary’ stage between the two versions appears in the Odbert Psalter196 where Christ stands above, yet slightly to the side of the Psalmist. In part, the change can be seen as the inevitable result of deploying the composition in the more restricted space of the margin of a page. At the same time, the more radically changed motif in the Bury Psalter seems to be assimilated to a diagrammatic representation of Uita and Mors— where Christ, the symbol of life, stands above a bestial demon, the symbol of death, both holding scrolls between their hands— which was current in late Anglo-Saxon England and appears in the form that we have described in the Tiberius Psalter.197 The title of this psalm, Uox Christi ad patrem de diabolo, may have encouraged the artist of the Bury Psalter to make the association between the inherited imagery and this diagram. Assuming that famili­ arity with this ‘chart’ indeed lay behind the treatment of the motif in the Bury Psalter, then we see the artist grafting a new dimension on to his image which makes it relate more intimately to the words of the text: the Lord is more clearly both illuminating the Psalmist’s eyes and preserving him from death. Yet, important though it undoubtedly was, one should not overestimate the popularity and significance of literal illustration. Alternative modes were some­ times used in the Bury Psalter, and when motifs from the Utrecht Psalter were redeployed in different ways in association with other texts, they almost inevitably became another type of image. The imagery in the canon tables of the Arenberg Gospels,198 which we mentioned earlier as an example of complementary illustration, is a case in point. Particularly interesting in this connection is the contribution of Artist F to the Harley 603 Psalter (Pis. 12, 16).199 This draughtsman was respons­ ible for the last folios of the manuscript as it survives and he worked after, though 195 Fo. 28r: Rickert, Painting in Britain, pi. 39a; Ohlgren, Anglo-Saxon Textual Illustration, 255. See further Heimann, ‘Three Illustrations’, 39-46; and Harris, ‘Marginal Drawings o f the Bury Psalter’, 144-51. ,% Boulogne , BM M S 20, fo. 19v. 197 Fo. 6' (F. Wormald, Collected Writings, i, ill. 125). It appears in a slightly different form, arranged on two pages in the Leofric Missal, Bodl. M S Bodley 579, fos. 49v-5 0 r (Temple, Anglo-Saxon Manu­ scripts, ills. 55-6). 19H For which see n. 55 above. 199 On whom see Backhouse, ‘Making’, 99; and Gameson, ‘Anglo-Saxon Artists’, 31-2; with the illustrations in Ohlgren, Anglo-Saxon Textual Illustration, 221-48.

The Illustration o f Texts


probably not very long after, the first group of artists.200 Whatever the time lapse, the nature of the project in this phase of production had changed very noticeably. No longer was any attempt made to follow the layout of the Carolingian book. Correspondingly, Artist F did not try to reproduce the ninth-century illustrations either in style or in iconography: instead he created his own picture for each psalm, influenced by and incorporating certain motifs from the Carolingian book. The quality of his work was uneven, and, what is of greater relevance in this context, his approach to the task of illustrating the text varied, as a few examples will demonstrate. On the one hand, we can point to instances where F worked independently in a literal mode. The picture which Artist F offered for Psalm 126201 adheres in general conception to the corresponding image in the Utrecht Psalter whilst de­ parting markedly from it in detail. Both the Carolingian and the Anglo-Saxon pictures literally illustrate the first verse of the psalm (‘Unless the Lord builds the house, those who build it labour in vain’) but they do so in different ways— attesting to Artist F ’s independent ability in this respect. In the Utrecht Psalter, God, holding a plumb-line and a trowel, is shown as a builder. In the Harley Psalter, the idea is far more dramatically rendered by the inclusion of a group of men who are shown engaged in building a series of temple-like structures; God, divested of his tools, blesses them from the heavens. (At the same time, it should be pointed out that whereas other phrases are depicted by various elements in the more complex Utrecht Psalter picture, this is the one and only verse that the Anglo-Saxon artist illustrated). The illustration to Psalm 136 provides a comple­ mentary case (PI. 12).202 Artist F ’s picture, which is much simpler than that in the Utrecht Psalter, was clearly extracted from the Carolingian image; nevertheless there is a detail within it which indicates that the artist was prepared to make a fresh literal response to the text that was in front of him. The start of Psalm 136 in Harley 603 reads: ‘By the rivers of Babylon there we sat and wept when we remembered Sion. On the willows. . . we hung up our organs’ (vv. 1-2). But hanging in the trees we see an organ and a harp— a motif with an interesting pedigree. The artists of the Utrecht Psalter referred both to a Gallicanum and a Hebraicum version of the Psalter text when creating their scenes; and although the book in which they were working had the Gallicanum text, they actually illustrated the Hebraicum on a few occasions when it lent itself more easily to visualization than did the Gallicanum.203 Now the Hebraicum more reasonably hung citharae in 200 That he worked later is suggested by the fact that he first appears on fo. 58' (Ps. 102), an unlikely place to start the project; and because his illustrations are related to Utrecht more loosely. That not much time can have elapsed between the work o f Artists A -D and his work is suggested by the fact that the text o f his section was written by the same scribe and, more significant, the rubric and psalm initials have been added (occasionally over the drawings) by the same hands that were responsible for them in the first section. 201 Fo. 66v (upper): Temple, Anglo-Saxon Manuscripts, ill. 207; Ohlgren, Anglo-Saxon Textual Illus­ tration, 234. 202 Fo. 701: Carver, ‘Contemporary Artefacts’, pi. 54c; Ohlgren, Anglo-Saxon Textual Illustration, 234. 203 See D. Panofsky, ‘Textual Basis’; further on the different versions see n. 30 above.


