BOLVIG - LINDLEY - History and Images - Towards A New Iconology

History and Images
MEDIEVAL TEXTS AND CULTURES OF NORTHERN EUROPE
5
MEDIEVAL TEXTS AND CULTURES OF NORTHERN EUROPE
EDITORIAL BOARD UNDER THE AUSPICES OF THE

CENTRE FOR MEDIEVAL STUDIES
UNIVERSITY OF HULL
Wendy Scase. Chair (University of Birmingham)

John Arnold (University of London. Birkbeck)
Alan Deighton (University of Hull)
Alan Hindley (University of Hull)
Chris Humphrey (University of York)
Brian Levy (University of Hull)
ADVISORY BOARD
Andrew Ayton (University of Hull)

David Bagchi (University of Hull)
Elaine C. Block (Misericordia International)
Olle Ferm (University of Stockholm)
Wim Hiisken (University of Auckland)
Peter Meredith (University of Leeds)
Bridget Morris (University of Hull)
Veronica O’Mara (University of Hull)
Istvan Petrovics (Jozsef Attila University. Szeged)
Oliver Pickering (University of Leeds)
Graham Runnalls (University of Edinburgh)
Brigitte Schludermann (University of Hull)
David Wallace (University of Pennsylvania)
EDITORIAL ASSISTANT
Elizabeth Wall
History and Images
Towards a New Iconology
edited by
Axel Bolvig and Phillip Lindley
BUEPOLS
British Library Cataloguing in Publication Data.
A catalogue record for this book is available from the British Library.
© 2003 Brepols Publishers n.v., Tumhout, Belgium
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without prior permission of the publisher.
D/2003/0095/29
ISBN 2-503-51155-4
Printed in the EU on acid-free paper.

Contents
List of Illustrations...............................................................................................vii
Acknowledgements............................................................................................. xix
Photographic Acknowledgments........................................................................ xxi
Introduction......................................................................................................xxiii
Im a ges and H is t o r y
Art and history: the legacy of Johan Huizinga.......................................................3

FRANCIS HASKELL t
Images and the historian.......................................................................................19

JEAN-CLAUDE SCHMITT
Nostalgia for the real: the troubled relation of art history to visual culture....... 45
KEITH MOXEY
Im a g e D a ta b a se s and H is t o r y
Pourquoi élaborer des bases de données d’image? Propositions pour une

iconographie sérielle............................................................................................ 59
JÉRÔME BASCHET
Image and word: systematic research into the relations between image and
word in Dutch culture (1500-1800).................................................................. 107
JÖRGEN VAN DEN BERG. HANS BRANDHORST. PETER VAN HUISSTEDE
vi
Six St Jeromes: notes on the technology and uses of computer lighting

simulations........................................................................................................ 131
SIMON NIEDENTHAL
The Lincoln CD-ROM Project: history, theory, conservation, and images..... 139
PHILLIP LINDLEY
Innovative hybride graphische Systeme zur Denkmalüberwachung und

-Verwaltung am Beispiel historischer Wandmalereien—Ein Erfahrungs
bericht............................................................................................................... 165
ROLF-JÜRGEN GROTE. ANNETTE HORNSCHUCH
Cutting off the king’s head: images and the (dis)location of power................ 187
FRANK COLSON. JEAN COLSON. ROSS PARRY. ANDREW SAWYER
‘Serra ex ferro’— ’Serra ex vitro’: medieval history—computers—image

messages reconsidered...................................... 209
GERHARD JARITZ
Quantitative image analysis: the Painter of wooden shoes.............................. 229

AXEL BOLVIG
Im a g e s as So u r c e M a t e r ia l
At the sign of the ‘Spinning Sow’: the ‘other’ Chartres and images of
everyday life of the medieval street...................................................................249
MICHAEL CAMILLE t
Les marges à drôleries des manuscrits gothiques : problèmes de méthode..... 277

JEAN WIRTH
‘Primitive’ paintings: the visual world of populus rusticus............................. 301

HELENA EDGREN
Man and picture: on the function of wall paintings in medieval churches....... 323
ANNA NILSÉN
Representations of Jews in Danish medieval art—can images be used as

source material on their own?............................................................................ 341
ULLA HAASTRUP
Anti-semitism, image desecration, and the problem of ‘Jewish execution’..... 357

NORBERT SCHNITZLER
Framing history with salvation..........................................................................379

S0REN KASPERSEN
On the epistemology of images..........................................................................415

LENA LIEPE
List o f Illustrations
Francis Haskell t
Art and history: the legacy of Johan Huizinga
1. ‘Arnolfini and his wife’, Jan van Eyck. 1434. Oil on oak, 81.8 x 59.7
cm. National Gallery, London. NG 186. © National Gallery. London
fPlate 1)
2. ‘Lucca Madonna’, Jan van Eyck. c. 1436. Oil on wood, 65.5 x 49.5 cm.
Städelsches Kunstinstitut, Frankfurt. © Städelsches Kunstinstitut,
Frankfurt (Plate 2)
3. 'Annunciation and Visitation’, Melchior Broederlam. Tempera on
wood, 167 x 125 cm. Musée des Beaux-Arts, Dijon. © Musée des
Beaux-Arts
4. ‘Singing angels’, Jan van Eyck. The Gent altarpiece. Oil on panel.
Cathedral of St Bavo. Gent. © Cathedral of St Bavo, Gent
5. ‘Shrine of St Ursula’, Hans Memling, detail. Wood. 37.5 x 30 cm.
Bruges Groeninge Museum. © Groeninge Museum, Bruges
Jean-Claude Schmitt
Images and the historian
1. Gospels of Otto III: majesty of kingship. Munich, Bayer. Staatsbiblio
thek, MS. Clm 4453, fol. 24'. © Bayer. Staatsbibliothek, Munich
2. Gospels of Otto III: divine majesty. Munich, Bayer. Staatsbibliothek,
MS. Clm 4453, fol. 34v. © Bayer. Staatsbibliothek, Munich
3. The mystical press (c. 1511): the image of the pope echoing that of God
the Father. © Church of St Gumbert. Ansbach (Plate 3)
4. Decretum Gratiani, decorated initial: the dominant figure of the
canonist. Baltimore, The Walters Art Gallery, MS. W. 133, fol. 123. ©
The Walters Art Gallery, Baltimore
5. St Louis’s Psalter: Jacob’s ladder, and the anointing of the stone of
Bethel. Paris, Bibliothèque nationale de France, ms. lat. 10525, fol. 13v.
© Bibliothèque nationale de France, Paris
6. St Louis’s Psalter: Jacob’s meeting with the angel, and Jacob wrestling
with the angel. Paris, Bibliothèque nationale de France, ms. lat. 10525,
fol. 14r. © Bibliothèque nationale de France, Paris
7. Psalter (13th century): the throne of grace. Cambridge, Trinity College.
MS. B. 11.4, fol. 119. © Trinity College, Cambridge
8. Conques, Abbey Church of Sainte-Foy, tympanum (detail): St Foy in
her church, prostrate before the hand of God
9. Henry the Lion’s Gospels: Henry and Mathilda crowned by the hand of
God. Wolfenbüttel, Herzog August Bibliothek, MS. Guelf. 105, Noviss.
2°, fol. 171'. © Herzog August Bibliothek, Wolfenbüttel (Plate 4)
10. Henry the Lion’s Gospels: divine majesty and the creation of the world.
Wolfenbüttel, Herzog August Bibliothek. MS. Guelf. 105, Noviss. 2°,
fol. 172’. © Herzog August Bibliothek, Wolfenbüttel (Plate 5)
Jérôme Baschet
Pourquoi élaborer des bases de données d’image? Propositions pour une
iconographie sérielle
1. Evangéliaire, Bibliothèque vaticane, ms. Vat. lat. 39, fol. 58, ‘Lazare
dans le Sein d’Abraham’ (troisième quart du xihc siècle). © Biblioteca
apostolica vaticana
2. Psautier de Blanche de Castille, Paris, Arsenal, ms. 1186, fol. 171, ‘le
Sein d’Abraham et l’enfer’ (c. 1225). © Bibliothèque nationale de
France, Paris
3. Légendier dominicain, Oxford, Keble College, MS. 49, fol. 239: ie
Sein d’Abraham’ (c. 1270). © Keble College Library, Oxford
4. Bible de Pampelune, Amiens, Bibliothèque municipale, ms. 108, fol.
255v, ‘le Sein d’Abraham’ (1197). © Bibliothèque municipale, Amiens
5. Psautier de Stephen de Derby, Oxford, Bodleian Library, MS.
Rawlinson G. 185, fol. 97, ie Crucifié dans le Sein du Père’ (milieu
xivc siècle). © Bodleian Library, Oxford
6. Santa Maria de Tahull, fresque absidiale (conservée au Musée d’Art de
Catalogne, Barcelone), ‘Vierge à l’enfant et adoration des mages’ (xuc
siècle). © Author
7. Heures de Jeanne de Navarre, Paris, Bibliothèque nationale de France,
ms. n. acq. lat. 3145, fol. 3V, ‘reine d’Angleterre priant devant la Vierge
à l’enfant et le Trône-de-Grâce’. © Bibliothèque nationale de France,
Paris
8. Sona, Santo Domingo, portail, au tympan : ‘le Fils dans le Sein du
Père’, au centre de la seconde voussure : les élus dans le Sein des
Patriarches (vers 1200). © Author
Illustrations ix
9. Décrétales, Bibliothèque vaticane, ms. Vat. lat. 1389, fol. 4, ‘le Fils
dans le Sein du Père’ (première moitié du xivL ' siècle). © Biblioteca
apostolica vaticana
10. Fribourg, Cathédrale, contrefort, ‘la Vierge au manteau’, xiv' siècle.
© Author
11. Heures de Marie de Bohun, Copenhague, Bibliothèque royale, MS.
Thott. 547, 4°, fol. 32v, ‘la Vierge au manteau protégeant les élus lors du
Jugement dernier’ (c. 1380-95). © Kongelige Bibliotek, Copenhagen
Jörgen van den Berg, Hans Brandhorst, Peter van Huisstede

Image and word: systematic research into the relations between image and word
in Dutch culture (1500-1800)
1. Device number 0223. © Institute of Information and Computing
Sciences, Utrecht University
6. Detail of device number 0223. © Institute of Information and
Computing Sciences, Utrecht University
sciences, Utrecht University
Simon Niedenthal
Six St Jeromes: notes on the technology and uses of computer lighting
simulations
1. Cornell box, default lighting. © Author (Plate 6)
2. Cornell box, raytraced rendering. © Author (Plate 7)
3. Cornell box, radiosity rendering. © Author (Plate 8)
4. Maya camera icon. © Maya (Plate 9)
5. ‘St Jerome in his study’, Antonello da Messina, c. 1450. © National
Gallery, London (Plate 10)
6. Wireframe. © Author (Plate II)
7. Untextured three-dimensional form. © Author (Plate 12)
8. Default light and textures, no shading. © Author (Plate 13)
9. Jerome. © Author (Plate 14)
Phillip Lindley
The Lincoln CD-ROM Project: history, theory, conservation, and images
1. The west front, Lincoln Cathedral. © National Buildings Record,
London
2. ‘The Death of Lazarus and Dives and his companions in Hell’ (panel 8.
according to Professor Zarnecki’s numbering system) after 1987
cleaning. © The Dean & Chapter, Lincoln
3. ‘The Death of Lazarus’ block has been copied, and missing detail (e.g.
the soul of Lazarus) supplied by the sculptor John Roberts. The lower
block of ‘Dives . . . in Hell’ is also a copy carving. © Author
4. Eighteenth-century ‘Torments’ panel after recent conservation. ©
Central Photographic Unit, University of Leicester
5. South side of west façade of Lincoln Cathedral in 1994. © Author
6. Detail of the above. © Author
7. Perpendicular-style window inserted by John de Welburn, in the
fourteenth century, into the framework of a thirteenth-century one. ©
Author
8. North side of central portal in May 1996. © Author
9. The panel showing the punishment of Lust from the ‘Torments’
sequence on the north side of the façade. © Central Photographic Unit,
University of Leicester
10. ‘The Deluge’ and ‘Giants’ (panels 16 and 17) in the Ringers’ Chapel.
© Author
11. View of left-hand side of ‘Noah building the Ark’ (panel 15) (photo
taken in 1996). © Author
12. Frontal view of ‘Noah building the Ark' (panel 15) (photo taken in
1994). © Author
13. Junction between ‘The Harrowing of Hell’ and ‘The Elect in Heaven’
(panels 4 and 5). © Author
14. ‘Abraham’s Bosom’ (panel 6). © Author
15. ‘Adam and Cain’ (panel 10). © Author
16. ‘The Feast of Dives’ (panel 7) after conservation and removal from the
west façade, now on exhibition in the cathedral. © Central Photographic
Unit, University of Leicester
Rolf-Jürgen Grote and Annette Hornschuch

Innovative hybride graphische Systeme zur Denkmalüberwachung und -
Verwaltung am Beispiel historischer Wandmalereien—Ein Erfahrungsbericht
1. Haus Altenkamp, Ansicht von Westen. © Landesamt für
Denkmalpflege, Hannover
Illustrations xi
2. Haus Altenkamp, räumliche Darstellung des Entrées überlagert mit

Schadenskartierung, Ansicht von Nordwesten. © Landesamt für
Denkmalpflege, Hannover (Plate 20)
3. Haus Altenkamp. Entrée, Wandmalerei-Ausschnitt der Südwand, Über
lagerung der Farbauszüge. Das Farbbild wurde aus den schwarz/weißen
Spektralbildern berechnet. Durch Kalibrierung des fotografischen
Prozesses läßt sich so eine langzeitstabile, rekonstruierbare
colormetrische Abbildung realisieren. © Landesamt für Denkmalpflege,
Hannover (Plate 21)
4. Haus Altenkamp, Entrée, Wandmalerei-Ausschnitt der Südwand, UV-
Fluoreszenz-Aufnahme. © Landesamt für Denkmalpflege, Hannover
5. Das Bild entsteht durch Beleuchtung mit 320-380 nm und durch
Filterung mit 440-660 nm. Die Verwendung unterschiedlicher Farben
bei früheren Restaurierungen der linken Figur wird deutlich sichtbar. ©
Landesamt für Denkmalpflege, Hannover
6. Haus Altenkamp, Entrée, Bildplan der Südwand, überlagert mit
Thermobild. © Landesamt für Denkmalpflege, Hannover
7. Das Thermobild setzt sich aus 4 Einzelbildern zusammen. Vom Objekt
reflektierte Strahlungen im Bereich zwischen 8 und 14 um wurden
registriert. Die Abbildung zeigt Kältebrücken an den Seiten der Wände
(Anschluß zur Außenwand bzw. zu unbeheizten Nachbarräumen). Der
wärmere (grüne) Bereich in Bildmitte weist darauf hin, daß durch einen
offenen Kamin im Nachbarraum Wärme abgesogen wird. © Landesamt
für Denkmalpflege, Hannover (Plate 22)
8. Haus Altenkamp, Entrée, Thermobild der Südwand, überlagert mit
Schadenskartierung. © Landesamt für Denkmalpflege. Hannover (Plate
23)
9. Die Überlagerung korreliert die Schadenskartierung mit
bauphysikalischen Aussagen. © Landesamt für Denkmalpflege,
Hannover (Plate 24)
10. Haus Altenkamp, Entrée, nach der Restaurierung. © Landesamt für
Denkmalpflege, Hannover (Plate 25)
Frank Colson, Jean Colson, Ross Parry, Andrew Sawyer