The Illustration o f Texts

the trees, and this was what the artist depicted. The text of Harley 603 is the Romanum which in this particular agrees with the Gallicanum. Artist F copied a cithara from the illustration in the Utrecht Psalter, but then added an organ in response to the text that was actually in the book that he was decorating. On the other hand, in contrast to these examples which indicate fresh, if highly selective, responses to the words of the text, occasionally we find more allusive elements directly juxtaposed with literal illustration. The picture for Psalm 127 is perhaps the most interesting to consider in this context.204 It consists of three distinct elements. The scene of the king and queen at table with three children serving them, within a building which is flanked by vines, illustrates verses 2-3: ‘You shall eat the labours of your hands . .. your wife as a fruitful vine on the sides of your house. Your children as olive plants round about your table’, and it was inspired by the Utrecht Psalter. Artist F modified the motif he had inherited in a number of respects: in particular he reduced the mass of vines and included a building (not present in the Carolingian drawing) which tailored the image more closely to the Psalmist’s phrase, ‘a fruitful vine on the sides of your house’. The rest of the illustration was subjected to altogether more dramatic changes. The other two elements which F supplied are a depiction of a king giving money to a group of paupers, and a scene of monastic foot-washing, the former set in the open, the latter within a building; neither has any parallel in the illustration in the Utrecht Psalter, nor an immediately apparent point of reference in the text of the psalm. In point of fact, far from being literal illustrations of the text, these motifs seem to have been inspired by St Augustine’s commentary on the psalm, and they visualize his words in terms of tenth- and eleventh-century practices. Augustine interpreted the children who are mentioned in the psalm as a reference to the giving of alms.205 This evidently made the artist think first of the Royal Maundy, the most conspicuous contemporary ceremony of alms-giving, and sec­ ondly of the foot-washing which was an important part of monastic ritual for Maundy Thursday. Practised from at least the seventh century, such foot-washing was certainly current in late Anglo-Saxon England, and it is prescribed in the Regularis concordia.m Now this rite was not simply associated with alms-giving through its connection with Maundy Thursday: on the contrary, as a profound form of serving others, done in direct imitation of Christ, it was itself a particularly potent type of charity. The point is underlined by the Vita Oswaldif1 At the climax of the Life, the dying Archbishop Oswald’s final act is to wash and dry the feet of the poor which he does not only with cloths but also with his hair. We are told that, whilst doing this, he thought on Christ’s dictum, ‘He who would be my ‘), 24b) was emphasized by the opportunity to contemplate broadly comparable images in which these details were omitted and others were present. When attempting to understand the impact of tenth- and eleventh-century images on those who beheld them it is important to 76 BL Stowe 944, fo. 6: Temple, Anglo-Saxon Manuscripts, ill. 244; col. pi.: Roesdahl et al.. The Vikings in England, 153. 77789 Fos. 6v- 7 : Temple, Anglo-Saxon Manuscripts, ills. 247-8. 78 Talbot-Rice, English A rt, 871-1100, pi. 11a; Taylor and Taylor, ‘Architectural Sculpture’, 13; Stoll, Architecture and Sculpture in Early Britain, pi. 84 (the best reproduction); and Coatsworth, ‘Late Pre-Conquest Sculptures’, 173-5 and pi. 3. 79 Green and Green, Saxon Architecture and Sculpture in Hampshire, 41-3; Taylor and Taylor, ‘Architectural Sculpture’, 12; Stoll and Roubier, Architecture and Sculpture tn Early Britain, pi. 62 (the best reproduction); and Coatsworth, ‘Late Pre-Conquest Sculpture’, 168-9, with pi. lb. 80 See further Schiller, Iconography o f Christian A rt, ii. 88-149; and Raw, Crucifixion Iconography, passim-, also O ’Reilly, ‘Rough-Hewn Cross’.


Repetition o f M o tif and Image

recognize the role of verbal expounding, and comment has been made on this here and elsewhere. Familiarity with other broadly comparable yet individually distinc­ tive versions of the same subject was no less important since this helped to accen­ tuate visually the iconographically significant, unique features of each. Because this aspect has received less attention, it is worth stressing it here. Certain renderings of the slumped figure of Christ crucified or of the Deity enthroned were particu­ larly powerful in their own right; yet it was arguably in relation to, and by comparison with, alternative renderings of the same general subject that they evoked the fullest response from the beholder. Thirdly, the presence of multiple depictions of a given subject enabled various aspects of it to be explored sequentially. It is extremely unlikely that the numerous crucifixes at Ely and Glastonbury had the same iconography. On the contrary it is highly probable that they explored and stressed different aspects of the subject and its spiritual implications. We are familiar with the juxtaposition of entirely differ­ ent though related subjects in or on a single work of art, and are sensitive to the role of their interaction in creating imagery that was broader-ranging and more comprehensive than the sum of its parts. One thinks, for instance, of the Alcester tau cross (PI. 28) and Judith of Flanders’ book-cover (PI. 4) whereon we find the Crucifixion juxtaposed with Christ in glory, and Christ triumphant over the beasts respectively. Much the same principal could apply to the juxtaposition of different iconographies of the same general subject. We mentioned above that there were four images of the Lord enthroned in the Boulogne 11 Gospels, and these demonstrate the point. Each of the images takes a different form and, considered in relation to its particular context in the book, has different nuances of meaning. The first, which is situated above Canons V and VI in the canon tables,81 shows the haloed Lord, enthroned in a mandorla, seated on an arc with his legs swathed in cloud. He is flanked by four figures, the two outermost holding a book, the innermost couple grasping a scroll. He blesses with both hands, apparently gesturing towards the evangelists beside him. This depic­ tion underlines how the four evangelists collectively witness to Christ’s life and significance, while simultaneously demonstrating the divine authority of the gospel writings. The next, a full-page depiction that occurs before Matthew’s gospel, thus immediately prefacing all four gospel accounts,82 is the image of the cosmic Christ which we discussed in Chapter 2 in connection with its use of inscriptions. Here the Lord is the only figure, he has a cross-nimbed halo, a golden headband, holds a book, and sits amid the sun, moon, and stars with his feet on the earth; inscrip­ tions further underline that he is the Lord of all creation. The third image, which is situated above the initial to Mark’s gospel,83 shows the Lord blessing and holding a book, seated in a mandorla which is upheld by two angels. Below him at his right side, beside the beginning of the gospel text which they complement, are 81 Fo. 7' (Ohlgren, Anglo-Saxon Textual Illustration, 313). 82 Fo. 10r (ibid. 318). 81 Fo. 56' (ibid. 325).

Repetition o f M o tif and Image


depictions of Isaiah and Mark with two long scrolls. Isaiah’s scroll is inscribed with his prophecy, ‘Uox clamantis in deserto: parate uiam domini’ (Is. 40: 3); Mark’s with his account of the fulfilment of it by John the Baptist (Mark 1: 3). The connection between prophecy and fulfilment is underlined visually by the fact that Mark grasps Isaiah’s scroll. Together the two scrolls lead the eye up to the Lord who is being heralded. Here the depiction of the Lord is part of an elaborate illustration of the start of the text, emphasizing its typology and Christological significance. Fourth and finally, the Lord appears surrounded by stars in a mandorla, flanked by angels, actually on the initial ‘I’ of John’s gospel (PI. 8).84 We discussed this remarkable conflation of word and image, illustrating the start of John’s text, in Chapter 1. Each of these four images is different in context, appearance, mes­ sage, and function, and each highlights distinct aspects of the Lord’s nature. Consequently, as a series they more fully expound the character of the Divinity and the direct relationship between him and the gospel texts. It was only by means of such repetition (or better: such themes and variations) that the mysterious nature of a triune Divinity and its manifold implications could adequately be represented.85 The numerous, variant depictions of these key sub­ jects can thus be seen to correspond to the other structures of religious life which during the course of the liturgical year sequentially explored, honoured, and cel­ ebrated different aspects of the Creator, his works, and his relationship with man. The complex of beliefs that are codified most succinctly in the verses of the Creed and the Te Deum received correspondingly sequential visual expression through a multiplicity of images. This leads on to our fourth, final, and perhaps most important point, namely that repetition of imagery encouraged reaffirmation of faith. Although the precise circumstances no doubt varied considerably from place to place, we can be confi­ dent that there were multiple depictions of the Deity in most if not all late AngloSaxon churches and monasteries. In some places a number of these images may have been included in one programme of decoration or on a single artefact; in general they will have ranged across the whole spectrum of artistic media. One will have represented one facet of the Deity, another a different aspect. While several may have resulted from a single patron or initiative, many others will have been acquired over the years from a variety of sources. They will have been located in different places and used in different contexts. We cannot now recover the details and catalogue the complete holdings of even one Anglo-Saxon foundation in this respect. Nevertheless, it is quite clear that depictions of the Deity were very numerous, and that consequently in a given ecclesiastical environment the faith­ ful soul will have encountered copious images of his Lord. God was around the walls, on and above the altars, in the books, and on the vestments, not to mention being carried in procession. He was literally everywhere. The significance of this M Fo. 107v (ibid. 330). 85 On the challenge this represented see, inter alia, Grabar, Christian Iconography, esp. ch. 5; and Henderson, 'Narrative Illustration and Theological Exposition’, esp. 31-5.


Repetition o f M o tif and Image

phenomenon can hardly be overstated. The physical presence of numerous such images impressed upon the beholder the fact that he was in the house of God. They reaffirmed the fundamental tenets of the Christian faith at every turn. They continuously underlined the multifaceted nature of the Deity. They reminded man that his every action was done in the sight of God. The risen Christ assured his disciples, ‘Behold, I am with you always until the end of time’ (Matt. 28: 20); St Benedict, in his chapter on ‘The Instruments of Good Works’, included the injunction: ‘To know for certain that God sees you everywhere’;86 and countless readings, prayers, and devotions declared God to be ‘omnipotens’ and ‘sempiternus’. Belief in the eternal omnipotence and omnipresence of the Deity was made all the more compelling by the ubiquitous depictions of him. God really was everywhere: repetition of his image fostered continual reaffirmation of faith. To sum up: the recurrence of decorative motifs and certain figurai subjects, although a commonplace phenomenon, is nevertheless worthy of attention. The use of similar decorative styles on a wide variety of works of art was aesthetically pleasing, and helped to create a visually attractive, co-ordinated environment. The repetition of images which shared basic elements of design although often quite distinct in detail, and the display of multiple interpretations of the same subject within one ecclesiastical environment, had a number of consequences. It enhanced the authenticity and resonance of all the works in question; it emphasized the individual features of each, thus stressing their particular significance; it facilitated the exploration and representation of a complex and awesome Deity; and, most important, it made him physically omnipresent. Numerous devotions, prayers, and homilies, and indeed the regular life as a whole, were designed to make man evermindful of his God and to encourage him to reflect upon his relationship with him. The repetition of imagery was one of the many devices and structures that helped man achieve this. w ‘. . . in omni loco Deum se respicere pro certo scire’, Rule o f S t Benedict, ed. Chamberlain, 25.

5 Pictorial Narrative A seminal work published in the early 1960s focused attention on the twelfth century as a formative period in the development of pictorial narrative in Eng­ land.1 The author, Otto Pächt, raised the profile of narrative depiction (according to his definition)2 and offered numerous insights into the field as a whole, while in particular he explored and placed in perspective the work of the Master of the St Albans Psalter.3 This emphasis shed invaluable light on English manuscript art of the twelfth century and on the work of an outstanding individual. However, its side-effect was to minimize the importance of other modes of pictorial narrative and to divert attention from the role of pictorial narrative in earlier English art.4 Pächt was principally concerned with cycles rather than individual images; the themes on which he concentrated were the ways in which artists incorporated the element of time into their images and the influence of liturgical drama on art;5 and he was dismissive of alternative approaches and other issues.6 But the alternatives are equally deserving of attention. Some of the forms of narrative art essayed by late Anglo-Saxon artists were less sophisticated than those of the St Albans Psalter. They were essentially inherited rather than newly developed and they were not refined in response to contemporary liturgical drama, but they were nevertheless an important part of their visual language. 1 Pacht, Pictorial Narrative. 2 For recent, more flexible approaches see inter alia Hahn, ‘Picturing the T ext’, and Passio Kiliani, Commentary, 4 -1 5 ; and Kupfer, Romanesque Wall Painting in Central France, 59-150. 3 Complementing his treatment o f the pictorial cycle in Pacht et a i, The Saint Albans Psalter. 4 His discussion o f Anglo-Saxon material (pp. 5-11) is very summary, consisting mainly o f a discussion o f one image in Junius 11 and some general observations on the Bayeux Tapestry. The crucially important prefatory cycle in Cott. Tib. C. VI is not discussed at all, being mentioned only in a footnote on p. 21. 5 Pacht, Pictorial Narrative, ch. 3. But note Schapiro, ‘Disappearing Christ’, 281-2, where it is suggested that the illustration o f Christ washing his disciples’ feet in the Tiberius Psalter was inspired by the Easter ceremony in which the abbot washed the feet o f his monks or the poor; also the (highly speculative) discussion o f MellinkofT, Homed Moses, ch. 4. 6 e.g. his comments on Greek art (Pictorial Narrative, 2-3): ‘the Greeks contrived to telescope a whole story into a single moment by focusing on the one point on which events revolved, a point situated at some important juncture o f past and future. But the Greek principle o f selecting the socalled pregnant or climactic moment reveals itself on closer scrutiny merely as an ingenious expedient for bypassing the issue, for it cannot be seriously maintained that the Greek device offered a truly genuine solution.’ In fact the ‘pregnant or climactic moment’ was extensively and effectively used in Christian iconography, as we shall see.


Pictorial Narrative

It should be stressed at once that much Anglo-Saxon art was not intended to be narrative. Subjects such as the Deity enthroned (cover, Pis. 4, 5a, 8, 18), the Virgin and Child, and even the Crucifixion (Pis. 4, 5b, 15£, 24, 28a) were designed to present an enduring vision of eternity rather than a particular happening. The context in which such cosmic subjects were set was spiritual not temporal, their imagery had to evoke eternal truths, not sequential narrative, and the task of the artist depicting them was to represent and reaffirm their nature as perpetually unchanging. The point applies also to the divine element in those pictures, such as the frontispiece to the New Minster Charter and the image of Saint Benedict and the monks in the Arundel 155 Psalter,7 which show the interaction of earthly and spiritual figures. In the New Minster Charter (PI. 2a) there is a clear contrast between King Edgar who is shown actively performing a particular deed— namely presenting the charter to Christ and adoring him—and the company of heaven who are frozen in an eternal tableau. In Arundel 155 the one heavenly figure, Saint Benedict, has a monumentality, stillness, and formality which sets him apart from the earthly community of monks who move and perform transient deeds around him. Christ, the company of heaven, and Saint Benedict are forever unchanging. Eternity and spirituality are conveyed in visual terms by formality and a minimum of specific action; and little action is needed because the supernatural figures are themselves the subject-matter. In addition, the distinction between the way the heavenly and earthly figures are portrayed draws a sharp, telling contrast between the eternal and the mortal, and stresses the dominant role of the divine in the affairs of men. The depiction of numerous other subjects, by contrast, could involve story­ telling. The events of the Pentateuch and the life of Christ were recorded as a series of stories. Anglo-Saxon artists illustrating such subjects were therefore faced with the task of depicting biblical narrative. Yet far more than the rendering of the recorded events in visual form was involved in this. Just as Ælfric was concerned in his homilies to explain the religious truths contained within biblical narrative, so it was highly desirable for the pictorial version of a story to indicate its religious significance. Indeed it was the spiritual meta-message of the event or events and not their mere physical occurrence that mattered. (Sometimes, when the imagery can be seen to reflect certain earthly preoccupations of its creators or patron, the meta-message might better be described as religio-political.) The philosophy of viewing earthly events as a medium for divine truths permeates biblical commen­ tary and is readily apparent in hagiography and monastic chronicles:8 it was equally important in determining the nature of visual renderings of scenes. (And just as chronicles reflect the contemporary concerns of their creators, so too could pictorial 7 BL M SS Cott. Vesp. A. viii, fo. 2'; and Arun. 155, fo. 133' (col. repro.: Backhouse et at.. Golden Age, pis. 4 and 18. 11 For a concise statement o f the nature and preoccupations o f monastic writings see Clanchy, From Memory to Wntten Record, 146-9. For general comments on hagiography see Heffeman, Sacred Bio­ graphy, esp. 3-37.