Cutting off the king’s head: images and the (dis)location of power.
1. ’Penning’ dated 1588: from H. E. Greve, De Tijd van den Tachtig-
jarigen Oorlog in Beeid (Amsterdam: Elsevier, 1908), p. 67
2. C. J. Visscher, ‘Afbeelding v a n ’t Nederlandts Bestandt’ [Image of the
Netherlands Truce], c. 1609. FM1267. Reproduced by permission of the
Rijksprentenkabinet, Rijksmuseum, Amsterdam
3. Detail, ‘Afbeelding van ’t Nederlandts Bestandt’ [Image of the
Netherlands Truce], c. 1609. FMI 267. Reproduced by permission of the
4. ‘Livre’, Revista 1Ilustrada, Rio de Janeiro, 14 October 1884
5. ‘Não vos aproximeis de mim! Vossas mãos ainda tintas de sangue dos
escravos manchariam as minhas vistas! Retirae-vos, tu não vos quero.’
Revista IIlustrada 15 June 1888 [original orthography]
6. Inigo Jones, ‘Pallas of Perfection’, Devonshire Collection, Chatsworth.
Reproduced by permission of the Duke of Devonshire and the
Chatsworth Settlement Trustees. © Photographic Survey, Courtauld
Institute of Art, London
7. Representation of a view from the west balcony of the first Jacobean
Banqueting House (c. 1607-1619) looking towards Inigo Jones’ scenery
for The Vision of Delight (1617). Computer-generated model, Ross
Parry, 1999. Details from Inigo Jones, ‘A Street in Perspective’ and
‘Proscenium and Hunt Scene’, Devonshire Collection, Chatsworth.
Reproduced by permission of the Duke of Devonshire and the
Chatsworth Settlement Trustees. © Photographic Survey, Courtauld
Institute of Art, London (Plate 26)
Gerhard Jaritz
‘Serra ex ferro’— ’Serra ex vitro’: Medieval history—computers—image
messages reconsidered
1. ‘Visitation of the Virgin’, Master of the Schottenaltar, panel painting,
Austrian, 1469/1480. Vienna, Schottenstift. © Institut für Realienkunde
des Mittelalters und der frühen Neuzeit, Krems
2. ‘Christ Visiting St George in Prison’, panel painting, 1516. Spisská
Sobota (Slovakia), parish church. © Institut für Realienkunde des
Mittelalters und der frühen Neuzeit, Krems
3. ‘The Dream of St Joseph’, Friedrich Pacher, (detail: the hard road to
Egypt), panel painting, Tirol, before 1500. Innsbruck, Landesmuseum
Ferdinandeum. © Institut für Realienkunde des Mittelalters und der
frühen Neuzeit, Krems
4. ‘The Temptation of St Martin by the Disguised Devil’, panel painting,
before 1500. Göflan (Südtirol), parish church. © Institut für Realien
kunde des Mittelalters und der frühen Neuzeit, Krems
5. ‘St Florian and St Sebastian’ (detail: Emperor Frederick III as St
Sebastian), panel painting, Styrian, c. 1480. Obdach (Steiermark),
church of the hospital. © Institut für Realienkunde des Mittelalters und
der frühen Neuzeit, Krems
6. ‘Parable of Dives and Lazarus’ (detail: the servant of Dives), panel
painting, Upper Rhine, end of the fifteenth century. St Paul im
Lavanttal (Kärnten), collections of the Benedictine monastery. ©
Institut für Realienkunde des Mittelalters und der frühen Neuzeit,
Krems
7. ‘Labours of the Months’, month of August (detail: hairstyle of an ‘im
Illustrations x iii
proper’ peasant), ink drawing, South German, 1475. Vienna, Östereich.

Nationalbibliothek, MS. 3085, fol. 5r. © Institut für Realienkunde des
8. ‘Duke Albrecht of Austria’, ink drawing, Tirol, last quarter of the fifteenth
century. Vienna, Östereich. Nationalbibliothek, MS. s. n. 12820, fol. 40v.
© Institut für Realienkunde des Mittelalters und der frühen Neuzeit, Krems
9. ‘Sermon of St John Capistran in Bamberg’ (detail: Bonfire of the
Vanities), panel painting, c. 1470. Bamberg, Staatsgalerie. © Institut für
Realienkunde des Mittelalters und der frühen Neuzeit, Krems
10. ‘Labours of the Months’, month of July (detail: peasants mowing and
hay-making), wall-painting, c. 1400. Trento, Castello Buonconsiglio,
Torre d’Aquila. © Institut für Realienkunde des Mittelalters und der
frühen Neuzeit, Krems
11. ‘Last Judgement’ (detail: the devil grips a peasant), panel painting,
Schnatterpeck-workshop, first quarter of the sixteenth century. Kortsch
(Südtirol), trustee church. © Institut für Realienkunde des Mittelalters
und der frühen Neuzeit, Krems
12. ‘Labours of the Months’, month of August (detail: one of the
‘improper’ field-working peasants), ink drawing, South German, 1475.
Vienna, Östereich. Nationalbibliothek, MS. 3085, fol. 5f. © Institut für
Realienkunde des Mittelalters und der frühen Neuzeit, Krems
Axel Bolvig
Quantitative image analysis: the Painter of wooden shoes
1. Gudme church. Wall painting, ‘wooden shoe’. 1488. © Author
2. 0rbæk church. Wall painting, ‘hand with a knife’. 1490. © Author
3. Stege church. Wall painting, ‘two wooden shoes’. 1494. © Author
4. Nibe church. Wall painting, ‘pig wearing pattens’. Early sixteenth
century. © Author
5. Gislev church. Wall painting, ‘Crown of Thoms’, detail, c. 1520. ©
Author
6. Sulsted church. Wall painting, ‘Flagellation of Christ’. 1548. © Author
Michael Camillef
At the sign of the ‘Spinning Sow’: the ‘other’ Chartres and images of everyday
life of the medieval street
1. Chartres. Distant view of the cathedral from the south east with ‘Basse-
Ville’ in foreground. © Author
2. Chartres, ‘Maison du Saumon’. © Author
3. Chartres, the ‘Truie qui file’ on the ‘Maison du Saumon’. © Author
4. The so-called ‘Truie qui file’ on the south tower of the cathedral, from
L’Abbé Bulteau, Monographie de la Cathédrale de Chartres, ii,
(Chartres, 1888)
X IV
5. Chartres Cathedral, north transept, right porch, right outer archivolts,

'December pig killing’. © James Austin
6. Chartres Cathedral, north transept, left porch, left outer archivolts.
‘Women carding and spinning’. © James Austin
7. ‘Woman with distaff defecating’. Wooden ‘aisselier’ from an apothe
cary’s shop, late fifteenth century, Nantes, Musée Dobrée. © Author
8. ‘Spinning sow’. Wood. Maiestroit, Brittany, fifteenth century. © Author
9. ‘Spinning sow and monkeys’, Livre des simples médecines, Paris
Bibliothèque nationale de France, ms. fr. 9136 fol. 1, late fifteenth
century. © Bibliothèque nationale de France, Paris
10. Chartres, ‘Salmon’ from the ‘Maison du Saumon’. © Author
11. Chartres Cathedral, north transept, right porch, left outer archivolts,
Pisces. © James Austin
12. ‘A la Truie qui file’ sign of a charcutier, early seventeenth century,
Paris, Musée Carnavalet. © Musée Carnavalet, Paris
Jean Wirth
Les marges à drôleries des manuscrits gothiques: problèmes de méthode
1. Heures de Jeanne d’Evreux, New York, Cloisters, MS. 54.1.2, fols 15-
16, ‘Arrestation du Christ et Annonciation’. © The Cloisters, New York
2. Bréviaire de Renaud de Bar. Verdun, Bibliothèque municipale, ms. 107,
fol. 138', ‘Siège d’un château’. © Bibliothèque municipale, Verdun
3. Psautier de Gorleston, Londres, British Library, MS. Add. 49622, fol.
162v, ‘Chevalier effrayé par un escargot’. © British Library, London
4. Psautier gantois, Oxford, Bodleian Library, MS. Douce 6, fol. 92v,
‘Couvade’. © Bodleian Library, Oxford
5. Psautier, Oxford, Bodleian Library, MS. Laud. lat. 84, fol. 19. ‘Initiale
Beams'. © Bodleian Library, Oxford
6. Livre d’heures, New York, Pierpont Morgan Library. MS. M. 754, fol.
48v, ‘Singes boulangers’. © Pierpont Morgan Library, New York
7. Psautier gantois, Oxford, Bodleian Library, MS. Douce 5, fol. 146,
‘Ecole des singes’. © Bodleian Library, Oxford
8. Psautier-Heures de Geoffroy d’Aspremont et d’Isabelle de Kievraing,
Oxford, Bodleian Library, MS. Douce 118, fol. 101', ‘David en singe
couronné’. © Bodleian Library, Oxford
Helena Edgren
‘Primitive’ paintings: the visual world of populus rusticus
1. Church of St Mary (Maaria), Southern Finland, ‘Primitive paintings’.
© Author
2. Church of Borgund, Norway, ‘Graffiti’, from Martin Blindheim, Grufati
in Norwegian Stave churches c. 1150-c. 1350 (Oslo: Universitetsforlaget,
1985)
Illustrations XV
3. Church of Nousiainen, Southern Finland, ‘Primitive painting’. ©

Finnish National Board of Antiquities, Picture Archives
4. Church of Nousiainen, ‘Primitive paintings’. © Author
5. Church of St Mary, ‘Saint Olav of Norway or Saint Erik of Sweden’. ©
Author
6. Church of St Mary, ‘Devil, primitive painting’. © Finnish National
Board of Antiquities, Picture Archives
7. Church of Korppoo, Southern Finland, ‘Ship’, probably depicting the
journey of St Olav of Norway to Finland. © Author
8. Church of St Mary. ‘Mary’. © Author
9. Church of St Mary, ‘Coat of arms’. © Author
10. Cathedral of Turku, ‘Primitive painting’. © Author
11. Church of Karjaa, Southern Finland, ‘Apostle’ from a credo series.
© Author
12. Church of Taivassalo, Southern Finland, ‘Coat of arms’. © Author
13. Church of St Mary. ‘Master mason and his assistant’. © Author
14. Church of St Mary. ‘Dominican friar'. © Author
15. Church of St Mary, ‘Devil’. © Finnish National Board of Antiquities,
Picture Archives
16. Church of Pyhtää, Southern Finland, ‘Figures from a credo series and
Christ on the Cross’. © Finnish National Board of Antiquities, Picture
Archives
17. Church of Pyhtää, ‘The Apocalyptic Madonna and other paintings’.
© Finnish National Board of Antiquities. Picture Archives
Anna Nilsén
Man and picture: on the function of wall paintings in medieval churches
1. Ärentuna church, Uppland (Sweden). © Author
2. Roslagsbro church. Uppland. © Author
3. Roslagsbro church. © Author
4. Ärentuna church, c. 1435. ‘The seven deadly sins with their roots’.
© Author
5. Härkeberga church. Uppland, c. 1485. ‘The Wheel of Fortune’.
© Author
6. Härnevi church, Uppland, c. 1485. ‘True and false prayer’. © Author
7. Vendei church, Uppland, painting in the porch, c. 1452. ‘The Trinity’.
© Author
8. Tensta church. Uppland. End thirteenth century. ‘Exterior from south’.
© Author
9. Tensta church. Paintings from 1437-38. © Author
10. Tensta church. ‘Bengt Jönsson, the donor of the paintings, at the feet of
St Peter’. © Author
11. Tensta church. ‘The Annunciation’. © Author
XVI
12. Tensta church. ‘Revelations of St Birgitta'. © Author
Ulla Haastrup
Representations of Jews in Danish medieval art—can images be used as source
material on their own?
1. Jelling church, fresco, 1080-1100, ‘St John the Baptist preaching to a
gathering of Jews’, Watercolour by the restorer J. Magnus-Petersen
1874. © Nationalmuseet, Copenhagen
2. Raasted church, fresco, about 1125, ‘judex barbatus’, in the scene of the
Magi at the court of King Herod. © Author
3. Altarpiece from Boeslunde church, now at the National Museum in
Copenhagen, about 1425-35, Lübeck. ‘The Virgin Mary’s Training in
Reading’. © Nationalmuseet, Copenhagen
4. Altarpiece in St Mary in Helsingborg, 1449-52. ‘The Circumcision’. ©
Author
5. Tuse church, wall painting, 1460-80. ‘The Circumcision’.
© www.kalkmalerier.dk
6. Fanefjord church, wallpainting, about 1500. ‘Jesus among the Doctors’.
© Author
7. Haggadah. German, c. 1460, London, British Library, MS. Add. 14762,
fol. T. ‘German Jews’ (from Metzger: Jewish Life, fig. 175)
8. Fanefjord church, wall painting, c. 1500. ‘The Circumcision’. © Author
9. Hebrew manuscript, Naples, Third quarter of the fifteenth century.
Nîmes, Bibliothèque municipale, ms. 13, fol. 181’. ‘The Circumcision’
(from Metzger: Jewish Life, fig. 333)
10. Master of The Tuscher-altar, Nürnberg, c. 1450, ‘The Circumcision’.
Aachen, Suermondt-Museum, from Mellinkoff, Outcasts, voi. ii, fig.
11.23
11. Stege church, wall painting 1494. ‘Kristen iude and figures with Jewish
hats’. © Author
Norbert Schnitzler
Anti-semitism, image desecration, and the problem of ‘Jewish Execution’
1. Engraving representing the panel-painting with the Miracle of Notre-
Dame de Cambrón, c. 1890, J. van Péteghem, Brussels
2. First field of the panel-painting: The image desecration, detail, c. 1890,
J. van Péteghem, Brussels
3. Fifth field of the panel-painting: Mary appears in a vision to the
blacksmith, detail, c. 1890, J. van Péteghem, Brussels
4. Tenth field of the panel painting: Execution of the baptized Jew capite
traverso, detail, c. 1890, J. van Péteghem, Brussels
5. Oath-Taking, from Ende rung vnd Schmach der Bildung Mariae von den
Juden bewiesen, Strasbourg c. 1515
Illustrations X VII
6. Torture, from Ende rung vnd Schmach der Bildung Mariae von den
7. Execution of the baptized Jew, from Enderung vnd Schmach der
Bildung Mariae von den Juden bewiesen, Strasbourg c. 1515
8. Study of a man hanging upside-down, fifteenth century, Andrea del
Sarto. © Gabinetto di disegni Uffizii, Florence
9. So-called ‘Schandbrief by the nobleman Alexander of Oberg addressed
to five noblemen who had answered for Duke Erich I of Calenberg-
Göttingen as debtors; their arms are shown upside-down. About 1540,
Wolfenbüttel, Niedersächsisches Staatsarchiv, Sig.: 1 Alt. 26. ©
Niedersächsisches Staatsarchiv, Wolfenbüttel
10. Death of Judas, c. twelfth century, from the so-called ‘Hildegard-
Codex’. © Bayerische Staatsbibliothek, Munich
Sdren Kaspersen
Framing history with salvation
1. Keldby Church. Decoration on the eastern part of the north wall of the
nave, c. 1325. © Lennart Larsen, Nationalmuseet, Copenhagen
2. Keldby Church. Decoration on the central part of the north wall of the
nave, c. 1325. © Nationalmuseet, Copenhagen
3. Keldby Church. The Last Judgement on the east wall of the nave, c.
1325. © Author
4. Keldby Church. The Flight into Egypt/Sotinen, c. 1325. © Author
5. Keldby Church. Prophet/chronicler? c. 1325. ©Author
6. Keldby Church. Tombstone for Henning Moltke, his wife and his
daughter (?), 13557/1360-70, from J. B. Löffler, Danske Gravslene fra
Middelalderen (Copenhagen, 1885)
7. Keldby Church. The Last Judgement: Purgatory, Resurrection of the
Dead, and coat of arms, c. 1325. © Author
8. (a-b) Hamburg, high altar from St Petri’s Church, 24 panels by Master
Bertram, c. 1379-83. © Kunsthalle, Hamburg
9. Hamburg, six panels on the high altar from St Petri’s Church,
uppermost ‘The Fall of the Rebel Angels’, ‘Thecreated Firmament /
palatium’, and ‘The Creation of Sun, Moon, and Stars’, last ‘The
Sacrifice of Cain and Abel’, ‘Cain Slaying Abel’, and ‘The Building of
Noah’s Ark', c. 1379-83. © Elke Walford, Kunsthalle, Hamburg
uppermost ‘God’s Curse on Adam and Eve’, ‘The Expulsion of Adam
and Eve’, and ‘The First Labour’, last ‘The Presentation in the Temple’,
‘The Massacre of the Innocents’, and ‘The Rest on the Flight into
Egypt’, c. 1379-83. © Elke Walford, Kunsthalle, Hamburg
uppermost ‘The Creation of Eve’, ‘The Admonition of Adam and Eve’,
x v iü
and ‘The Fall’, last ‘The Annunciation’, ‘The Nativity’, and ‘The
Adoration of the Three Kings’, c. 1379-83. © Elke Walford,
Kunsthalle, Hamburg
Lena Liepe
On the epistemology of images
1. Unknown artist: frontispiece, Giles of Rome, De regimine principum,
translation to French by Jean Wauquelin, copy from 1452 by Jacques
Pilavaine. Brussels, Bibliothèque Royale Albert I", ms. 9043, fol. 2r.
© Bibliothèque Royale Albert I", Brussels
2. Unknown artist: frontispiece, Jacques de Guise, Annales Hannoniae,
translation into French by Jean Wauquelin, copy from 1448 by
Wauquelin and Jacotin du Bois. Brussels, Bibliothèque Royale Albert
Ier, ms. 9242. fol. Ir. © Bibliothèque Royale Albert 1", Brussels
Acknowledgements
oth editors would like to record their great debt of gratitude to Dr Brian
B Levy, Reader in French at the University of Hull, for his exemplary

translation of Jean-Claude Schmitt’s article, his checking of the French
throughout the book, his assistance with many of the other translations and for
his general care and attention throughout the book’s progress to the printers. No
general editor could have done more. We also would like to thank Dr Simon
Forde of Brepols Publishers for his great interest in the present volume.
Axel Bolvig would also like to record his thanks to Annedorte Vad for all her
practical work.
The Editors, and the Editorial Board of Medieval Texts and Cultures of
Northern Europe, deeply regret that the essays of Michael Camille and Francis
Haskell are here published as posthumous testimony to the scholarship of two
eminent and sadly missed colleagues. This volume is dedicated to their memory.
Photographie Acknowledgements
Austin, James
Bayerische Staatsbibliothek, Munich

Biblioteca apostolica vaticana
Bibliothèque municipale, Amiens
Bibliothèque municipale, Verdun
Bibliothèque nationale de France, Paris
Bibliothèque Royale Albert I", Brussels
Bodleian Library, Oxford
British Library, London
Church of St Gumbert, Ansbach

Cloisters, New York
Courtauld Institute of Art, London
Dean & Chapter. Lincoln

Duke of Devonshire and the Chatsworth Settlement Trustees
Gabinetto di disegni Uffizii, Florence

Groeninge Museum, Bruges
Herzog August Bibliothek, Wolfenbüttel
Institut für Realienkunde des Mittelalters und der frühen Neuzeit, Krems
Institute of Information and Computing Sciences, Utrecht University
xxü
Keble College Library, Oxford

Kongelige Bibliotek, Copenhagen
Kunsthalle, Hamburg
Landesamt für Denkmalpflege, Hannover

Larsen, Lennart, Nationalmuseet, Copenhagen
Musée Carnavalet, Paris

Musée des Beaux-Arts, Dijon
National Board of Antiquities, Finland

National Buildings Board
National Gallery, London
Nationalmuseet, Copenhagen
Rijksprentenkabinet, Rijksmuseum. Amsterdam
Trinity College, Cambridge
University of Leicester
Walford, Elke, Kunsthalle. Hamburg

Walters Art Gallery, Baltimore
Introduction
A xel Bolvig
n 1999. the History and Images Congress was held at the University of
Copenhagen. It was arranged as part of a research project initiated by the
B. Faculty of Humanities under the heading The Visual Construction of
Realities. Historians and art historians presented a variety of new approaches to
research on visual material. The papers of the congress form the basis of this
volume. The main object was, and remains, not only to show how the images
may be understood as an equal partner to the text in historical research but also
to demonstrate that imagery constitutes a separate category of source material
with its own category of meaning and information and requiring its own
interpretative methodologies. The subtitle ‘Towards a New Iconology’ indicates
that traditional approaches to images are becoming inadequate. Historians who
have expected art to reveal directly aspects of the historical society under
investigation as a mimetic reproduction of the past, art historians who have
looked solely at the aesthetic qualities of images, those who have investigated in
iconography or iconology, all those who from Gregory the Great to the present
day have linked image to text, all now seem insufficient.1*The contributions to
the present volume demonstrate that the visual construction of realities
represents a unique temporally located world of our own times as well as the
‘reality' of an historical past. And they demonstrate that this world of the eye is
as complex and ubiquitous as the world of the word.
1Concerning an evaluation of the famous statement of Gregory the Great, see Jérôme
Baschet. ‘Introduction: Limage-objet’, in L'image. Fonctions et usages des images dans
l'occident médiéval, ed. by J. Baschet and J.-C. Schmitt (Paris: Le Léopard d'Or, 1996),
pp. 7-26; Mary Carruthers. The Book o f Memory. A Study of Memory in Medieval Culture
(Cambridge: Cambridge University Press. 1990: repr. 1999). pp. 221-29.
X XIV A xel B olvig
It was written that in the beginning was the Word, but God soon felt obliged
to place the prohibition of the making of images in the Ten Commandments on
equal terms with the prohibition of killing. Modern research has discovered that
among our five senses the eye is the most active. The eye sends ten million bits a
second to the brain, the skin sends one million, the ear 100,000, the sense of
smell 100,000, and the sense of taste only 1000 bits a second.2The sense of sight
is then by far the strongest of our senses, sending enormous amounts of
information to the brain. This in itself is a weighty argument for dealing with
images. The text is also apprehended as an image by the eye, but being a visual
representation of the linear system of the spoken words, the contents of a text is
only understood in the course of time. A picture, in contrast, represents all its
information within the same syntax.3
In the present volume, Keith Moxey argues for the establishment of Visual
Studies: ‘Rather than seek consensus, it benefits from the radical disagreement
between different styles of interpretation’. The contributions to this volume
present many different approaches to research in pictures as images and art and
as visual communication and should, therefore, accord with his definition of
visual studies. The film director Michelangelo Antonioni’s credo reads:
But we know that behind every image revealed there is another image more fateful
to reality and in back of that image there is another and yet another behind the last
one and so on up the true image of that absolute mysterious reality that no one will
ever see.4
This is what modern researchers on the visual world of the historic past have also
realized: every picture has many layers and contains many scientific ‘truths’. The
end of the last millennium was accompanied by the pluralism of deconstruction,
which initiated the ‘meltdown’ of historical and art historical methods, described
by Jean-Claude Schmitt in ‘Images and the historian’. Are we facing a new
constructive era based on a deconstructivist fragmentary inheritance?
In 1928, René Magritte painted a picture with visual configurations which re
sembled a pipe. Over this simple and rather un-aesthetic depiction he wrote: ‘Ceci
n’est pas une pipe’. The title of the picture reads: ‘La trahison des images’. It took
decades for researchers in history and art history to realize that their work with
images was embedded in ‘treachery’. In my opinion Magritte’s simple and ‘un
’ Bent Fausing, Synet som sans (Copenhagen: Samlerens Bogklub. 1995), pp. 17, 54.
’ Michael Baxandall argues that analysing an image is equal to scanning a text (M.
Baxandall, Patterns o f Intention. On the Historical Explanation of Pictures (New Haven:
Yale University Press. 1985: repr. 1992). p. 3). In my view this has nothing to do with the
non-linearity of an image. The textualizing of an image is caused by the fact that when
analysing we ‘translate’ the visual contents to a linear linguistic system.
4 Michelangelo Antonioni (dir.), Al di là delle nuvole (Beyond the clouds). 1995: lines
spoken by John Malkovich (The Director).
Introduction XXV
artistic’ combination of a mimetic configuration and a linguistic statement is one of

the many examples showing that artistic creativity most often forms the avant-garde
in raising questions. In my view, the contributions to this volume demonstrate an
inspiring willingness during the last couple of decades to deal with the ‘treachery’ of
images by accepting the uniqueness of visual information. The ‘treachery’ we find is
within the words, which are connected to an image as a title or as a description, not
within the image itself. And why not think that medieval artists, too, often formed an
avant-garde?
One could argue that iconographie studies for centuries have realized the
‘treachery’ of images if we define iconography as a conventional codified
distance between expression and content. From the ancient hieroglyphs and
culminating in medieval Christian imagery, such conventions have dominated
image making. The more or less arbitrary linkage between content and
expression is a ‘treachery’ towards the apparent analogue visual presentation.
This is not the intriguing point in Magritte’s canvas. He is, so I think, alluding to
our belief in the analogue aspect in visual representation. With the rise of portrait
and especially landscape painting, iconography lost influence in the visual arts.
At least its role is no longer as clear-cut as it used to be.5 Photography changed it
all in such a way that, according to Roland Barthes, a photo in a strange way
represents an un-coded message.67As a matter of fact, when looking at a photo of
a pipe we all in our verbal descriptions admit ‘it is a pipe’. It is we who commit
treachery towards the image—or to use Gerhard Jaritz’s explanation: ‘Using
verbal descriptions of pictorial sources (or also original objects), we only rarely
have the opportunity to deal with anything like the original wording of the
source of full-text. We are creating our own text’.'
The development of photography and the establishment of history and art
history as scientific occupations go hand in hand. When in 1839 the French
government bought the rights to the daguerreotype process, the purpose was to
support scientific research. Photographers were sent to Egypt to make ‘real’
copies of the hieroglyphs. We still use photographs as a mimetic or at least an
acceptable equivalent to the original image; as a matter of fact, the plates in the
present volume are based on this assumption. But things have changed. Today
we know that even Robert Capa composed ‘fakes’: the falling man from the
Spanish Civil War is not a dying republican soldier. Nobody believes in
photography any longer. Benetton made the final post-modern showdown with
5 Brendan Cassidy, ‘Introduction. Texts, and Audiences’, in Iconography at the Cross

roads. ed. by Brendan Cassidy (Princeton: Index of Christian Art, Princeton University,
1993). pp. 3-15.
6 Roland Barthes, ‘Rhétorique de l’image", Communications, 4 (1964), 40-51.
7 Gerhard Jaritz, Images. A Primer o f Computer-Supported Analysis with Kleio IAS (St
Katharinen: Max-Planck-Institut für Geschichte, 1993), p. 16.
XXVI A xel Bolvig
the mimetic representational power of the most realistic visual medium,

photography. Furthermore, the simulation of a painting in three-dimensional
(3-D) space strikes a blow against the ascendancy of photorealism in computer
rendering. Simon Niedenthal demonstrates this by colliding a painting by
Antonello da Messina, ‘St Jerome in his Study’, with the most recent rendering
technology and suggesting the use of radiosity processing to check a 2-D
representation against a 3-D structure.
The development of landscape and portrait painting gradually moved art away
from iconography. The invention of photography marked a culmination of this
movement. The final exposure of the ‘treachery’ of images has questioned the
mimetic aspect. In a way these movements create a point of departure for
modem research in pictures. With the loss of iconography and the mimetic
aspect of realism in all kinds of pictures (dare I call it a pre-iconographic level of
an image?) we are free to turn to the images of the past with new eyes. Is this
what Erwin Panofsky meant when at the end of his life he gave up iconology?
And is it not fair to think that the same situation has forced Irving Lavin to say
that he sometimes thinks iconography is an invention of the devil?s Many of the
articles in this volume demonstrate a marginalization of traditional iconographie
studies and introduce not so conventional visual codes. Norbert Schnitzler’s
concern is significant for many of the articles when he demonstrates the mutual
stimulations that combine new juridical concepts like ‘deterrence’ with
corresponding concepts of representation. Is he elaborating another iconography
or does he work at a new iconological level? None of the articles deals with the
mimetic aspect as a documentary.*9
The post-World War situation with the decline of the colonial powers has
erased (art) historical explanations based on a centre-periphery theory. In the
modern world there are no real centres of cultural or art historical activities.
Finally the Internet has shown us an absolute decentralization of culture and
information.10 Hundred of thousands of images put on the Internet on equal terms
make it almost impossible to point to specific artistic centres of the past. Such
experiences are reflected in Michael Camille’s essay ‘At the sign of the
“Spinning Sow”: The “other” Chartres and images of everyday life of the
s Irving Lavin, ‘Iconography as a Humanistic Discipline (“Iconography at the Cross

roads”)', in Iconography at the Crossroads, ed. by Brendan Cassidy (Princeton: Index of
Christian Art. Princeton University, 1993), pp. 33, 38.
9 Cf. Keith Moxey, ‘Reading the “Reality Effect” ’, in Pictura quasi fictura. Die Rolle
des Bildes in der Erforschung von Alltag und Sachkultur der Mittelalters und der frühen
Neuzeit, ed. by Gerhard Jaritz (Vienna: Verlag der österreichischen Akademie der
Wissenschaften. 1996). pp. 15-21.
,0 Axel Bolvig, ‘Ars longa—vita brevis’, in Medium Aevum Quotidianum, ed. by
Gerhard Jaritz (Krems: Gesellschaft zur Erforschung der materiellen Kultur des
Mittelalters. 1998). pp. 7-20.
Introduction x x v ii
medieval street’. The famous cathedral of Chartres is not necessarily the only
centre of the city. Its art is not the only one worthy of research. The
preoccupation with so-called ‘marginal’ images has been so intense during the
last decades that Jean Wirth treats the subject in an overall discussion of the
research into the marginal figures of manuscripts from Emile Mâle’s negation of
meaning to Sharon Davenport’s reading a spiritual meaning into every example.
After Marcel Duchamp’s ‘Ready Mades’ there can be no real distinction
between high art and low art or non-art. In ‘ “Primitive paintings: the visual
world of populus rusticus’, Helena Edgren writes: ‘The primitive paintings still
suffer from the label of “second-rate art” that they received a hundred years ago,
and consequently they have not been able to compete with finer paintings for the
attention of art historians’. Now the competition is vigorous. Michael Camille’s
‘Spinning Sow' and Axel Bolvig’s ‘Wooden Shoes’ have nothing to do with art
with a capital A.
Another questioning of the traditional preoccupation with the intentional
meanings of an image, be they those of the artist or the donor, has been—it is my
conviction—inspired by the modern world’s use of pictures. No contemporary
visual media can survive without reflecting on ‘what the spectator sees’." Could
and should the spectator see and experience one and only one intentional
meaning? Anna Nilsén reflects on conditions of visual communication in her
article ‘Man and picture: on the function of wall paintings in medieval churches’:
\ . . the interiors of the churches were too dark for the whole body of paintings
to be identified by the congregation’.
Since the establishment of the technique of printing, historians have had
access to huge amounts of written source material. This is one of the
explanations why most historians fastened themselves in linguistic irons. Since
Leopold Ranke, historical methodology has bound itself to the Gutenberg
Galaxy.*12 In her article, Lena Liepe, an art historian, notes that ‘after all, it is not
the image itself that interests the historian’.13 Today the situation has turned
around 180 degrees. You can write a history of world art in one volume but you
can only print a small selection of the works of art. In this new millennium—in
principle—you can store millions of pictures on huge hard disks but nobody will
have time or energy to read about each single object. Image databases offer
" The expression refers to Richard Wollheim, ‘What the Spectator Sees’, in Visual
Theory: Painting and Interpretation, ed. by N. Bryson. M. A. Holly, and K. Moxey
(Cambridge: Polity Press in association with Basil Blackwell. 1991), pp. 101-50.
12 Axel Bolvig, ‘Zur Kritik digitalisierender Geschichtsproduzenten', in Electronic
Filing. Registration, and Communication o f Visual Historical Data. Abstracts for Round
Table no 34 o f the 18th International Congress o f Historical Sciences, ed. by Axel Bolvig
(Göttingen: Max-Planck-Institut für Geschichte. 1995). pp. 1-7.
Cf. Keith Moxey. The Practice o f Theory. Poststructuralism, Cultural Politics, and
Art History (Ithaca and London: Cornell University Press, 1994).
xxvüi A xel B olvig
possibilities for comparative analysis of historical image information.14 Several

of the present articles deal with the use of image databases. Jörgen van den Berg
demonstrates the new potentiality of ICONCLASS as a common indexing
system within a digitized world. Jérôme Baschet argues for an ‘iconographie
sérielle’ which is ‘indispensable pour faire apparaître l’épaisseur historique et la
complexité des phénomènes, dès lors qu’on la conçoit. . . comme la production
de gammes de variations, au sein d’environnements complexes et mouvants’.
Gerhard Jaritz shows how the database Real brings new light to visual
phenomena such as cleanliness and men’s hairstyles. In the image database we
are confronted with connections of the qualitative and the quantitative, opening
up possibilities of searching for patterns, their role, rules, and connotations. Axel
Bolvig uses the database of Danish wall paintings to demonstrate that in the
pictorial world wooden shoes were not used by peasants but by evil people and
devils, and that they connote rebellion.
The growing acceptance of a unique and individual visual meaning
independent of the linearity of linguistics is stressed by Lena Liepe when writing
that ‘the main potential of images as historical sources lies precisely in their
visual formation’. Ulla Haastrup demonstrates the historical power of images.
She challenges the historians’ positivistic belief in the superiority of written
material when using images as evidence of the existence of Jews in medieval
Denmark in spite of written silence. Spreti Kaspersen in his ‘Framing history
with salvation’ indicates a medieval, politically tinted, connotative use of visual
language in detailed programs. With three different examples Frank Colson, Jean
Colson, Ross Parry, and Andrew Sawyer demonstrate a superiority of the image.
‘When we are considering power (to use Bryson’s phrase) “not as a monolith,
but as a swarm of points traversing social stratifications and individual
persons”,15 it appears to be (in these three examples at least) the image that
thrives as a representational framework’. This acknowledged superiority of
image to word is behind Francis Haskell’s article on J. Huizinga showing that
The Autumn of the Middle Ages is ‘the first major work of history to have been
inspired, both directly and self-confessedly, by visual sources’.
An established acceptance of images as invaluable material in our
understanding of the past stresses another treachery of the image. We do not look
at the same pictures as did our forefathers. The content of a text is the same
whether it is written by hand or printed in Palatino or in Times. This means that
we can copy a text without changing the content. But we cannot copy an
14 Gerhard Jaritz, ‘Comparative Analysis of Historical Image Information’, in