Pictorial Narrative


narrative, as recent studies of illustrated saints’ lives have amply demonstrated.’) The prefatory cycle in the Tiberius Psalter demonstrates the point (Pis. 13/», Mb). Although largely composed of events from the lives of David and Christ, this is not simply an attempt at pictorial biography. On the contrary, as was noted in Chapter 1, the sequence of images was chosen and structured to form a typological and liturgical cycle; and the collective message of the scenes is as much about the fulfilment of the Old Law, the nature of rule, power, and Christian kingship, and the conquest of evil,910 as about the earthly doings of David and Christ. But then the ‘true’ significance of the deeds and sayings of Christ, his prototype, and his saints was eternal and spiritual, not temporal; ideally, art reflected this. One response to the challenge of representing spiritual meaning was to replace narrative with (or, perhaps better, sublimate it into) an icon—a single, hieratic image that indicated the spirituality and could be taken to allude to the recorded sequence of events. Many of the images in the Benedictional of St Æthelwold, for instance, are predominantly of this nature. Thus we are not shown Saint Etheldreda resisting the advances of her second husband or founding her monas­ tery:11 rather she is depicted as a stately standing figure holding a book and a foliate rod (PI. 23a).12 The image asserts her sanctity, while the details of the iconography and the accompanying inscription (‘Imago sanctae Æþeldryþae abbissae et perpetuae uirginis’) allude to the key elements behind it, namely her status as virgin and abbess. On the other hand, certain images created by Anglo-Saxon artists did tell a story, or some part of it, by visual means and it is these that are of particular concern to us here. In analysing this material we should be careful not to ask misdirected questions or to apply false standards. We should focus our attention on the ways in which visual narrative was created in the surviving images produced by tenth- and eleventh-century artists without being preoccupied with the origin­ ality of a given response or solution. Furthermore, although the purpose of this section is to explore the mechanics of Anglo-Saxon visual language, it would be misleading to divorce this part of that language from its context—a context of words, speech, and familiarity with the accounts that were depicted. It was not the role of pictorial narrative single-handedly to explain events that were otherwise unknown.13 On the contrary, its function was to accompany written texts, to recount familiar stories, to fulfil informed expectations, and to activate easily triggered responses; and it did this with the help of writing, inscriptions, or an 9 For recent studies o f Saints’ Vitae see e.g. Abou-El-Haj, ‘Bury St Edmunds Abbey between 1070 and 1124’, and Carrasco, ‘Spirituality in Context’. For discussion o f the influence o f contemporary thought and politics on imagery in the Benedictional o f St Æthelwold see Deshman, ‘Christus Rex et Magi Reges’. 10 The last o f these themes is expounded by Openshaw, ‘Battle between Christ and Satan’. 11 Compare Ælfric, Aelfric's Lives o f Saints, ed. Skeat, i. 4 3 2 -4 0 (derived from Bede, Historia ecclesiastica, 4. 19). 12 Add. 49598, fo. 90': F. Wormald, Benedictional o f S t Ethelwold, pi. 6. 13 Cf. Heffernan, Sacred Biography, 18-22, on the relationship between the writer o f a saint’s Life and his audience.


Pictorial Narrative

explicator. It is crucial to keep all this in mind as we consider the various purely visual devices that are the subject of our present enquiry. In contrast to the more leisurely recounting of events in a written source, the Anglo-Saxon artist had to encapsulate both the essence of a narrative and its significance in a single image. There were two essential aspects to doing this successfully. The first was selecting a suitable narrative moment to illustrate; the second was conveying the action and the implication of that moment in pictorial form. As action happens through time, while the figures in an image are frozen in one position, not everything that a text recorded as happening could easily be illustrated: to be effective a depicted narrative moment had to show something that could be seen to happen.14 However, that one moment could also imply a number of others. A key incident could evoke both what had happened beforehand and what was going to occur afterwards, and it might thus effectively represent an entire story. The selection of the moment or moments to depict was thus crucially important, and it should be stressed that this (which may be termed the ‘research side’ of Christian pictorial narrative) had been studiously undertaken in late Anti­ quity15 and the most acceptable solutions had been worked out. In most cases there were models to show late Anglo-Saxon artists which moments best lent themselves to depiction and how to approach the task of depicting them. Before turning to examine the principal techniques used in Anglo-Saxon nar­ rative imagery, we should draw attention to one potentially important device that was largely ignored: the face. Facial features were generally of minor importance for creating and supporting pictorial narrative in Anglo-Saxon art. When render­ ing a face, most Anglo-Saxon artists preferred to repeat one of a strictly limited number of favourite formulas than to invent different physiognomies. Thus in the prefatory cycle to the Tiberius Psalter we see three head-types— profile with prominent nose and chin, three-quarters bearded, three-quarters beardless (PI. 17b)— used over and over again. Two basic head-types, one bearded, the other beardless, are repeated many times in the Benedictional of St Æthelwold. The former is at least dignified (the latter is banal and ugly) but with neither does the artist attempt even the most simple variation of expression. Thus the same inex­ pressive, beardless face that was used for the figure of Thomas in the Appearance to Thomas was given to St Etheldreda (PI. 23a), to the angel at the sepulchre, to Christ entering Jerusalem, to the bishop depicted using the Benedictional (PI. 13a), to both the maid and the Virgin in the Nativity, and to the Virgin, St Peter, a second apostle, and the four angels in the Ascension.16 Such monotony and standardization was acceptable because, with a few exceptions, faces were rarely 14 For discussion o f some o f the general issues involved sec Gombrich, ‘Moment and Movement’, and ‘Action and Expression'; and Arnheim, Art and Visual Perception, ch. 8. 15 For an introduction to methods of pictorial narrative in Antiquity and late-Antiquity see Weitzmann, Roll and Codex, passim; Brilliant, Visual Narratives, passim; and Pollitt, Art in the Hellenistic Age, ch. 9. For the methods of early Christian art see Weitzmann (ed.), Age o f Spirituality, 396-555. 16 See the facsimile: Warner and Wilson, Benedictional o f St Æthelwold.