Electronic Filing, Registration, and Communication of Visual Historical Data (as note
12), pp. 29-33.
15 N. Bryson, ‘Semiology and Visual Interpretation’, in Visual Theory: Painting and
Interpretation (as note 11), pp. 61-73 (p. 71).
Introduction X XIX
image—we create a new image with a resemblance to the original.16 Each

individual picture is the result of an inevitable connection between form and
content. A change in the former involves a change in the latter. Therefore we are
forced to reconsider the physical conditions of pictures. Rolf-Jürgen Grote and
Annette Hornschuch write about the use of IT and the need for interdisciplinary
research in ‘Innovative hybride graphische Systeme zur Denkmalüberwachung':
Selbstverständlich können die im Verlauf der Forschungsprojekte entwickelten
oder geplanten rechnergestützten Verfahren der Visualisierung und Dokumenta
tion zunächst nur vorläufige Interpretationshilfen geben. Sie sind als Service-
Leistungen aufzufassen, die einem interdisziplinären Untersuchungsteam den
Einstieg in vertiefende Zustands- und Schadenserfassungen erleichtern und erste
Hinweise für Konservierungs- und Restaurierungsmaßnahmen ermöglichen.
The title of Phillip Lindley’s article is significant and explains why research
in images is such a broad-spectrum project: ‘The Lincoln CD-ROM Project:
history, theory, conservation, and images’. The chief objective of the Lincoln
Project was to document the evolution of conservation and the progress of
conservation work. The intention behind this volume is to demonstrate an
interrelated fan of approaches to understanding the visual world of the past.
To stress the diversified but exiting instability of visual communication I
cannot resist quoting Against Art and Artists by Jean Gimpel. As its struggle
against the Reformation developed, Rome decided to use art as a means of action
against Protestant heresy. Art and artists were to be indoctrinated, and their work
supervised. In 1573, Paolo Veronese was summoned to appear before the Holy
Office because of his Last Supper. He had continued to paint religious subjects,
ignoring the instructions of the Council of Trent. Into his picture he had
introduced all kinds of human and animal figures, all of which the Holy Office
considered incompatible with the gravity of the subject.
Inquisitor: Does it seem to you proper in a painting of the Last Supper of our Lord
to represent fools, drunken Germans, dwarfs and other trifles?
Veronese: No.
Inquisitor: Why did you do it?
Veronese: I did it with the idea that these people were outside the room where the
supper was taking place.
Inquisitor: Do you know that in Germany and other places infected with heresy
they are in the habit of using various paintings full of low subjects and similar
inventions to tear to pieces, vilify and make fun of the things of the Holy Catholic
Church, in order to teach false doctrine to foolish and ignorant people?
Veronese: The Commission was to decorate the picture as 1 thought fit. It is large
16 Spreti Kjprup. ‘Billcdkommunikation’, in Visuel Kommunikation, vol. 1. ed. by Bent

Fausing and Peter Larsen (Copenhagen: Medusa, 19S0). pp. 58-82.
XXX A xel B olvig
and can hold many figures. . . . If there is space left in a picture I decorate it with
invented figures... . We painters claim the license that poets and madmen claim.
The inquisitors were not satisfied with Veronese’s explanation. They ruled that
within three months he must correct and amend his picture.
Paolo Veronese did not correct his picture. He changed the title. The Last
Supper became The Feast in the House of Levi.'1
Who is treacherous towards images? In my opinion the many possible
answers lie in a new Iconology.
17 Jean Gimpel, Against Art and Artists, rev. edn (Edinburgh: Polygon. 1991), pp. 55-
56. This is a translation of the original French, Contre Tart et les artistes, ou la naissance
d'une religion (Paris: Seuil, 1968).
Art and history: the legacy o f Johan Huizinga
t F r a n c is H a s k e l l '
he present reputation of Johan Huizinga is a curious one. His masterpiece,
T The Autumn of the Middle Ages, remains in print, as it has done more or
less ever since it first appeared over seventy years ago,1 and a casual
glance at paperback bookshops in the U.K., France, Italy, and Germany suggests
that foreign translations are still in strong demand (can one say the same of any
other major historical work of our century?); in his native Netherlands a new
research institute has just been named after him and, on a recent visit to
Amsterdam, I got the impression that just about half the student population there
was engaged in investigating some aspect of his achievement; and any current
book or article which refers to him, invariably, and as a matter of course,
attaches the adjective ‘great’ to his name. And yet scarcely a single modern
scholar accepts either the methodology or the conclusions on which his
reputation is based. Above all, as far as I can see, no art historian now agrees
with the central tenet of his book to the effect that the magnificent art of
fifteenth-century Flanders—the art of Jan van Eyck and Rogier van der
Weyden—represents not an innovative ‘Renaissance’ but an autumnal, if
ostentatious, decline from the Middle Ages. Just how that art did emerge remains
very controversial, but Huizinga’s view has either been explicitly rejected—as
Sadly, Francis Haskell died before he was able to revise this essay; minor
emendments have been made by the editors with the assistance of Nicholas Penny.
! I will be referring throughout this article to the new, complete English translation of
the book by Rodney J. Payton and Ulrich Mammitzsch. published by The University of
Chicago Press in 1996. despite the fact that, as 1 have written in The New York Review of
Books (4 April 1996). very occasionally the English version produced by F. Hopman
(London: E. Arnold. 1924). under Huizinga’s supervision, does seem to incorporate new
ideas which are of real value.
4 F rancis H askell
by the distinguished connoisseur of Northern Art, Horst Gerson, in an otherwise

appreciative appraisal2*—or (more usually) totally ignored.' By contrast, Panofsky
in a note in his Early Netherlandish Painting, which, ever since its publication in
1955 has been recognized as constituting a milestone in studies of the period,
does refer with respect to Huizinga’s achievement.
Nevertheless, one would not realize that at least one theory in that book—a
theory of crucial importance to Panofsky—may well have been suggested to him
by a reading of Huizinga. Indeed, perhaps this was not something that Panofsky
himself would have acknowledged. The theory I have in mind is, significantly,
just the one in Early Netherlandish Painting, which has been most trenchantly
criticized and ultimately dismissed by art historians: the theory of "concealed or
disguised symbolism’. As is well known. Panofsky argues that the conquest of
natural appearances by Jan van Eyck and some of his contemporaries was so
effective that it carried with it the supreme risk of destroying what had hitherto
been the principal purpose of a picture: ‘to instruct, arouse pious emotions and
awaken memories’.4 A painting such as Jan van Eyck’s Portrait o f Giovanni (?)
Amolfini and his Wife, in the National Gallery in London (fig. 1, plate 1). might
be thought of as devoid of meaning in that sense,5 and a way had therefore to be
found of reconciling ‘the new naturalism with a thousand years of Christian
tradition’. That way was to be the conscious application of disguised symbolic
interpretations ‘to each and every object, man-made or natural’. Thus when we
look at van Eyck’s Lucca Madonna in Frankfurt (fig. 2, plate 2) we—and still
less fifteenth-century spectators—are not invited to be awe-struck by the
apparently natural positioning of the Virgin and Child seated under a three-
dimensional canopy in a simply furnished room around which we could find our
way without difficulty, in contrast to the stylized and artificial conventions
adopted by Flemish artists in the depiction of similar scenes a generation or so
earlier, such as Broederlam’s Annunciation and Visitation in Dijon (fig. 3).
Panofsky emphasizes instead the significance of the candlestick which
2 H. Gerson, ‘Huizinga und Kunstgeschichte', in Johan Huizinga 1872-1972, papers

delivered to the Johan Huizinga conference. Groningen 11-15 December 1972. ed. by W.
R. H. Koops. E. H. Kossmann. and Gees van den Plaat (The Hague: Nijhoff, 1973), pp.
207-22.
' See, for instance, Jan Bialostocki, L'art du XV siècle, des Parler ¿i Durer (Paris:
LG F. 1993).
4 Erwin Panofsky. Early Netherlandish Painting (Cambridge. MA: Harvard University
Press, 1955). p. 141. For Panofsky’s reference to Huizinga, see p. 74. n. 2.
5 See National Gallery Catalogues: The Fifteenth-Century Netherlandish Schools, ed.
by Lome Campbell (London: National Gallery Publications. 1999), pp. 174-211.
A n amt history: the legacy of Johan Huizinga 5
Fig. 3. ‘Annunciation and Visitation Melchior Broederlam. Tempera on wood,

167 x 125 cm. Musée des Beaux-Arts, Dijon. © Musée des Beaux-Arts. Dijon
6 F rancis H askbll
supposedly denotes the Virgin (who supports her child as the candlestick
supports the candle); of the glass carafe and the basin, which serve for ‘an
indoors substitute for the most typical symbols of the Virgin’s purity ‘the
fountain of gardens’ and ‘well of living waters’ of The Song of Songs’; and of
the four lions on the armrests of the Virgin’s throne which ‘bring to mind’ the
Throne of Solomon, as described in The Book of Kings; and so on/’The effect of
all this is to neutralize the impact of van Eyck’s vivid and real novelty and to
make the painting appear far more ‘medieval’ than it strikes us at first sight.
Indeed, Panofsky points out that van Eyck is still making use—though on a far
more systematic and sophisticated basis—of the same type of symbolism that
Broederlam had earlier employed in his Annunciation, where, for instance, the
three windows on the cornice allude to The Trinity and the circular structure
behind the Virgin’s shrine has been designed to look like a tower because this
was a recognized symbol of chastity.6 7
The case that van Eyck was not the ‘realist’, and hence ‘progressive’, painter
so admired by the nineteenth century had, of course, been made very forcefully
by Huizinga, who had stressed, again and again, the overriding importance of
symbolism in his art and in that of his contemporaries—a symbolism that had by
now lost its original significance and served only ‘to give concrete shape to
every conception. Every thought seeks expression in an image, but in this image
it solidifies, and becomes rigid’.8 For Huizinga, this feature of fifteenth-century
Flemish art had been characteristic of the decline of an overripe civilization
rather than of the lusty birth of a new one. Panofsky’s conclusion is not, of
course, the same as this, but its implications do seem to me to be close enough to
it to warrant the hypothesis that the two could have been related—and I was
delighted to come across the suggestion made by Gerson (who was writing at a
time when the concept of ‘disguised symbolism’ was still generally accepted)
that Huizinga would have been gladdened by Panofsky’s theory had he been able
to know of it.9When reading Panofsky’s emphasis on the intellectually designed,
rather than realistically observed, features of van Eyck’s Virgin and Child in
Frankfurt, and similar pictures elsewhere, I cannot help recalling that Huizinga
had repeatedly claimed that the horror vacui, the over-elaborate detail, the
exaggerated finish in the paintings of van Eyck and his contemporaries signified
an end, rather than a beginning—often magically beautiful, but sometimes
pompous and artificial. Well before Panofsky had singled out the deliberately
6 Panofsky (as in note 4), pp. 137, 143-44.

1Ibid., p. 132.
8 For the following analysis of Huizinga, see (where not otherwise indicated) Francis
Haskell, History and its Images (New Haven: Yale University Press, 1993). pp. 468-95.
9 See Gerson (as in note 2). p. 212.
Art and history: the legacy o f Johan Huizinga 7
Fig. 4. ‘Singing angels’, Jan van Eyck. The Ghent altarpiece. Oil on panel.
Cathedral of St Bavo, Gent. © Cathedral of St Bavo. Gent
s F rancis H askell
archaizing and contrived elements in the Virgin and Child in Frankfurt, Huizinga
had warned his readers not to be misled by the Singing Angels in the Gent
altarpiece (fig. 4) who appear to be so straightforwardly and freshly rendered:
‘the heavy garments, all dark red and gold with sparkling gems, the overly
emphasized, distorted face, the somewhat pedantic ornaments of the music
stand— all are the painterly equivalent of the dazzling bombast of the literary
Burgundian court style’.'“
A few words of Otto Pächt’s polemical review of Early Netherlandish
Painting silently demolish the principal thesis behind the great works of both
Huizinga and Panofsky: ‘W hile it is arguable whether one should start the
history of Italian Renaissance painting with Giotto or with Masaccio, no doubts
are permissible as to where the caesura between the Middle Ages and modern
times has to be made in Northern painting’"— for it is, of course, precisely the
sowing of such doubts that characterizes the works of both these cultural
historians.
Cultural historians working outside the field o f painting have been just as
sceptical (perhaps even more) of Huizinga’s views. In particular, a number of
distinguished Italian scholars, of whom the most conspicuous is probably
Eugenio Garin, have barely been able to conceal their resentment beneath a thin
veneer of guarded respect, and it has been plausibly argued that this hostility
may have been aroused by a feeling of disquiet that Huizinga's concentration on
northern Europe (disenchanted though it was) may have distracted attention from
the achievements of the Italian Renaissance as celebrated by Burckhardt—-the
more so as it was the North that was to play a more influential role than Italy in
the creation of the modem world.*12
Within the last few years, two important English studies have gone out of
their way to dispute Huizinga's claim— perhaps the most influential of all his
claims— that by the fifteenth century the notion of chivalry, with its attendant
enthusiasm for tournaments and displays of the most ritualized yet extravagant
nature, had lost its original purpose of training virtuous warriors to defend
Christianity and had sunk into morally meaningless prodigality, and dreamlike
evasion of reality. On the contrary, we are now told, the tournament still retained
its value as a preparation for actual warfare, and in any case the nostalgia for the
chivalric values o f three hundred years earlier— a nostalgia which, it is admitted,
certainly does pervade Burgundian court culture—served to uphold those values
111Huizinga (as note I ), p. 342.

" See The Burlington Magazine (1958). 275.
12 Alberto Varvaro. 'Riconsiderando l’Autunno del Medioevo’, in Filosofia e storia
della cultura. Studi in onore di Fulvio Tessitore, ed. by G. Cacciatore. M. Martirano. and
E. Massimilaa (Naples: Morano, 1998). pp. 785-97.
against later distortions of them."

My reading of modem literature on the religious beliefs and practices of the
time is too scanty for me to judge whether Huizinga’s view that they too had
become sterile and artificial still commands respect, but I do know that he has
been criticized in this field also and that he has certainly been made to pay dearly
for the hostages that he offered to fortune by once admitting that he had made no
serious reference to Joan of Arc on the grounds that she would have upset the
balance of The Autumn o f the Middle Ages.
Most surprising and devastating of all, perhaps, are the couple of paragraphs
which in 1941 the fine French historian Lucien Febvre devoted to undermining
the very basis of Huizinga’s approach—surprising both because Febvre
expressed such warm admiration for the book which, he complained, had been so
'prosaically' translated as ‘Déclin du moyen âge’ and because, only a year later,
he had echoed so closely some of the opening phrases of The Autumn of the
Middle Ages in his own masterpiece. Le Problème de l'incroyance au XVI
siècle—la religion de Rabelais. Thus Huizinga states:
When the world was half a thousand years younger all events had much sharper
outlines than now. The distance between sadness and joy. between good and bad
fortune, seemed to be much greater than for us... Just as the contrast between
summer and winter was stronger than in our present lives, so was the difference
between light and dark, quiet and noise. ...iJ
If we now turn to Febvre we find:
Let us conjure up François Rabelais’s contemporaries—their violence and
capriciousness, their inability to resist surface impressions, their extraordinary
changes of mood, their astonishing quickness to anger, to lake offence, to draw the
sword and then kiss and make up [...]. These are the natural products of a life full
of contrasts, much sharper than we can imagine: the contrast of day and night,
unknown to us in our electrically lit homes; the contrast of winter and summer,
tempered for us in normal times by a thousand devices, whereas they were
subjected to the seasons’ rigours and necessities practically without letup, for
weeks and months at a time...."
Yet despite this homage to the great Dutch historian Febvre questions the very
notion of singling out some particular sentiment—the violent oscillation, for in
stance, between cruelty and tenderness, extreme hatred and extreme goodness,*145
" Malcolm Vale. War and Chivalry (London: Duckworth. 1981): Maurice Keen.
'Huizinga. Kilgour and the Decline of Chivalry’, in Medievalia et Humanística, new
series, no. 8 (Cambridge. 1977). pp. 1-20.
14Huizinga (as note I ). pp. 1-2.
15 Lucien Febvre. The Problem of Unbelief in the Sixteenth Century—The Religion of
Rabelais, trans, by Beatrice Gottlieb (Cambridge. MA: Harvard University Press. 1982), p.
99.
IO F rancis H askell
which exudes ‘the mixed smell of blood and roses’—as having been
characteristic of the decline of the Middle Ages. Cannot the same sentiment be
detected just as clearly at the dawn of the Middle Ages—or, indeed, at the dawn
of modern times?16
Some recent writers have indeed gone much further than this. Thus in a subtle
and admirable study of the imagery associated with the early years of François
r of France, the French art historian Anne-Marie Lecoq quite often finds
Huizinga’s concepts relevant to her discussion, despite the fact that the subtitle
of her book contains the phrase l ’aube de la Renaissance, which seems to pose a
deliberate challenge to the Autumn of the Middle Ages. On one occasion she even
claims that Huizinga’s analysis of a popular feature of decaying Burgundian
rhetoric—the use of ‘reverse symbolism' whereby ‘the lower does not point to
the higher, but rather the higher to the lower, since, in the mind of the inventor,
the earthly things that he intends to glorify with some heavenly ornamentation
are foremost’—is more applicable to the court of Henri IV early in the seven
teenth century than to that of François Ier in the sixteenth.17 The case against
Huizinga is thus formidable: and yet he lives.
Are we to assume that this is because he belongs with those great, imaginative
and self-confident pattern-makers whom we associate mainly, but not
exclusively, with the nineteenth century: men such as Burckhardt and Taine and
the anthropologist Sir James Frazer and (dare I say it?) Marx and Freud, who
have—often with spellbinding virtuosity—constructed syntheses, whose inner
consistency continues to mesmerize us even after the experts have convincingly
proved to us that they are frayed at the edges and often decayed at the core? To
some extent, this is certainly true of the work of Huizinga. And it is also all too
easy to understand why, ever since 1914, Europeans and Americans should,
however reluctantly, be particularly attracted by the poignant spectacle of a
decaying but flashy culture—especially when that spectacle is evoked with
incomparable poetry, as Dutch colleagues assure us is the case of The Autumn of
the Middle Ages in the original.
And yet the survival of the book depends on much more varied causes than
that, and I want to devote the remainder of this essay to the two features of
Huizinga’s work that strike me as still being both original and challenging—and
to show, incidentally, how both of them have indeed also been challenged.
The Autumn of the Middle Ages is, I believe, the first major work of history to
have been inspired, both directly and self-confessedly, by visual sources—and
here I must at once apologize for the need I will frequently have to succumb to
16 Lucien Febvre. ‘Comment reconstituer la vie affective d'autrefois? La sensibilité et

l’histoire'. Annales d'Histoire Sociale, 111 (1941). reprinted in Combats pour l’histoire
(Paris: A. Colin, 1953). pp. 221-38 (pp. 226-27).
17 Anne-Marie Lecoq, François I" imaginaire: Symbolique & politique à l’aube de la
Renaissance française (Paris: Macula. 1987). p. 353 and elsewhere.
from now on: that is, to draw on a recent book of my own—History and its
Images—that tries to explore the strengths and weaknesses of the use of such
sources by writers ever since the Renaissance. For it is, of course, true that, long
before Huizinga, much historical writing had, to a greater or lesser extent, been
significantly affected by responses to the visual arts. In 1864, Hippolyte Taine
had planned ‘to study Italian history by making use of painting for my
documentation instead of written sources’, perhaps aware of the fact that some
ten years earlier John Ruskin had already claimed to do just that in The Stones of
Venice. I wish that I knew enough about Huizinga’s development to discover
whether or not he had (as seems likely) read this astonishing and perverse
masterpiece and, if so, whether he remembered from it the brief, but eloquent,
passage in which Ruskin recalls visiting Venice before the days of the railway
and being so ‘entranced by the visionary charm of a scene so beautiful and so
strange’ that he feared running the risk of forgetting that the past of the city had
been marked by dark cruelty and fear and that the evidence of art could therefore
be misleading. Above all, I would like to know whether Huizinga was
acquainted with a famous aphorism of Ruskin’s that seems to me to encapsulate
much of the methodology that was to serve him for The Autumn o f the Middle
Ages: ‘Great nations write their autobiographies in three manuscripts: the book
of their deeds, the book of their words, and the book of their art. Not one of these
books can be understood unless we read the two others; but of the three, the only
quite trustworthy one is the last’. For reasons that will later become obvious I
want to draw particular attention to a few words of that phrase: ‘Not one of these
books can be understood unless we read the two others’: we will return to their
implications.
The Stones of Venice can, however, hardly be described as a work of history,
however great its influence on the writing of history may or may not have been,
and we must turn now to the explanation provided by Huizinga himself of the
inspiration for his own spiritually akin, but far more coherent, masterpiece.
Writing in the very last years of his life he tells us that although he had, as a
child, been fascinated by the Middle Ages and, as a student, by the visual arts, in
later years his professional interests had moved to the remote field of ancient
Indian theology and literature until, in 1902, he paid a visit to the remarkable
exhibition of Flemish Primitives held that year in Bruges. Some five years or so
later, on a Sunday afternoon between 1906 and 1909, probably 1907, ‘when my
wife’s time was taken up by looking after the young children’, it occurred to
him, in a flash, while walking through the countryside, ‘that the late Middle
Ages were not the herald of something new that was to come, but the fading
away of something that had already passed. This thought, if it can be called a
thought, revolved above all around the art of the van Eycks and their
contemporaries, which considerably occupied my mind at that time’ and which
was generally being interpreted as the dawn of a Northern Renaissance. ‘My
perception was directly opposed to this. Some years passed, however, before I
12 F rancis H askell
began to work it ou t . . . [and] I began in 1910 to read as much as possible of the

Burgundian and French historians of the period..
It is the chronology of these stages that I want to stress. If we are to trust
Huizinga, he saw the early Flemish pictures in 1902, reacted against the
prevailing view of them in 1907, and only in 1910 began to read the literature of
the time, which—nine years later—led to The Autumn of the Middle Ages. In
other words it was the visual source that stimulated his interest in the written
ones—and not the other way round. I emphasize this as strongly as I can,
because in a very famous and enthralling lecture (at which I had the honour to be
chairman and which was subsequently published as book called In Search of
Cultural History) Ernst Gombrich argued in 1967 that instead of starting 'from
the impression [that] van Eyck’s paintings made on him’, Huizinga 'took the
opposite line. He simply knew loo many facts about the age of van Eyck to find
it easy to square his impression of his pictures with [what he knew of the
culture]. He felt he had rather to re-interpret the style of the painter’ so that van
Eyck’s realism ‘could no longer be seen as a harbinger of a new age; his jewel
like richness and his accumulation of detail were rather an expression of the
same late Gothic spirit that was also manifested much less appealingly in the
prolix writings of the period which nobody but specialists read any more’.
But if Huizinga is telling the truth about his development—and I cannot see
why we should doubt that he is—it was only after he had had the perception
(whether valid or not is another matter) that Flemish painting was affected, over-
refined, and fussy that he followed up the literary sources and chronicles which
led him to conclude that the whole period was one of decline. It would therefore
naturally be of interest to examine for ourselves just what were the pictures that
made this impression on him at the decisive Exhibition of 1902—and, to some
extent, we can do this thanks to surviving catalogues, descriptions and
photographic records. It would be pedantic, absurd in fact, to pursue this line of
enquiry too far, because quite obviously he had plenty of later opportunities to
see Flemish pictures elsewhere, but we are all the same entitled to believe that
the first, overpowering, revelation came to him at the Bruges Exhibition, so that
it is worth noting that only one certain, and incidentally very small, work was to
be seen there by the artist who is now usually called Robert Campin and
recognized as one of the founders of the new Flemish painting—an artist whose
domestic, almost ungainly, realism conveys so very different an image from the
delicate but luxurious elegance of van Eyck. And although as many as seventeen
pictures were attributed to Rogier van der Weyden in the catalogue, it was
agreed that in none of these cases was the attribution wholly convincing, while it
was in any case not possible to borrow from Spain his large and dramatic
Deposition from the Cross (now in the Prado), which exemplifies just that
boldness of conception and originality of design that Huizinga professed himself
unable to discover in the art of the period. There were, on the other hand, two
Fig. 5. ‘Shrine o f St Ursula'. Hans Memling, detail. Wood. 37.5 x 30 cm. Bruges
Groeninge Museum. © Groeninge Museum. Bruges
14 F rancis H askf.ll
wings of an altarpiece showing no fewer than eight scenes from the legend of the
martyrdom of St Ursula (fig. 5) by an anonymous master of the second half of
the fifteenth century who had at one time been thought to be Memling and more
recently (for easily understandable reasons) has been considered to be a
miniaturist from Bruges working on a larger scale than the one he was
accustomed to. In fact, I get the impression that although the stars of the
Exhibition were van Eyck, Memling, and Gerard David, it was appealing but
relatively minor pictures of this sort, combining picturesque and fanciful
architecture with eminently pretty, though slightly gawky and awkward, figures
engaged in scenes of ceremony and cruelty—pictures of which there were large
numbers—that would have been most apparent to visitors to the Exhibition.
Indeed, by far the most famous of the many genuine Memlings to be seen there
was the small shrine in the form of a tabernacle depicting episodes from the
same legend, painted in the spirit of a chivalrous fairy story of the very kind
which plays so prominent a part in Huizinga’s conception of the Middle Ages. It
is, therefore, tempting to suggest that he was so struck by works of this nature
that he hardly bothered to look at the numerous portraits by Memling that were
also present and that appear to represent sober, devout. God-fearing merchants
who created the riches needed for the exotic pastimes and costumes of court life
but who are of no interest to the author of The Autumn o f the Middle Ages. More
in keeping with the spirit of that book was a large and horrifying picture by
Gerard David of the Flaying of the Unjust Judge (in Groningen) which, although
he does not mention it, must surely have helped to stimulate Huizinga’s
evocation of the atrocious tortures and executions freely carried out in the Low
Countries during the fifteenth century. Similarly, although he does not refer to
Hieronymous Bosch (and, in other writings, tends to play down his importance),
who was represented by a handful of paintings (two or three of which were
authentic), it is hard not to see his so-called ‘diableries’ as a formative influence
on Huizinga’s view of the period as credulous, perversely child-like and
ultimately detestable.
It is thus not unreasonable to take Huizinga at his word when he tells us that
he first formed an impression of Flemish and Burgundian civilization of the
fifteenth century as unbalanced, neurotic, cruel, and fanciful on the basis of the
pictures that he suddenly came across at the Bruges Exhibition in 1902.
But if this rescues Huizinga from one of the traps into which Gombrich claims
he had fallen, it may seem to thrust him still more vigorously into the other—the
fallacy, that is, of adopting the Hegelian viewpoint whereby if the nature of the
art cannot be deduced comfortably from ‘the spirit of the age’ in general, then at
least that ‘spirit of the age’ can itself be deduced from the nature of the art. It is a
reproach that has quite often been made against Huizinga both explicitly and
implicitly—most unsparingly (and misguidedly), I believe, by another even more
encyclopaedic historian of the Middle Ages, the German Ernst Robert Curtius—
an historian of literature and a philologist, who had little if any interest in the
visual arts, but who in many other ways was rather similar to Huizinga. Both
were conservative and pessimistic by instinct, disgusted as well as horrified by
the twentieth century—yet (understandably) reluctant to commit themselves too
far in view of the certain dangers to which such outright opposition would have
led—and, above all, both men were at home with the whole of European
civilization: along with Panofsky among the very last of their kind. I used the
phrase that ‘I believe’ that Curtius had Huizinga in mind when indulging in
rather a muddled outburst in the first chapter of his European Literature and the
Latin Middle Ages although this is not explicit, and he was certainly even more
concerned to correct the influence of Wölfflin,"* Curtius stigmatized the concept
of ‘art history as a super discipline’ and insisted that:
literature has different forms of movement, of growth, of continuity, from art. It
possesses a freedom which is denied to art [...] with the literature of all times and
peoples I can have a direct, intimate, and engrossing vital relationship, with art
not. Works of art I have to contemplate in museums. The book is more real than
the picture. [. . .] To understand Pindar’s poems requires severe mental effort—to
understand the Parthenon frieze does not. The same relation obtains between
Dante and the cathedrals, and so on. Knowing pictures is easy compared with
knowing books. Now if it is possible to learn the ‘essence of Gothic’ from the
cathedrals, one need no longer read Dante.. . .
One of the more surprising features about this very questionable statement is
that Huizinga himself seems to have anticipated—and to have shared—some of
the anxieties that gave rise to it. As early as 1919—well before Curtius—he
himself pointed out that:
our perception [today] of former times, our historical organ, so to say, is more and
more becoming visual. [. . .| The change of our ideas about the Middle Ages is
due less to a weakening of the romantic sense than to the substitution of artistic
for intellectual appreciation.
And by 1940—at much the same time as Curtius—he was deploring the fact that
‘as more and more visual material for the appreciation of the past became quite
generally available, so thinking and writing about the past fell into increasing
neglect’. It is true that he does not mention the fact that it was he himself who,
more than any other historian, had been (unwittingly) responsible for stimulating
such a ‘one-sided aesthetic view’ of the past, but it is also true that no-one less
than he deserved Curtius’s reproach (if, indeed, it was directed at him) of
forgetting that ‘there are essential differences between the book and the picture’.
On the contrary. One of the reasons that makes it difficult to admire the
approach of most later historians to visual evidence lies precisely in the fact that*1
ls See Ernst Robert Curtius, European Literature and the Latin Middle Ages, trans,
from German by Willard R. Trask (Princeton: Princeton University Press, 1990), Chapter
1, especially pp. 11-16.
16 F rancis H askeli.
none of them seems to have taken sufficient account of some of the weaknesses
of his own methodology pointed out by Huizinga and of the dichotomy between
art and literature outlined by Curtius. It was, after all, Huizinga himself, and not
one of his critics, who emphasized how very limited was the material available
to the historian who wished primarily to understand the past on the basis of
visual evidence—an awareness, incidentally, that makes it all the more
astonishing that he paid almost no attention whatsoever to tapestries and
miniatures. None of the recorded bathing and hunting scenes by Jan van Eyck or
Rogier van der Weyden had survived—let alone the elaborate costumes and
ephemeral decorations of the time: in contrast, written records of absolutely
every kind were to be found in abundance. It was Huizinga also who analysed
with a subtlety worthy of Lessing (even if not always convincingly) the differing
expressive possibilities open to the artist and the writer in the Middle Ages.
Love, he argued for instance, could not be depicted adequately in fifteenth-
century painting or miniatures, because it always appeared as hieratic, solemn,
artificial—whereas we know from surviving literature that the sentiment itself
certainly existed. That said, it is only through painting that we learn that people
could respond to the effects of light breaking through darkness or the freshness
of the countryside in spring, because the literary conventions were too contrived
to allow for the expression of such feelings.
And it is through this sort of analysis that Huizinga reaches a conclusion of
absolutely fundamental importance for anyone interested in cultural history:
It is a general phenomenon that the idea which works of an give us of an epoch is
far more serene and happy than that which we glean in reading its chronicles,
documents, or even literature. Plastic art does not lament. [ ...] [Hence] the vision
of an epoch resulting from the contemplation of works of art is always incomplete,
always too favourable, and therefore fallacious.1'’
I doubt if I could work up much enthusiasm for any historian aiming to make use
of visual evidence (as so many historians now claim that they do) who did not, at
the very least, ponder or, or challenge, this penetrating axiom—for it is. of
course, open to challenge—and it brings me to what seems the central issue in
my discussion of why we should continue to take Huizinga very seriously today.
For in certain fundamental ways he differed radically from those great creative
historians and systematizers to whom I compared him a little time ago. Unlike
them, Huizinga was always aware of the potential weaknesses in his own
approach. He was tempted by the notion of an all-embracing Hegelian
synthesis—what cultural historian is not?—but he could not stifle his doubts
about its validity in the face of conflicting testimony. 1 read in a book review
recently that it is now fashionable for scholarly books to shun conclusions and to
end on a note of enquiry. I was dismayed by this comment, partly because I had
19 The Waning o f the Middle Ages (New York: Doubleday Anchor, 1954), pp. 242-43.
Art and history: the legacy of Johan Huizinga 17
thought that, as I grew older, I myself had developed this procedure in the light
of the experience of a lifetime—and it is always disconcerting to discover that
one’s apparently original and deeply pondered contributions to scholarship are,
in fact, merely the consequences of fashion. I do not, in any case, believe that
Huizinga thought of his book in this way, but 1 do believe that, fashionably or
not, we can approach it in such a spirit if we choose to. His Dutch critics—and
he has some pretty ferocious ones*1—have been scathing about his timidity, the
tentative nature of his aesthetic involvement, his vacillations, and his reluctance
to commit himself forcefully enough. I must confess that it is partly these very
weaknesses—if that is what they are—that help to keep The Autumn of the
Middle Ages so stimulating for me. In 1905, many years before he had even
begun to think of embarking on that masterpiece. Huizinga delivered his
inaugural lecture at the University of Groningen and chose as its title The
aesthetic element in historical thought’. In it he argued that it is through images
that we can best approach the past. He pointed to the example of Michelet in
perfecting this method and told a story from Michelet’s history of the French
Revolution that has haunted me ever since I first came across it. Long after the
revolution a young man asked the aged Merlin de Thionville how he had brought
himself to condemn Robespierre. The old man seemed to be regretting what he
had done and then suddenly he arose brusquely: ‘Robespierre’, he said,
‘Robespierre. If you had seen his green eyes, you, too, would have condemned
him as I did". For Huizinga also, the image came first, enabling us to see the past
'more lucidly, more sharply and more colourfully—in short, more historically',
but this was only a first stage, and in The Autumn of the Middle Ages he was able
to create a book which demonstrates, with incomparably evocative power but
also with incomparable critical insight, just what further structures can be built
on foundations whose intrinsic nature would still repay the most serious
investigation.
*’ See especially Pieter Geyl, 'Huizinga as Accuser of his Age', reprinted in