Pictorial Narrative


used to express very much. Context, inscriptions, and iconography, not facial features, establish identity; only occasionally in late Anglo-Saxon art do we find the intense, engaging gazes that distinguish early Christian figures and Byzantine icons and mosaics;17 and in pictorial narrative communication between figures, their mood, and sometimes even the direction of their attention is primarily in­ dicated by means of gestures. Speaking figures point at each other but have their mouths closed; adoring figures raise their arms, proffer their gifts or bow their heads with impassive faces. The contribution of heads to a picture was often limited to facing in the right direction. Even when this was effected in a striking manner, as on certain folios of the Bury Psalter (where characters crane their necks to look at figures above them: PI. 15a)1819and in the Trinitarian frontispiece to the Harley 603 Psalter (where the heads of the Father and Son, along with that of the Dove, are pressed closely together expressing their unity),1,1 the actual facial fea­ tures remain inexpressive, if not inappropriate (judged that is by the irrelevant standards of physical realism). One notable exception that should be mentioned is the image depicting the ‘fate of the soul’ in the New Minster Liber Uitae (PI. lb ).20 In this composition a complex narrative is created and the responses of the viewer are guided not only by gesture and pose, but also by appropriate and communicative facial expressions. St Peter, at the top of the recto, is smiling and he is clearly jubilant to be able to greet a throng of the blessed (whose faces show them to be happy and expectant) and he invites them to enter the gates of heaven. St Peter appears again in the register below, and this time his face expresses the determination with which he is battling for the salvation of a soul. The devil he is worsting looks appropriately aggrieved, while the distress and shame of the sinful souls to the right is under­ lined by their bowed heads, downcast faces, and wringing hands. In the lowest register it is evident that the demon is not just afflicting three souls, but that he is doing so with inhuman frenzy. This is a rare example of facial characterization being used to enhance the narrative thrust of an image, and to emphasize its spiritual message— which is basically that it is wonderful to enter Heaven but terrible to be condemned to Hell. The composition on folio T of the New Minster Liber Uitae consists of three distinct scenes that are intimately related to each other thematically. Their juxta­ position heightens the drama of the struggle for the soul that is being enacted in the middle register. We see that on the outcome of this conflict between Saint Peter and a devil depends whether he will join the blessed who are adoring Christ 17 See Demus, Byzantine Mosaic Decoration, 5-10; van der Meer, Early Christian A rt, 102-7; and Weitzmann, The Icon, passim. 18 e.g. fos. 29r, 30r, 62r, 62v, 68v, 73v, and 80r (Ohlgren, Anglo-Saxon Textual Illustration, 256-7, 265 — 6, 269, 274, and 277). 19 The composition would appear to derive from an image of the Virgin and Child, though it is difficult to agree with the conclusions Hasler, ‘Ältesten Kopie des Utrecht-Psalters’ draws from this. 20 BL MS Stowe 944, fos. 6v- 7 : Temple, Anglo-Saxon Manuscripts, ills. 247-8; facsimile: Schools o f Illumination, i, pi. 13.


Pictorial Narrative

in Heaven or be cast down and locked into the torments in Hell. And the physical arrangement itself underscores the moral: it is up to Heaven but down to Hell. On certain pages in other manuscripts, in the Bayeux Tapestry and, we may presume, on many lost wall-paintings and hangings and possibly also carved friezes21 and doors22 similar juxtapositions add or added appreciably to the narrative force of each individual scene involved. They create a visible progression, a progression which may be thematic, temporal, or both. The scenes in question can be arranged in one continuous setting, in different registers or areas within a single space or page, or even on separate but contiguous areas or pages. An example which is set on one space divided into two registers can be seen on page 31 of the Junius 11 manuscript of Old English poetry.23 Here a drawing of Eve giving the fruit to Adam appears directly above a separately framed depiction of the sinful Adam and Eve prostrating themselves. The be­ holder advances rapidly through the events of the story as his eye moves from the upper picture to the lower. Furthermore, the moment depicted in the first image provides the explanation for the action shown in the second. The juxtaposition thus enlarges by visual means the temporal context of the events depicted in the separate scenes, and emphasizes the causal relationship between them. A more fluid example of the same phenomenon appears eleven pages earlier in the same manuscript, this time set within a single, continuous space. In the bottom half of the picture area on page 20 we see the Devil, shackled in Hell, sending out an agent, while in the top half are depicted the temptation of Eve by the serpent, and Adam and Eve pointing at a tree.2425In this case the flow of narrative from one scene to the next is enhanced not only by their close proximity, but also by the compositional structure of the picture as a whole. The diagonal angle of the Devil points the eye towards the figure of his agent, by whom it is led upwards to the tree with the serpent. The serpent projects to the right, leading the eye towards Eve, whose hand points it on in turn to the final scene of Adam and Eve. Here, as in the first example, both an elapse of time and the element of causality are built into the imagery. There are several other cases in the extant material where sequential narrative moments are extensively juxtaposed. On page 49 in Junius 11, for instance, five scenes from the story of Cain and Abel appear within one frame;23 on folio 38r of the Old English Hexateuch there are three scenes tracing the events of the sacrifice of Isaac within one frame (PI. 29);26 on folio 42r in the same manuscript Jacob’s 21 The surviving evidence for narrative friezes in late Anglo-Saxon England is limited to the frag­ ment from the Old Minster, Winchester, for which see Backhouse et a i, Golden Age, cat. 140. 22 If wooden doors were produced like the 5th-c. example at Santa Sabina, Rome (Volbach, Early Christian A rt, pis. 103-5), or bronze ones like those cast at Hildesheim in the early 11th c. (Tschan, Bemvpard o f Hildesheim, ii. 141-20 with iii, pis. 115-36; Leisinger, Romanesque Bronzes, pis. 12-35; Brandt and Eggcbrccht (eds.), Bemward von Hildesheim, ii, cat. 7. 33). 23 Facsmilc: Gollancz (ed.), Caedmon Manuscript; Ohlgren, Anglo-Saxon Textual Illustration, 541. 24 Ibid. 538. For discussion o f the relationship between poetic text and imagery in this section o f the book see Ch. 1 above. 25 Ibid. 550. 26 Col. repro.: Campbell (ed.), The Anglo-Saxons, ill. 191.

Pictorial Narrative


deception of Isaac is narrated in four scenes in one picture space; while three stages in the release of Peter are depicted in three separate registers on folio 22r of the Caligula Troper.27 Turning to other media, mention may be made of the oval box which is carved with a cycle of scenes showing monks around an altar,28 and of the Bayeux Tapestry, a strip of narrative art 68 metres long.29 Differences in the arrangement or mode of presentation do subtly effect the nature of the narrative: formal divisions lead the viewer to perceive a series of distinct moments; a continu­ ous space encourages the perception of ongoing time. The designer of the Bayeux Tapestry used both techniques (trees and buildings providing dividers as required) according to whether he wished to convey temporal and spatial continuum or disjunction. Yet even in works like these where the meaning of each scene is enhanced by its sequential or thematic relationship with its immediate neighbours, the essence of the narrative remains the action depicted in the individual figurai unit. The most successful individual scene encapsulates a narrative in itself. It does this by em­ bracing more than one moment of a story. The narrative iconographies that AngloSaxon artists inherited were generally designed to evoke a number of moments. There were three principal means by which this was achieved. The first was simply to depict the protagonist more than once. Thus in the picture on folio 36r of the Old English Hexateuch which illustrates Hagar being directed to a well by an angel, we see Hagar twice: first receiving the message from the angel; then going to the well. Again, on folio 29v, the episode of Abraham seeing three men, then prostrating himself before them, is conveyed by two Abrahams—one on the left looking out of his tent towards three men on the far right; the second prostrate before them in the middle of the picture. God’s cursing of the serpent in Junius 11 provides another example.30 The illustration includes a double depiction of the reptile—it is shown both standing upright on its tail facing God, and writhing on the ground with its head away from him, having been cast down by his power. The second method was to represent the transition between one instant and the next. For an example of this we may turn to the final episode of the story of the sacrifice of Isaac illustrated on folio 38r of the Old English Hexateuch (PI. 29).31 The moment depicted is that when the angel stops Abraham from slaying his son. As both the dramatic climax and the moral focus of the episode this was a natural point for an artist to choose to represent. However it should also be noted that this 27 Col. repro.: Backhouse et al., Golden Age, pi. 21. One may contrast this with the two-scene version o f the event, as in the mosaics of the Cappella Palatina, Palermo, and Monreale (Borsook, Messages in Mosaic, ills. 49-51, and 74 and 76). 28 V. & A. 268-1867: Longhurst, English Ivories, 31 and 94 with pi. 34; and Catalogue o f Carvings in Ivory, 90 with pi. 70; Beckwith, Ivory Carvings, cat. 19, ills. 25 and 42-5; and Backhouse et al., Golden Age, cat. 116. T he story depicted has yet to be identified satisfactorily. 29 For discussions which include important comments on the narrative mode o f the Bayeux Tapestry and comparisons with various written texts see Pacht, Pictorial Narrative, 9-11; Dodwell, ‘Bayeux Tapestry’; Henderson, Early Medieval, 168-78; Brooks and Walker, ‘Authority and Interpretation’; S. A. Brown, ‘Bayeux Tapestry’; and Brilliant, ‘Bayeux Tapestry’. 30 Junius 11, p. 41: Ohlgren, Anglo-Saxon Textual Illustration, 545. 31 Tem ple, Anglo-Saxon Manuscripts, ill. 270; col. repro.: Campbell (ed.), The Anglo-Saxons, ill. 191.


Pictorial Narrative

moment offered the additional narrative benefit of being a turning point. By focusing on a turning point in a story, one image could illustrate the transition between two immediately consecutive moments: in this case the attempted slaying and the reprieve. Attention should also be directed to the disposition of the figure of Abraham in this scene. Although he holds down Isaac with one hand and raises aloft the exaggeratedly large sword with the other, his head is turned away from his son, and he looks towards the angel on the left. His hands indicate that he is in the act of slaying Isaac; his head shows him receiving the crucial command that prevents the impending tragedy. The efficiency of the iconography in embracing two con­ secutive moments is enhanced by what may be termed the duality of Abraham’s pose—the fact that he is partly engaged in one action, and partly in the moment that immediately succeeds it. Similarly effective dual poses are used to hold to­ gether and to ease the transition between consecutive moments illustrated in single scenes elsewhere in this manuscript. On folio 7r, for example, the consecutive moments of Eve’s eating the apple and then encouraging Adam to do likewise are synthesized in the one figure of Eve who is shown taking fruit from the tree to the left of the composition with one hand, whilst turning her head towards Adam who is eating the fruit on the right.32 And, more subtly, on folio 36r the effectiveness of the illustration of Abraham’s dismissal of Hagar and Ishmael is enhanced by showing simultaneously both the verbal exchange and the subsequent separation. Although Abraham’s head looks at Hagar and Ishmael and he points at them with one hand, his body is shown walking away from them. They are likewise depicted looking at him (receiving the command) whilst walking away from him (having received the command). The same technique is used in the Bayeux Tapestry where it also serves to link one scene to the next. Other scenes embraced an extended temporal span without resorting to dual poses. The representation of an exchange of dialogue lay at the heart of many such depicted narrative moments. This was the third method. In certain cases, as we saw in a previous chapter, relevant words were actually introduced into the picture itself through the device of a scroll. However the mere implication of dialogue effectively enlarged the time continuum that could be contained in a single image. It was rare before the twelfth century for an artist to try to evoke the specific topic of a dialogue by visual means,33 and the ‘deficiency’ of the pictorial arts in this respect was commented on by the author of the Libri Carolini.34 Nevertheless it may be assumed that the tenth- or eleventh-century beholder to whom the phrases of the relevant scriptural narrative were very familiar would readily ‘supply’ the 32 T his is generally represented as two separate incidents with two figures o f Eve in the Cotton Genesis tradition, as for instance in the Moutier Grandval Bible and the Salerno ivories: see Weitzmann and Kessler, Cotton Genesis, ills. 45-9. 33 As pointed out by Pacht, Pictorial Narrative, 54. 34 Carolt Magni Capitulare de Imaginibus, 3. 13, ed. Bastgen, 153. For a recent discussion o f the treatment o f some aspects o f speech in early medieval art see Hahn, ‘Purification, Sacred Action, and the Vision o f G od’.

Pictorial Narrative


appropriate words for verbal exchanges that were clearly indicated by hand pos­ tures. Most beholders of the image of God gesturing towards the serpent in Junius 11, for example, would surely have recalled the gist or even the very words of the divine curse on that beast.*5 The same is likely to have been true of the depiction of God addressing Adam and Eve which appears immediately below.3536 In these two pictures only one character has implicit speech; but there are other cases where a more complex dialogue is to be understood. In the images of the three women at the sepulchre in the Rouen Benedictional and the Rouen Sacramentary,37 for instance, the angel’s gesture relates to a particular and wellknown verbal exchange and hence creates a narrative continuum. The uplifted, blessing hand shows that the angel is addressing the women; but more than that it specifically evokes the women’s original enquiry concerning the whereabouts of Christ, the angel’s triumphant reply that he is risen, and his command to the women to go and tell the good news: a narrative sequence. Furthermore, the monastic beholder of these images may well have expanded the implied exchange in terms of the relevant liturgical forms that were used and dramatized in the service for Easter Day:38 the trope Quem quaeritis in sepulchro, 0 Christicolae and the antiphons Uenite et uidete locum and Surrexit Dominus de sepulchro. The illus­ tration in the Benedictional of St Æthelwold of Christ’s appearance to Thomas provides another example.39 The image simply shows the apostle putting his hand into Christ’s side and Christ raising his hand in acknowledgement or blessing, but clearly it also implies and embraces Thomas’s earlier statement that he would not believe in the Resurrection unless he put his hands into the wounds,40 Christ’s subsequent instruction to him to do just that, and Christ’s final blessing of those who believe in him without such an opportunity. Unlike certain illustrations in the Old English Hexateuch which may depict single, brief incidents, these scenes were chosen and their iconographies were designed not as a glimpse of a single temporal moment, but as the focus of a cluster of moments, as a node of related events. To an age which habitually looked beyond the literal meaning of biblical texts to their exegetical and homiletic significance, undue preoccupation with mere phys­ ical happenings was a distraction from the ‘true’ meaning of the events in question. (Arguably one of the most remarkable aspects of the illustrated Hexateuch is the 35 Gen. 3: 14-15; ‘Quia fecisti hoc, maledictus es inter omnia animantia et bestias terrae: Super pectus tuum gradieris, et terram comedes cunctis diebus uitae tuae. Inimicitias ponam inter te et mulierem, et semen tuum et semen illius; Ipsa conteret caput tuum, et tu insidiaberis calcaneo eius.’ Cf. Genesis, 11. 906-17 (Anglo-Saxon Poetic Records, i, ed. Krapp, p. 30). 36 Gen. 3: 16-19; cf. Genesis II. 919-38 (Anglo-Saxon Poetic Records i, ed. Krapp, pp. 30-1). 37 Rouen, BM M SS Y. 7, fo. 21v (Temple, Anglo-Saxon Manuscripts, ill. 89) and Y. 6, fo. 72' (Pacht, Book Illumination, col. pi. 19). 38 Regularis Concordia, ed. Symons, 49-51; The Winchester Troper, ed. Frere, 17. For a general discussion of the ‘Quem queritis’ liturgical drama see Stevens, Words and Music, 330-6. 39 Fo. 56v: Temple, Anglo-Saxon Manuscripts, ill. 86; col. pi.: Backhouse, Illuminated Manuscripts, ill. 7. T his iconography can be traced back to the early 5th c. when it appears (in reverse) on the last o f the four ivory passion-scene plaques in the Brit. Mus.: Dalton, Catalogue o f Early Christian Antiquities, no. 291, pi. 6(d); Weitzmann (ed.), Age o f Spirituality, cat. 452. 40 John 21: 25.