Encounters in History (London: Collins. 1963).
Images and the historian-
J ean -C laude Schmitt
growing number of historians have for some years now been taking an
A interest in art and in the image, an area of research traditionally reserved

for ‘art historians’. This development in the historian’s craft has very
clearly had consequences for both disciplines. It has widened the historian’s field
of study, obliging him to reflect on objects and on values—in particular aesthetic
values—which he has hitherto not been wont to take into account. It has in turn
obliged art historians to think about the achievements and the future course of
their own discipline, and about the social dimension of works of art: moving, one
might say, in the opposite direction from that taken by the historians in response
to their new requirements. It is now accepted that collaboration is not only
desirable but necessary; this present volume is but one example of the trend.
What remain to be laid down are some precise guidelines, to ensure that
successful collaboration bears fruit. Three such guidelines seem to me to stand
out as vital:
1. we must first adopt a historiographical perspective, and ask the proper
questions concerning the developments which have both hindered and made
possible the rapprochement of our two disciplines;
2. we must agree on the methodology to apply to the analysis of images;
' Translated into English by Brian J. Levy. The original first appeared as ‘L’historien
et les images’, in Der Blick auf die Bilder. Kunstgeschichte und Geschichte im Gespräch,
cd. by Otto Gerhard Oexle, Max-Planck-Institut für Geschichte (Göttingen: Wallstein,
1997), pp. 9-51.
20 J ean -C laude S chmitt
3. we must outline the historical problems and issues relating to the role
accorded to the image in social activities.1
H isto rio g ra p h y
Francis Haskell has recently reminded us of the important part played ever since
the Renaissance by images, from ancient coinage to paintings hanging in
museums, in historians’ reconstruction of the past.2 His monograph does not
really develop the theoretical aspect of his subject, nor does he take account of
all the latest developments in historical research; it is, however, an extremely
rich work and reveals very well indeed how historians, from the beginnings of
their discipline, have hesitated between three possible standpoints.
Some have sought (or indeed still seek) to find in images a more or less
faithful representation—and thus more or less acceptable to historical scholar
ship—of social realia, whether it be the portrayal of warfare (the scenes in the
Bayeux Tapestry, for example), of agricultural techniques (drawn from illus
trated calendars in manuscripts), or medieval domestic interiors (based on certain
Nativity scenes). It cannot be denied that some at least of these images are of
documentary relevance; however, this direct appropriation of images by some
historians tells us nothing of the images themselves, of their raison d'être, or,
above all, of the far more complex nature of the whole process of
‘representation’.3 It is dangerous self-deception to assume that for people in the
past (or, for that matter, for us today) a sense of what was ‘real’ could exist
independently of their attitudes as performers on the social stage, or of the way
in which they might portray it in their works. The problem with this ‘positivist’
approach is that it gives the illusion that all art (at least until the arrival of
abstract art) has a referential function, quite distinct from its other possible func
tions, and thus capable of being isolated by the historian. To be sure, an image is
11 am here adopting the point of view of a social historian who, for the past decade and
in company with other historians and art historians, has been reflecting on the status and
functions of the image in medieval culture, under the aegis of the Western Medieval
Historical Anthropology Group at the Paris Ecole des Hautes Etudes en Sciences Sociales.
This research project has been essentially a collective one: my gratitude is due to my co
researchers, friends, and colleagues J. Baschet, J.-Cl. Bonne. A. Debert, and M.
Pastoureau.
2 Francis Haskell, Histoiy and its Images. Art and the Interpretation of the Past (New
Haven and London: Yale University Press, 1993).
3 A survey of recent scholarship in this area may be found in Jean-Claude Schmitt,
‘Representations’, in Georges Duby. L ’écriture de l'Histoire, ed. by Claudie Duhamel-
Amado and Guy Lobrichon, Bibliothèque du Moyen Age. 6 (Brussels: De Boeck-
Université, 1996), pp. 267-78.
Images and the historian 2!
always the image of something; but it is this fact that may delude us into thinking
that we merely have to state what the image represents in order to say everything
about its representation. In any case, this is beside the point; time and again, the
images themselves are perfectly capable of reminding us that their function is less
that of ‘representing" an outside reality than of adapting this reality to their own
particular mode.4 For the historian, it should be a matter less of isolating and of
‘reading’ the contents of an image than of appreciating first and foremost that
image in its totality: fonti and structure, functions and functioning.
For other historians, art bears witness to a Zeitgeist on which, by analogy with
the styles familiar to art history, they bestow the names ‘Romanesque’, ‘Gothic’,
‘Flamboyant’ or ‘Baroque’. They do not, however, take the trouble to analyse
any actual works; generally speaking, they barely look at them. Francis Haskell
calls them ‘philosophers' (as opposed to ‘antiquarians’) and has gone so far as to
include among their number Jules Michelet, Jacob Burckhard, and Johan
Huizinga. As far as the latter is concerned, he notes that he very rarely ever turns
to a work of art to depict a specific aspect of medieval society, with the
exception of Chapter Nine of The Autumn o f the Middle Ages, devoted to ‘the
vision of Death". While praising Huizinga for having been in his opinion the last
great historian to make a major contribution to the subject of history and its
images, and for recognizing that through the image we can see the past ‘more
lucidly, more sharply and more colourfully, in short more historically’, Haskell
criticizes him for reverting in the final analysis to the traditional prejudice of
devaluing the importance of images, compared with texts, because ‘works of
literature offer one more criterion than do works of art: they make it possible for
us to appreciate the spirit as well as the form’.
The most common attitude has surely consisted, above all, of one long
process of avoidance, between history (social, political, and even cultural) and
art history, with each enjoying divergent lines of development. This avoidance is
not fortuitous. It has its roots in the privileged position of the tongue, of speech,
over all the other symbolic functions of mankind. It stems from the constitution
of history as a ‘literary’ discipline in the eighteenth century, and as a positive
science in the nineteenth, with pride of place given to written documents, judged
to be more objective and reliable than pictures.
In exactly the same way as positivist history, based on ‘the text’, was able to
align itself with archaeology, whose irrefutably objective foundations were laid
down by the very material nature of the objects and inscriptions studied, so art
history, all preoccupied with the appreciation of ‘masterpieces’, styles, and
4 The ability of images, particularly in the modern period, to signify that the only
reality they portray is that of representation has recently been studied by Victor 1.
Stoichita, L 'Instauration du tableau. Métapeintiire à Taube des Temps modernes (Paris;
KJincksicck, 1993): and by Klaus Krüger, ‘Der Blick ins Innere des Bildes. Aesthctische
Illusion bei Gerhard Richter", Pantheon, 53 (1995), 149-66.
22 J ean -C laude Schmitt
artists, found itself comfortably associated with philosophy, its mission being to
determine ‘Beauty’, in all its timelessness. As a result, in the 1888 inaugural
lecture of his Archaeology course, Charles Diehl was able to draw a distinction
between philosophy and art history on the one hand, with their seeking after the
‘laws’ governing universal aesthetics, and, on the other hand, history and
archaeology, standing shoulder to shoulder and studying ‘facts’. On art histo
rians, he said: ‘An ideal of supreme beauty serves as their yardstick when they
come to study and to judge every age’s particular ideal of beauty’. While ‘art
history is the close, and inseparable, kin of philosophy and aesthetics’, in
contrast ‘archaeology is linked to history, and is its indispensable and utterly
reliable adjunct’.5
A little later, following on from Alois Riegl (Stilfragen, 1893), Heinrich
Wölfflin (Kunstgeschichtliche Grundbegriffe, 1915), and Henri Focillon (La vie
des formes, 1934), the priority given by art historians to form and its sui generis
development could only serve to widen still more the gulf between their artistic
Heavenly Spheres and the (as it were) down-to-earth preoccupations of social
and political history. Art history had its own ‘cycles’ (from the Classical to the
Baroque), quite independent of the chronology of reigns or social systems.
‘Styles’ were simply a matter of ‘schools’, ‘workshops’, or ‘influences’ —even
of ‘national traditions’6—and there was hardly any attempt to break out of the
closed province of art history.7
For anyone seeking to do more to link artistic development with that of
society as a whole, such a relationship could only be achieved in a metaphorical
or reductionist sense. Thus, Emile Mâle, stating once and for all that thirteenth-
century religious art (with the exception of ‘purely decorative work’, which he
deliberately discarded from his field of study, on the grounds of its lack of
‘symbolic meaning’)8 could be compared to a ‘book’, their aim being to make
Church teaching visually manifest, particularly to the illiterate masses. From this
standpoint he was able to distinguish levels of meaning and function, by analogy
with the Speculum of Vincent of Beauvais: the Mirror of Nature, the Mirror of
5 Cited by Jean Hubert, ‘Archéologie médiévale’, in L 'Histoire et ses méthodes, cd. by

Charles Samaran (Paris: Gallimard - Pléiade, 1961), pp. 275-328 (317-20).
6 In some cases, this involved all the ideological drift only too well illustrated in
nineteenth-century history. See the frank study by Eric Michaud, 'Nord-Sud (Du
nationalisme et du racisme en histoire de l’art. Une anthologie)’, Critique, no. 586 (March
1996), 163-87.
7 Louis Pressouyre, ‘Histoire de l’art et iconographie’, in L ’Histoire médiévale en
France. Bilan et perspectives, Société des Historiens Médiévistes de l’Enseignement
Supérieur; edited by Michel Balard; with a preface by Georges Duby (Paris: Le Seuil,
1991), pp. 247-68.
8 Emile Mâle, L'art religieux du XIIIe siècle en France (Paris: Armand Colin, 1898;
new edn., Paris: Le Livre de Poche, 1948), p. 24, n. 5.
Images and the historian 23
Science, the Mirror of Moral Instruction, and the Mirror of History.

Nowadays we can see the obvious conceptual limitations of such a project: an
entire literate culture cannot be seen through images, any more than the image
can be reduced to just a reflection of culture. In its time, however, this enterprise
proved highly tempting to those contemporary historians who were conscious
that art was an important card to play in general cultural history, yet who (by
virtue of their training as historians) felt ill-equipped to embark upon such a
study. Emile Mâle’s synthesis at least seemed to deal with medieval art in cate
gorical terms which they found intelligible.
Marc Bloch, for example, referred to Mâle when he made his own suggestions
concerning the parallel development of society and contemporary artistic
achievements. This great historian’s ambiguous attitude towards medieval art has
recently been well noted by Ulrich Raulff.9 On the one hand, history, above all
rural history, was for him a science of observation, to be carried out by means of
maps, plots, and aerial photography, which he had learned to use during the
Great War. He also took a close interest in Lefebvre des Noëttes’s research on
the Bayeux Tapestry, relating to the history of military costume. On the other
hand, Marc Bloch might be considered representative of that type of French
intelligentsia who have been accused of being less than sensitive to visual
stimuli.1012 However that may be, Bloch effectively confesses his lack of
competence, in those pages of his 1939 La société féodale devoted to ‘Façons de
sentir et de penser’, in which—and here is where he cites Emile Mâle—he
stresses the need for the historian to look into society’s ‘expression plastique’,
but hesitates himself to set out on such a hazardous path." Nonetheless, for his
projected 1933-34 collection of essays, he had envisaged grouping a number of
items under the rubric ‘Images et imagination collective’, which would have
included inter alia his fine article on ‘La vie d’outre-tombe du roi Salomon’,
originally published in 1925, a year after Les rois thaumaturges.'2 In both article
and monograph, Bloch amply demonstrates his skill in looking very closely at
images and in understanding their ideological impact, and he even sketches a
comparative history of Western Christian and Byzantine images. But he went no
9 Ulrich Raulff, Ein Historiker im 20. Jahrhundert: Marc Bloch (Frankfurt am Main:
Fischer, 1995), p. 93.
10 This is the theory put forward by Martin Jay, Downcast Eyes. The Denigration of Vision
in Twentieth-Century French Thought (Berkeley: University of California Press, 1993).
11 See Jean-Claude Schmitt, ‘Façons de sentir et de penser. Un tableau de la
civilisation ou une histoire-problème?’, in Marc Bloch aujourd'hui. Histoire comparée et
sciences sociales, ed. by Hartmut Atsma and André Burguière (Paris: Ecole des Hautes
Études en Sciences Sociales, 1990). pp. 407-18 (408).
12 This proposed collection, containing this article, has at long last been published:
Marc Bloch. Histoire et historiens. Textes réunis par Etienne Bloch (Paris: Armand Colin,
1995), see in particular pp. 167-90.
24 J f.an -C laudf. S c hmitt
further down this path, nor did he really pursue the ethno-historical approach
illustrated by his book. In both cases markers were laid down, but medievalists
were only later to follow them.
The harmful effect of the lack of communication between the developing
disciplines of history and the history of art respectively may be well measured by
the fact that, at this precise time when Marc Bloch was wishing devoutly for a
proper analytical use of images by historians, art history, first in Germany, then
in Great Britain and the U.S.A. (with Aby Warburg and his successors, headed
by Fritz Saxl and Erwin Panofsky), was undergoing an unprecedented conceptual
sea-change, but one which took a long time to influence the preoccupations and
methods of historians. It would appear that Marc Bloch himself failed to note
this transformation in the fields of theory and methodology.1314
However, our modern retrospective judgement of the Warburg heritage and of
the New History bom in France with the setting-up of the Annales School, does
enable us to appreciate the extent to which intellectual projects, albeit belonging
to different cultural and linguistic traditions, were sufficiently analogous as to
allow for future collaboration. Still, such collaboration was by no means
automatic: disciplinary conservatism, and the legitimate priority accorded by
historians to other matters (in particular, in the fields of economic and
demographic history), have for a long time had the opposite effect of keeping
problems and issues apart, instead of enabling them to come together. In another
area, however, in the current configuration between human and social sciences,
some concepts—despite their long-standing definitions—have shown themselves
to be highly heuristic, and highly efficacious.
Such is the case of the concept of ‘symbolic form’, which Erwin Panofsky
borrowed from the philosophy of Ernst Cassirer and applied historically to the
question of plane perspective.IJ Panofsky postulated a decisive break with the
history of styles, which was supposed to offer very little by way of interest to the
social historian. He rejected the deduction that plane perspective stemmed from
the subjective psycho-physiological experience of vision: on the contrary, he
stressed the absolute abstraction of the nature of its twin constituent principles
(the vanishing-point to infinity, and the homogeneity of the field of representa
tion). He also saw the origins of the discovery of perspective in the context of
advances across the range of intellectual history (in optics and mathematics),
rather than in the single field of art history; and finally, he showed the effects of
perspective in turn on the rationalisation of subjective sight, including the
This fact makes Haskell’s point (History and its Images (as in note 2), p. 8 and n. 1,
p. 496) that the Annales School did not as a whole pay much attention to the importance
of art for the historian. All this was to change for a later generation, as witness the work
of Georges Duby, for example.
14 Erwin Panofsky, La perspective comme forme symbolique et autres essais. Précédé
de ‘La question de la perspective par Marisa Dalai Emiliani (Paris: Minuit, 1975).
Images ami lhe historian 25
extreme visionary experience of mystics. In this way Erwin Panofsky made an

essential contribution to the promotion of perspective as a subject of the ‘general
history’ which informs all our current efforts. The aim of this ‘general history’,
so ardently desired by Michel Foucault, is to articulate all the various ‘symbolic
forms’ governing the functions of a society at any given period.
This is precisely the way in which Warburg’s notion of ‘Kulturwissenschaft’
broke with the concept of a ‘cultural history’ still linked (even for Karl
Lamprecht) to the ideas of progress and evolution, or to a strictly national frame
work.15 In its origins at least, it really had nothing about it of the purely
intellectual history, which it would become for Panofsky, in whose hands it
would tend towards a reduction of the ‘iconological’ element and towards a
study of those intellectual traditions which would enable all a work’s multiple
meanings to be deciphered. Initially, the notion took in the various aspects of a
society’s ‘Weltanschauung’: all ‘the issues of social history and social psy
chology, of society and politics’.16 It is impossible not to recognize here some
common features with the twentieth-century’s other great landmark in the field
of historiography: the contemporary, but totally (or mainly) independent work of
the Annales School.
One of the most famous projects, and one which indeed engaged all the
attention of the Annales historians, was Erwin Panofsky’s attempt to incorporate in
one single explanation the principles of Gothic architecture and Scholasticism.17
His point of view is structural and relational, seeking to uncover, in two
contemporary, but otherwise apparently heterogeneous phenomena—the
University disputation and the system of Gothic vaulting—common principles of
clarification and conciliation of opposing forces. This attempt is, however, open
to criticism, for being still organized around the idea of a basic parallelism
between the various forms of human thought: figurative and mythic, architectural
and philosophical. In a ‘general history’, there does not always have to be
harmony between the different ‘symbolic forms’ or the different ‘languages’ of a
given society: there may also be contradictions between them, or even conflict,
in historical circumstances where unity will always be a problematic factor.18
The problems of the specificity of art, and of the sometimes conflicting relations
15 See Otto Gerhard Oexle, in the Introduction to Memoria als Kultur, ed. idem
(Göttingen: Vandenhoeck & Ruprecht. 1995), p. 25.
16 Michael Dicrs, "Mnemosyne oder das Gedächtnis der Bilder. Über Aby Warburg', in
ibid., pp. 79-94 (87).
17 Erwin Panofsky, Architecture gothique et pensée scolastique, précédé de l ’Abbé
Suger de Saint-Denis. Traduction et postface de Pierre Bourdieu. 2nd edn (Paris: Minuit,
1967).
IS See Hubert Dänisch, "Art (Histoire de 1')’, in La Nouvelle Histoire, ed. by Jacques
Le Goff, Roger Chartier, and Jacques Revel (Paris: Retz-C.E.P.L., 1978), pp. 68-77.
between the different ‘symbolic forms’ of a single society, should be at the very
forefront of the work of the ‘image historian’.
M ethodology
Our historian is thus faced with the double challenge of analysing art both in its
own right and in its dynamic relationship with the society which produced it.
The work, in France, of someone like Pierre Francastel, a sociologist of art
(hostile to all pseudo-sociology) and a meticulous analyst of figurative thought,
and, in the United States, of Meyer Shapiro, a highly perceptive specialist in the
field of Romanesque art,19 seems to me to be particularly illuminating for the
present-day historian.
For the medievalist, almost always working on images which are explicitly, or
at least implicitly, in a relationship with a text (first and foremost, the Biblical
text), an essential task is to bring out the specificity of the illustrated work, and
to draw from it all possible inferences. The structures of the fixed image (the
only kind of image available to Western civilisation until the invention of
cinematography) and of language are totally different. The former strikes the eye
simultaneously in all its constituent parts, even if it subsequently calls for a
longer period of ‘decipherment’ and of comparison with other, analogous but
older images. It constructs its own space or, as Francastel would put it, its own
constituent system of figures and places. Language, however, written or spoken,
extends over time, from the duration of the sentence to that of the whole
discourse, while creating the illusion that its meaning has imprinted itself
instantly upon our minds. Francastel has observed this very acutely:
The words which issue from our lips over a period of time form a snare, leading us
to assume that the thoughts they contain are on the contrary unified and
simultaneous. The snare of images which have appeared fixed throughout the
centuries is to evoke within us a combination of perception, knowledge and
idealized imaginings which only sporadically go together. That is why the
mechanisms of language and pictorial representation cannot be geared together.20
The respective specificity of image and language rules out any possibility of
the former being thought of as merely ‘illustrating’ a text, even in the case of a
miniature painted facing the text and in direct relationship with its sense or
meaning. Text conveys the signified temporally, via a succession of words;
image ensures, in spatial terms, the emergence of a ‘figurative thought’ which is
19 Meyer Shapiro, Words and Pictures. On the Literal and Symbolic in the Illustration
of a Text (The Hague and Paris: Mouton, 1973).
20 Pierre Francastel, La figure et le lieu. L ’ordre visuel du Quattrocento (Paris:
Gallimard, 1967), p. 351.
totally different. Besides, the construction of the image’s space and the way the
figures relate to each other are never neutral: at one and the same time, they
express, and produce, a classification of values, hierarchies and ideological
choices.
In Panofskian terms, an image’s meaning stems in the first instance from
iconography (in the strict sense of the word): in a single manuscript such as the
Gospels of Otto III, a portrait of the king in majesty analogous to that of Christ in
Majesty is full of significance, both for the medieval representation of kingship
and for that of the Divine (figs. 1 and 2).21 But no less important is the way in
which the structuring of the image may explicitly link these elements. For
example, in the 1511 Ansbach Retable, the fact that God the Father, operating
the press which crushes his Son, is wearing the same tiara as Pope Gregory the
Great, who—symmetrically opposed to him in relation to the suffering Christ—
is collecting the hosts coming out of the press, means that the image suddenly
becomes (in its historical context) a very meaningful summary of an entire
ideological programme, by bringing together the mystery of the Incarnation, the
dogma of the Trinity, and the sacrament of the Eucharist, and above all by
asserting the legitimacy of pontifical authority confronted with the first throes of
the Protestant Reformation (fig. 3, plate 3).22 Analysis of an image must
therefore take into account, as well as iconographical motifs, all the relationships
governing its structure and characteristic of modes of representation specific to a
given culture and period. The following are some principles to aid such analysis:
1. ‘In-depth study’. One should be aware of the stratification of the scene, from
the background, which in miniatures sometimes consists of a gold-leaf
ground, to the figures placed in the ‘foreground’: the first to strike one’s eye,
often seeming larger than the others, completely visible from top to toe, and,
within the value-system set up by the image, enjoying a privileged position
over the figures ‘taking up the rear’ and partially concealed by them. In just
such a way, the illuminated initials of a manuscript of Gratian’s Decretum
play on the characters’ size and posture (sitting, standing), on the choice and
brightness of the colours of their clothing, and above all on their position ‘in
front o f or ‘behind’ the letter framing these figures. All this may suggest a
social hierarchy by no means indicated explicitly in the text itself (fig. 4).2j
21 Munich, Bayerische Staatsbibliothek, MS. Clm 4453, fols 24 and 34v.

22 Le Pressoir mystique. Actes du colloque de Recloses (27 mai 1989). ed. by Danièle
Alexandre-Bidon (Paris: Cerf, 1990), p. 4 (Church of St Guinbert, Ansbach Retable,
c. 1511, school of Dürer).
2' Baltimore, Tire Walters Art Gallery, MS. W. 133, fol. 123 (Gradan, Decretum, Paris or
Northern France, c. 1300). Cf. Jean-Claude Schmitt, ‘Le miroir du canoniste. Les images et le
texte dans un manuscrit médiéval’, Annales E.S.C., 48 (1993), 1471-95.
Fig. I. Gospels o f Otto III: majesty o f kingship. Munich. Bayer. Staatsbibliothek.

MS. Chn 4453, fo i 24'. ©Bayer. Staatsbibliothek. Munich
2. The image should also be considered as an ‘inscribed surface’, with a set of

hierarchies, both vertical (ascending or descending) and horizontal (from
right or from left: from the point of view normally of the image, but
sometimes of the spectator). There is above all a sense of compartmentali-
sation, of rhythm, and of internal dynamics, achieved by means of
representational features (such as architectural motifs contributing to the
organisation of the image, as well as setting each scene within a formal
Fig. 2. Gospels o f Otto III: divine majesty. Munich. Bayer. Staatsbibliothek. MS.
d m 4453, fot. 34'. ©Bayer. Staatsbibliothek. Munich
rmfuiG ut«^î-cçinmr«i?2ctojc]0. t udic qf né WW

qir* icr«ííumm>aiic erma noi ¿muo itroráT II
trfh ^
m n o n t X n m v in c a in .n ib iitt& s fo M ir.
m t it f e f o m i t a c n i C tm o n e jx c b ir ciba
li le g m m t ‘U i i n d i w u x o n c fb u o m fa (>
m c$.7 n t t n iig io f t eucu taxcn ftfrctttö
T r io n i D fu â jT tn tc îïiz iâ lit& b in im ic o '
q c v p im x c iá Caá vent tnunoCo tu b iti
: _________ rr_____ — -------
Fig. 4. Decretum Gratiani, decorated initial: the dominant figure o f the canonist.
Baltimore, The Walters Art Gallery, MS. W. 133, fol. 123. © The Walters
Art Gallery, Baltimore
frame), and also of the colour scheme allotted to each image and to the
whole work in which it appears. In the miniatures (and we may assume that
the same was true with wall-paintings and polychrome church sculpture),
these colours excel in producing alternating, chiasmic, echoing, and repeat
effects which give the image its dynamic quality and provide a visual link
between certain figures. They also impart a degree of temporality to the
representational space, where appropriate, stressing a narrative line. The
linked sequence of scenes from Genesis in the St Louis Psalter does not
simply represent a faithful ‘illustration’ of the historia as present in Holy
Writ but is achieved through a continual repetition, from one image to
another, of lines, forms, and colours suggesting the continuity of the
narrative itself (figs. 5 and 6).24
3. Representational figures, ornamental motifs, forms, and colours acquire full
meaning only when considered in relationship: their relative positions;
contrasting or comparative features; the distance separating them; or else the
ways in which they come together, are juxtaposed, or even blend one with
the other. A single figure may be a composite, a dreamlike meld of several,
otherwise distinct figures, so as to express, through the contradiction of
posture or movement, the dialectic of the artist’s aim and meaning.25 The
attempts made from the twelfth century onwards to achieve a suitable
representation of the logical paradox of the Trinity, by combining the unity
of Divine Essence (‘Three-in-One-ness’) with the necessary distinction of
the three personae (‘Threefold-ness’), provide ample evidence of the power
and creativity of medieval ‘figurative thought’. In fact, we can detect a
hesitation, mainly between two logically opposed patterns: three distinct and
matching figures (the triandrous pattern),26 and three figures fused together
(the ‘Throne of Grace’) (fig. 7).27
4. No image is ever completely present in isolation. It often forms part of a
series: for instance, of full-folio miniatures in a single Psalter manuscript, or
of historiated initials commencing each Life in an illuminated manuscript of
the Golden Legend, or each causa in a manuscript of Gratian’s Decretum.
24 Paris, BnF, ms. lat. 10,525, fols 13' and 14r (St Louis Psalter, Paris, mid-thirteenth
century).
25 See. for example, the fusion within the historiated initial of the two figures of Jacob,
one dreaming, the other wrestling with the angel, in Oxford, All Souls College, MS. 6,
fol. 96 (Amesbury Psalter, thirteenth century). Cf. Jean-Claude Schmitt. ‘La culture de
Vimago'. Annales. Histoire. Sciences sociales, 51 (1996), 3-36 (p. 8, fig. 1).
26 Paris, BnF, ms. esp. 353, fol. 13 (Bréviaire d'Amour, fourteenth century).
2‘ Cambridge, Trinity College, MS. B. II. 4. fol. 119 (Psalter, thirteenth century). Cf.
François Boespflug and Yolanta Zaluska, ‘Le dogme trinitaire et l’essor de l'iconographie
en Occident de l’époque carolingienne au iv1- Concile de Latran (1215)’, Cahiers de
Civilisation Médiévale, 37 (1994), 181-240, illus.
Fig. 5. St Louis's Psalter: Jacob's ladder, and the anointing o f the stone of
Bethel. Paris, Bibliothèque nationale de France, ms. lat. 10525, fol. 13'.
© Bibliothèque nationale de France, Paris
Fig. 6. St Louis 's Psalter: Jacob ’s meeting with the angel, and Jacob wrestlin
Ö0-
with the angel. Paris, Bibliothèque nationale de France, ms. lat. 10525, fol. 14
© Bibliothèque nationale de France. Paris
The only truly meaningful work is perhaps that of the entire series: isolating
one image will always be arbitrary, and misguided.
5. An ‘original’ series of this kind must be distinguished from those ‘con
structed’ by the historian, based on iconographical, formal, structural, them
atic, or chronological criteria. The possibilities of constructing such series,
and above all of cross-referencing them, are infinite: for example, a
chronological series of Romanesque sculpted tympana can be created; this
may then be cross-indexed with one of Last Judgement scenes, not only in
sculpture but also in manuscript illustrations and wall-paintings. Or again, a
series of representations of Abraham’s Bosom can be tailored, as
exhaustively as possible, to reflect an historical anthropologist’s study of
‘Divine parentage’;2sor one may concentrate on structure, seeking out all the
graphic examples of a right-left division which may correspond to a crucial
ideological patterning in medieval society.*29
6. It is all the more imperative to think hard, and now, about the principles of
the construction of iconographical series, because the facilities of IT
documentation (via computers and digital images) are on the point of
transforming utterly our conditions of scholarly research and our opportu
nities of gaining access to material; this offers the possibility, through the
cross-indexing of image-selection criteria, of creating unlimited numbers of
new, and virtually complete archives. But these technological developments
raise as many new theoretical problems as they offer unheard-of practical
possibilities.30
7. Finally, a crucial point must be made: an image does not simply express
some cultural, religious or ideological ‘signified’, deemed to exist well
before or beyond this particular artistic expression. On the contrary, it is the
image itself which creates this element, just as we perceive it: creates it by
its very structure, form, and social effect. In other words, any analysis of a
work of art, of its form and structure, is inseparable from a study of its
functions. There must be unbroken continuity between analytical research
and historical interpretation.31
2S Jérôme Baschet, "Inventivité et sérialité des images médiévales. Pour une approche
iconographique élargie’, Annales. Histoire, Sciences sociales, 51 (1996), 93-134.
29 Jean-Claude Bonne, L'art roman de face et de profil. Le tympan de Conques (Paris:
Le Sycomore, 1984).
30 Over the past few years, several different systems of image-indexing have been
proposed. Our own suggestion has at least the merit of simplicity: Thésaurus des images
médiévales (published by the Groupe d’Anthropologie Historique de l’Occident
Médiéval, Paris, 1996).
On these points, see L'image. Fonctions et usages des images dans ¡'Occident
médiéval, ed. by Jérôme Baschet and Jean-Claude Schmitt (Paris: Le Léopard d’Or,
1996).
Fig. 7. Psalter (thirteenth century): the throne o f grace. Cambridge, Trinity