Pictorial Narrative

number of mere happenings it represents.41) Thus in his homily for Easter Sun­ day42 Ælfric interprets the actions of the three Maries as a model for contemporary behaviour: ‘We who believe in the resurrection of Christ come assuredly to his sepulchre with precious ointment if we are filled with the breath of holy virtues and if we with the fame of good works seek our Lord. The women who brought ointment saw angels, for they see the heavenly angels who with the breath of good works yearn for the upward journey.’ He expounds details like the moving of the stone to express theological truths: ‘The angel rolled the lid from the tomb not to make way for Christ’s departure, but to manifest to men that he was risen. He who came mortal to this world, bom of the closed womb of the Virgin, he without doubt might, when he arose immortal, though in a closed tomb, depart from the world.’ And he reads ascetic, monastic values into the dialogue: ‘The angel cheered the women saying, “Be not afraid”, as if he had thus said: Let those be afraid who love not the advent of angels; let those be terrified who are beset with fleshly lusts, and have no joy in the host of angels.’ Correspondingly, in addition to creating a visual narrative, the individual Christian image had also to evoke the spirituality and ‘message’ of the actions depicted. Indeed in many cases this was its raison d ’être. The exegetical writer could quote his source passage and then proceed to ex­ pound it; the artist had to present simultaneously the temporal occurrence and its spiritual significance. This was achieved in the first instance by incorporating ‘divine’ elements in the imagery. The halo indicated the holiness of an individual character, while the presence of attendant angels (found for example in the Bap­ tism of Christ in the Benedictional of St Æthelwold,43 and, most interestingly, in the washing of the apostles’ feet in the Tiberius Psalter44), or the hand of God (as seen in the depiction of Pentecost in the Rouen Benedictional)45 served to advertize the participation of the divine in the event as a whole, and to link the temporal happening to a spiritual plane. At the same time the position, scale, and formality of the key holy figure underlined the dominant role of the divine in the happening. Supernatural characters are sometimes larger than the humans with whom they interact46 (the angel at the tomb, for instance, is appreciably bigger than the three women) and they are generally distinguished by their serenity and formality, as we noted earlier. Moreover the holy figure or agent is very often centrally placed in the visual field with the rest of the composition arranged around him. This effectively presents the divine as the focus and moving force of the events in question. 41 See the summary in Dodwell and Clemoes, Old English Hexateuch, 65. 42 Ælfric, Sermones Catholici, ed. Thorpe, i. 2 2 0 -9 , esp. 2 2 0 -4 . 43 BL Add. M S 49598, fo. 25'. Col. repro.: Grabar and Nordenfalk, Early M edieval Painting, 180. 44 BL M S Cott. Tib. C. vi, fo. 1 l v (F. Wormald, Collected Writings, i, ill. 134). T his is a most unusual feature which is not present in any other well-known versions o f the scene (cf. Schiller, Iconography o f Christian Art, ii, ills. 117-33 and 135-40). 45 Rouen, BM M S Y. 7, fo. 29' (Temple, Anglo-Saxon Manuscripts, col. frontispiece). 44 See further Ch. 6 below.

Pictorial Narrative


The gestures and poses of the protagonists could also make an important con­ tribution to elucidating the religious significance of an event. The spiritual signifi­ cance of the appearance to Thomas centred on Christ’s wounds. In the first place, as Ælfric emphasized, Thomas’s touching of the wounds was God’s confirmation to man that Christ had really risen from the dead;47 secondly, as is stressed in Judgement Day imagery, man must recognize in the wounds the fact that Christ suffered for him;48 thirdly, as the hymn for the Feast of St Thomas preserved in the Bosworth Psalter succinctly states, the wounds themselves represent man’s release from sin.49 The inherited iconography of the appearance to Thomas, which emphasizes the touching, the wound, Christ’s divine yet physical presence, and his general blessing, effectively encapsulate these ideas and could undoubtedly recall them to the mind of a receptive beholder. The spiritual significance of a scene like the Ascension, where the departure of Christ’s human body incorruptible to heaven represented the reopening of Para­ dise to mankind, would seem to present more of a challenge. Nevertheless the inherited iconography as exemplified in the Benedictional of St Æthelwold could evoke the requisite range of ideas for the informed viewer.50 The striding posture of Christ emphasizes his movement and physicality. The colour and patterning of the background of the upper area clearly distinguishes the zone to which Christ has ascended bodily from the earthly realm that he has left, the presence of the angels shows it to be Heaven, and the extended hand of God indicates the Father’s reception of the Son. The inclusion of Mary among the human company (not specified in the biblical account) serves as a visual reminder of Christ’s incorrupt humanity, while the fact that the two lowest angels point down to earth as well as up at Christ underlines the fact that his transition has great implications for the relationsip between Heaven and earth. The image of the Ascension in the Benedictional of St Æthelwold introduces another important facet of the role of gesture in pictorial narrative: the fact that one gesticulating hand could simultaneously indicate and connote a number of different things. As the individual scene could be a focus of events, so the indi­ vidual gesture could represent a node of responses and hence, by extension, could imply a spectrum of ideas. The upraised hands of the watching figures in the Benedictional of St Æthelwold express both amazement and adoration. Returning 47 Dominica Prima post Pascha (Sermones Catholici, ed. Thorpe, i. 302). 48 Vercetli Homilies, VIII, ed. Scragg, 145-6; Christ III, 11. 1029 fF. (The Anglo-Saxon Poetic Records, iii. The Exeter Book, ed. Krapp and Dobbie, 31-2, and 36-7); The Homilies oflVulfstan, ed. Bethurum, 121-2. Note also the ivory Last Judgement from North Elmham (Okasha, Handlist, cat. 97; Beckwith, Ivory Carvings, cat. 18; discussed in Ch. 2 above) with its inscription: o v o s OM S : v i [ d e ]t e m a n u s e t p(.). 49 The Canterbury Hymnal, ed. Wieland, 90. 50 Talbot-Rice, English Art, 871-1100, pi. 50b, and (much clearer) Beckwith, Ivory Carvtngs, ill. 33. Except for the hand o f God, the design is broadly comparable to that o f the 9th-c. Carolingian ivory carving, V. & A. 254-1867, for which see Longhurst, Catalogue o f Carvings in Ivory, i. 65 with pi. 44; also Beckwith, Ivory Carvings, ills. 50 and 52 (where the piece is mistakenly claimed as Anglo-Saxon). The ‘disappearing Christ’ iconography is far less suited to evoking the spiritual significance o f the event.


Pictorial Narrative

to the image of the appearance to Thomas in the same manuscript, it can be seen that the raised right hand of Christ (which exposes and draws attention to the wound in his side) expresses both communication and blessing; furthermore it gives an air of calm control to the person of Christ himself. This is complemented by his left hand which, in addition to holding the cross staff, points upwards, alluding to his risen status and to the promise of eternal life for those who believe on him.51 In the Old English Hexateuch, figures who are being addressed are not infrequently given upturned hands which serve to indicate that they are receiving the message (PI. 14)— Sarah has her hands thus on folio 22r where she is being instructed by Abraham to pretend to be his sister.52 When this conceit was used in certain contexts, such as for Hagar when she receives the directions to the well, or for Abraham as he is told that his descendants will be as numerous as the stars, the ‘receiving’ gesture also usefully conveys prayerful praise.53 The posture thus evokes the receiving of the message and the reaction to it, and also alludes to its divine origin. It is difficult to be sure that this is a deliberate device when the same gesture is insensitively used on other pages,54 but the visual effect of the examples cited is unequivocal. The gesture that can perhaps most commonly be seen performing a multiple role in this way is that of blessing. When this was used as an indication of pointing, addressing, commanding or healing it could impart a wholly appropriate mood of liturgical solemnity to the scene depicted, reminding the beholder of the spiritual dimension to the narrative moment before him. Thus the angel addressing the three women at the sepulchre, for instance, does so with a blessing hand.55 Inter­ estingly in the Bayeux Tapestry, Harold is shown swearing his oath on the relics with a blessing hand.56 Given the sensitive use of gesture throughout the Tapestry, 51 Cf. Cott. Tib. C. VI, fo. 14v (F. Wormald, Collected Writings, i, ill. 140) and see further n. 39 above. One may contrast, inter alia, the version, in Berlin, on the late 10th-c. typological diptych o f Moses receiving the Law and Thomas feeling the wound (A. Goldschmidt, Elfenbeinskulpturen, ii, no. 24; Brandt and Eggebrecht (eds.), Bernward von Hildesheim, ii, cat. 4. 35) where Christ’s hand is raised but, owing to the shape o f the panel, is curled over his head; and that in the early 13th-c. Psalter and Horae, BL MS Arun. 157, fo. 1 l r, where Christ’s hands are engaged in holding his robe open to reveal the wound. 52 In Orleans, BM M S 175, fo. 149r, (Temple, Anglo-Saxon Manuscripts, ill. 144) the small monk has his hands disposed thus, indicating his hopeful receptiveness o f spiritual blessing. As a Carolingian example one might cite the giving and receiving o f judgement image in St Paul in Carinthia, M S XXV. 4. 8, fo. l v. Respectively fos. 36r and 73'. In Abraham’s case it also means he is gesturing towards the stars, the focus o f the implied dialogue. For its use exclusively as a gesture o f prayer (an alternative orant pose) see e.g. the Byzantine serpentine roundel o f 1078-81 in the V. & A. (Williamson, Medieval Treasury, 9 0 -1 ). 54 e.g. fo. 68r, for Joseph riding out in his chariot. 55 Thus in the Benedictional o f St Æthelwold (Temple, Anglo-Saxon Manuscripts, ill. 89; col. repro.: M. P. Brown, Anglo-Saxon Manuscripts, ill. 72), and the Rouen Benedictional and Sacramentary (Talbot-Rice, English A rt, 871-1100, pis. 51b and 55b). Contrast the late llth -c . Wadham Gospels (Oxford, Wadham College, M S 2 (A. 10. 22), fo. 104v: Talbot-Rice, English Art, 871-1100, pi. 64a; Kauffmann, Romanesque Manuscripts, ill. 18) and the Tiberius Psalter, fo. 13' (F. Wormald, Collected Writings, i, ill. 138; discussed in the text and n. 60 below). >6 Wilson, Bayeux Tapestry, pi. 26.

i a. St Mary’s Church, Breamore, Hants, (viewed from south east).

lb. St Laurence’s Church, Bradford-upon-Avon; interior looking north east.

2a. New Minster Charter, pictorial frontispiece. British Library, Cott. Vesp. A. viii, f. 2V. Page size: 206 x 162 mm.


20. New Minster Charter, descriptive verses. British Library, Cott Vesp. A. viii, f. 3r.

3- The Harley 603 Psalter; illustration and part of text for Psalm 13. British Library. Harley 603, f. f . Page size: 380 X 310mm.

4-Judith of Flanders’ Gospels, jewelled binding (front cover). New York, Picrpont Morgan Library, M 708. Size of manuscript: 296 x 197 mm.

5a. Ivon plaque with Christ enthroned. Height: 9.5 cm.

5b. Ivon plaque depicting the Crucifixion. Height: 8cm. Width: 6.5 cm.

6. The New Minster Liber Uitae, frontispiece British Library, Stowe 944, f. 6r. Page size: 255 x 150mm.

7 a, ~jb. New Minster Liber Uitae, folios 6 and 7 . British Library, Stowe 944, ff. 6Nand y .

8. The Boulogne 1 1 Gospels, beginning of John’s Gospel. Boulogne, Bibliothèque Municipale, 1 1 , f. 107'. Page size: 330 x 260mm.

ç. The Old English Hexateuch, the fall of the rebel angels. British Library, Cott. Claud. B. iv, f. 2r. Page size: 342 x 217m m .

io. The Benedictional of St Æthehvold, the Annunciation. British Library, Add M S 49598, f. 5'. Page size: 293 x 225 mm.

11 a. The Benedictional of St Æthehvold, Dormition of the Virgin Mary. British Libran Add. M S 49598, f. 102'.

i ib. The Rouen Benedictional, Dormition of the Virgin Mary. Rouen, Bibliothèque Municipale, Y. 7, f. 54'. Page size: 323 x 245 mm.

"o ro

O sO

JJ i 'c cd

1— 1


J JS _c/d ‘u, OQ sd ro E 13 C/3 CU O c

_o «3 Un C/î



QJ C/5


ro O 0 U> "u cð 1 (U JS H N

iT,b. The Tiberius Psalter, the Harrowing of Hell. British

Library, Cott. Tib. C. vi, f. 14'. Page size: 248 x 146mm.






§ c

§ -S

S & 2

.y . •

Cl "a 0 0

-c < l> .52 jc

x> u •H

t O



5 ^

= -

00 ^

2 ^

§ .£ eg

6 SS S 2