College, MS. B. llA .fo l. 119.© Trinity College, Cambridge
H istory
Hans Belting makes a distinction, in Western tradition, between a medieval ‘age’

of the image and its ritual and religious uses (‘Bild und Kult’), and an age of ‘art’
(‘Zeitalter der Kunst’) which dawned in the third decade of the fourteenth
century in Flanders and Italy and which is marked, notably, by ‘the invention of
painting’.32 The idea of the ‘Bild’, as he uses the term, operates on two different
levels. First the historical level, the author appearing to deny any aesthetic intent
in images created ‘before the artistic era’. And second, the historiographical
level: Belting states that social and cultural historians (of all periods) talk
nowadays of ‘images’, in order to treat them as they would other documents with
which they are familiar, and to distinguish their approach from that of ‘art
'2 Hans Belting, Bild und Kult. Eine Geschichte des Bildes vor dem Zeitalter der Kunst
(Munich: C. H. Beck, 1990); Hans Belting and Christiane Kruse, Die Erfindung des Gemäldes.
Das erste Jahrhundert der niederländischen Malerei {Munich: Hirmer, 1994).
Fig. 8. Conques, Abbey Church o f Sainte-Foy, tympanum (detail): St Foy in her

church, prostrate before the hand o f God
historians’ mainly concerned with identifying workshops, dating works, and

recognising styles.
Hans Belting has solid reasons for categorising, if not all then at least a large
number of, medieval images by their ‘cultic’, religious function. This feature
certainly does distinguish them from modem easel painting and its continually
developing aesthetic and secular functions. But one should not be too
categorical: not all medieval images were ‘cult’ objects (in the sense that the
enthroned statue of St Foy at Conques or the ‘Veronica’ cloth undoubtedly
were). One must also distinguish between a whole range of different ‘cultic’
items. At Conques, for instance, the sculpted tympanum at the very entrance to
the abbey church very probably did play a part in the veneration of the famous
statue-reliquary of St Foy preserved in the choir. The figure of St Foy in fact
appears on this tympanum, prostrate before the Hand of God as though she has
just left her throne in the body of the church, which is also depicted in the
sculpture, as are the chains hanging from the capitals under the vaults, left as ex
voto offerings by miraculously-cured pilgrims (fig. 8). The tympanum thus
invited those who were ready to proceed into the building to prepare themselves
for the imminent vision of the Majestas, by having them anticipate (but without
yet actually revealing it) the miraculous treasure they had come to ‘adore’. As a
result, carved on the tympanum and in her own tri-dimensional Majesty, the saint
was present in two images, which played complementary yet very different roles
in the development of her cult. If we turn to the frescoes in Italian courts of law,
these were associated not with any ‘cult’ in the religious sense of the word but
with civic ceremonies : at Siena, for example, justice was dispensed beneath the
painted representation of Good and Bad Government. Conversely, not all
modern images manage to escape the clutches of various sacred or profane
‘cults’; nowadays, a visit to a museum or to a major art exhibition often takes the
form of a ritual act, to which social pressure lends a note of obligation.
Over and above all this, however, if we deny medieval images the status of
‘art’ (‘Kunst’), we are faced with many difficulties. Costly material, expensive
labour, glittering gold, gemstones, and colours, and an item’s manifest beauty
would all combine simultaneously to exalt the glory of God and the prestige of a
rich and powerful sponsor. All these qualities would enhance the aesthetic value
of the work, which was considered inseparable from its religious or social
functions.33 Therefore, one should not make a distinction between ‘cult’ and ‘art’
but, rather, note how the one takes on the attributes of the other and is thereby
fulfilled. Clearly, one of the most difficult, and urgent tasks allotted to modem
historians and art historians is to understand the aesthetic function of works
(insofar as they may be considered ‘works of art’) as part and parcel of their
historical significance: be it in a ‘cultic’, political, legal, or ideological context).34
In actual fact, it is very difficult to see how the three terms set out in Belting’s
thesis can line up in a one-to-one relationship: one given period, one type of
image, one exclusive function. In every period there are many types of images,
each one with many possible functions.
That is why I also wish to retain the word image when referring to the Middle
Ages, not to distinguish it from ‘art’ but, rather, to restore to it all its meanings
and to group together the three components of the medieval imago: the world of
material images (imagines); that of the imagination (imaginatio), composed of
mental, oniric, and poetic images; and finally that of anthropology and Christian
theology, founded on the concept of Man created ad imaginem Dei and promised
salvation through the Incarnation of Christ imago Patris. If one only considers
one of these aspects, one will inevitably end up with a warped and incomplete
" Meyer Schapiro, ‘On the aesthetic attitude in Romanesque Art', in his Romanesque
Art (New York and London: Thames and Hudson. 1977). pp. 1-27.
■’4 On this point, see the recent work by Jean-Claude Bonne: ‘Pensée de Fart et pensée
théologique dans les écrits de Suger’. in Artistes et philosophes: éducateurs?, cd. by
Christian Descamps (Paris: Centre Georges Pompidou, 1994), pp. 13-50; ‘Formes et
fonctions de l’ornemental dans l’art médiéval (viT-xii siècle)’, in L'image (as in note
31), pp. 206-49; and ‘Les ornements de l’histoire (à propos de l’ivoire carolingien de
saint Rémi)’, Annales. Histoire. Sciences sociales. 51 (1996), 37-70.
38 Jean-C laude Schmitt
view of the history of the medieval image. The common task of historians and art
historians must be to study ‘imago als Kultur’.35
We can thus agree with Belting when he observes that our acceptance of the
word image has coincided with the widening by historians of a field of study
traditionally reserved for art historians. From now on, this should be our
common aim: to situate images within the overall context of ‘social imagina
tion’, with all its factors of power and memory,36 without in any way rejecting
the specific contribution made by art historians to our knowledge of artistic
works and traditions.
To achieve this, we are fortunately not lacking in conceptual equipment.
Thanks (again) to Aby Warburg, we no longer separate the study of a work’s
function from the analysis of its structure. Pierre Francastel expressed very
similar ideas when he compared the representational structures, with their
respective functions, of the Giotto frescoes in the Upper Church at Assisi, and
Duccio’s ‘Maestà’ in Siena: ‘Here again, function has dictated the composition
and the systematic ordering of the constituent parts’.37 In the relationship
between an image’s form and function, we can see the expression of the aims
and objectives of the artist, of the sponsor-patron, and of the entire social group
responsible for the work’s creation; and this work enshrines, from the very
outset, the attitudes of the person or persons to whom it was addressed, and the
uses (liturgical, for example) to which the image was put. Not only should the
type of image be taken into account, but also the place for which it was intended
(very different from the museum or library where it is often to be found today),
its possible mobility (whether, for instance, it could have been carried in a
procession), and also the ‘interactive performance’ between the figures within
the image exchanging glances with each other and with the spectators in the
world outside.
Similarly, the thoughts of someone like Michel Foucault teach us that no
image is first and foremost a simple ‘document’ for the historian (any more in
fact than is a charter or a chronicle).38 Nor is it a ‘monument’ reserved for the art
35 See Jean-Claude Schmitt. ‘La culture de l’imago’ (as in note 25). p. 4.

36 Andrea von Hülsen-Esch, ‘Zur Konstituierung des Juristenstandes durch Memoria:
Die bildliche Repräsentation des Giovanni da Legnano’, in Oexle, Memoria als Kultur (as
in note 15), pp. 185-206.
j7 Francastel (as in n. 20), p. 195.
,s Michel Foucault. L 'archéologie du savoir (Paris: Gallimard. 1969), p. 13. Foucault
secs a link between the appraisal of a document and a projected ‘general history’ along
the lines of the ‘Kulturwissenschaft’ discussed above. My own idea of applying the notion
of ‘document-cum-monument’ to medieval images owes much to Jérôme Baschet’s ‘Les
images: des objets pour l’historien?’, in Le Moyen Age aujourd'hui: trois regards
contemporains sur te moyen âge: histoire, théologie, cinéma: actes de la rencontre de
historian! It is for both of them, and in every detail, a ‘document-cum-

monument’, providing information about the historical circumstances of its
production and simultaneously revealing itself as a statement of religious faith or
as a proclamation of social prestige. Every image aims visibly at becoming a
‘receptacle of memory’, a monumentum, all the more so since memory itself, the
individual memoria as defined by St Augustine at the dawn of Christian
culture,39 and also the collective memoria, in all its social and cultural dimen
sions, consists above all of ‘images’.40 All this may easily be recognized (just to
take one prestigious example) in the miniatures—replete with the affirmation of
dynastic glory but still overseen by God, to Whom they are equally offered-—of
Henry the Lion’s Gospels (fig. 9, plate 4). Here once more, following the
Ottonian model which we have already seen (figs. 1 and 2) and responding to the
depiction of political sovereignty at the very moment of its divine legitimiza
tion—coronation by the hand of God—we find, on the facing folio, the glorious
manifestation of divine Majesty (fig. 10, plate 5).41
The historian must right from the start study a work in all the synchronic
depth of its social, cultural, and ideological underpinnings. But he must also
compose a history, take account of diachronic movement, tackle the delicate
question of the division into periods, propose a chronology. We have already
seen how highly unsatisfactory metaphorical references to a succession of styles
may appear: ‘Romanesque’, ‘Gothic’, etc., and yet, conversely, an attempt to
tack upon art a chronology based purely on matters of social history may equally
lead us to underestimate the rhythmical patterns of evolution which are very
Cerisy-la-Salle, juillet 1991, ed. by Jacques Le Goff and Guy Lobrichon (Paris: Le
Léopard d'Or. 1998).
39 Augustine, Confessions, X, 8: *[. . .] in the vast halls of my memory, where are
stored the treasures of all those innumerable images that entered by the gateway of my
senses [. . .]’. See also the very interesting case (as regards the relationship between
memory and image) of the monk Arnold of St Emmeran, cited by Patrick J. Geary,
Phantoms o f Remembrance. Memory and Oblivion at the End o f the First Millennium
(Princeton: Princeton University Press, 1994), pp. 16-19, and 159.
40 See, in Oexle, Memoria als Kultur (as in note 15), as well as the above-cited piece
by A. von Hülsen-Esch (as in note 36): Bernhard Jussen, ‘D ohr und Memoria.
Trauerriten, gemalte Trauer und soziale Ordnungen im späten Mittelalter’, pp. 207-52;
and Martial Staub, ‘Memoria im Dienst von Gemeinwohl und Öffentlichkeit.
Stiftungspraxis und kultureller Wandel in Nürnberg um 1500’, pp. 285-334.
41 Otto Gerhard Oexle, ‘Das Evangeliar Heinrichs des Löwen als geschichtliches
Denkmal’, in Das Evangeliar Heinrichs des Löwen. Kommentar zum Faksimile, ed. by
Dietrich Kötzsche (Frankfurt am Main: Insel Verlag, 1989), pp. 9-27; and Joachim M
Plotzek, in Heinrich der Löwe und seine Zeit. Herrschaft und Repräsentation der Welfen
1125-1235. Katalog der Ausstellung. Braunschweig 1995, ed. by Jochen Luckhardt and
Franz Niehoff, i (Munich: Hirmer, 1995), pp. 206-10.
40 JRAN-CLAUDI: SCHMITT
much part of images and their uses. A good example of these difficulties is given
by the first volume of the Histoire artistique de l ’Europe, devoted to the Middle
Ages and edited by Georges Duby.-12 Duby puts forward a chronology of Western
images which, after an initial five-hundred-year block (from the fifth to the tenth
century), suddenly becomes refined and subdivided into far shorter phases, with
precise dating: 960-1160, 1160-1320, 1320-1400. The terms of reference for
each of these periods are perfectly justifiable. And yet the individual essays
which follow, signed by a galaxy of an historians, specialists in Ottoman
manuscript illustrations, Romanesque façades, English alabasters, church
treasuries, enamels, stained-glass windows, funerary sculptures, and many other
items, do not really take much account of the chronological frame laid down
from the outset. The point is that chronology is never a simple matter; it is not
necessarily the same for all products of a given society: there is no real reason
why the art of stained glass should develop at the same pace as, say, a noble
lineage. In these circumstances we should conceive of historical time as a system
of various chronological strands, with advances in one sector being counter
balanced by sluggish development in another. The integration of art history in
our projected ‘total’ cultural history should result in some far more complex
divisions into periods than historians and art historians have independently been
used to.
This view of historical time as a multiple and contradictory entity must
necessarily precede any consideration of the function of images. Once he is
prepared to side-step his discipline’s traditional questions, and strive towards a
better understanding of the relationship set up between his objects and their
social context, the art historian will find himself, quite naturally, blending two
chronologies: that of the evolution of styles, and that of the evolution of society.
This was Millard Meiss’s working hypothesis when he started out on his study of
Florentine and Siennese painting after the Black Death:43 what could have been
the impact of the epidemic (which in a few months carried off more than half of
the population of those wealthy cities) on the art of this period? In fact, counter
to all expectations, the Plague left no trace on Tuscan painting, except perhaps in
the negative sense, as a horror that may certainly not have been unspeakable
(since writers and chroniclers, headed by Boccaccio, did speak of it), but was
unpaintable, and so was excluded by the artists. Above all, however, Meiss’s
research has allowed us to uncover the ‘fundamental anachronism of painting*4
42 Le Moyen Age, ed. by Georges Duby, in Histoire artistique de ¡'Europe, ed. by

Georges Duby and Michel Laciotte (Paris: Seuil, 1995). Apart from Duby himself,
twenty-six specialists from five different European countries collaborated on this work.
4'’ Millard Meiss. Painting in Florence and Siena after the Black Death (Princeton:
Princeton University Press, 1951); translated into French, with a preface by Georges
Didi-Huberman, as La peinture ci Florence el à Sienne après la Peste Noire. Les arts, la
religion, la société au milieu du Xive siècle (Paris: Hazan, 1994).
and its subject-matter’, in the words of Georges Didi-Huberman, who has shown
that the image of the Triumph of Death (beginning with the great fresco of
Buffalmacco, in the Campo Santo at Pisa) actually predates the Black Death. The
‘event’ of the Black Death was not the cause of an evolution in painting and its
themes; any cause is to be sought earlier, more deeply-rooted in the structural
development of mental and religious images of the Hereafter and of the fate
awaiting people after death.'14
The matter of connecting images to their contemporary social background is
also a prerequisite to a further field of research, nowadays still largely unculti
vated, but one which Marc Bloch had already declared to be an essential goal of
new historical studies: comparative history. Images and attitudes towards images
in different societies represent for me one of the best means of observing the
divergent courses of development followed by neighbouring civilisations,
stemming sometimes from common roots. Christian image-culture gradually
defined itself in opposition to the ‘idols’ of classical antiquity, which were
considered all the more shamefully evil for the powerful fascination which they
still exerted.44' It also had to justify itself in the face of Judaism’s proscription of
the graven image, while at the same time reiterating its affinity with the Old
Testament and the Ten Commandments. Finally, in the course of the Middle
Ages, Western Christendom and the Greek Church found themselves in radical
opposition over the matter of religious images, their formal characteristics, their
ritual uses and their theoretical underpinnings.
Byzantium moved from the most violent iconoclasm to the most fervent
iconolatry; this contrastive history led to a theology of the icon (with John the
Damascene, Theodore of Studion, and Nicephore the Patriarch) which has
known no equivalent in the West. In the long term, these icons have become
characterized by a relatively conservative formalism, explainable by the fact that
they are seen to bear, in their very matter, the imprint of divine emanation. Their
beauty is no independent quality: it is that of the invisible Power manifest in
them; and the wondrous stories of their divinely-inspired creation has quite
overshadowed the act of the artist who painted them.
In the Western Church, on the other hand, the history of the image has
been both calmer and more inventive: it has barely strayed at all from the via
media laid down by Gregory the Great, and the waves of iconoclasm
(towards 600, at the beginning of the ninth century, at times of medieval
44 Jérôme Baschct. Les Justices de l'au-delà. Les représentations de l'enfer en France

et en Italie (XII -XV siede) (Rome: École Française de Rome, 1993), p. 293; and 'Image
et événement: Part sans la peste (1340-1400)’, in La pesta nera: dati di una realtà ed
elementi di una interpretazione (Spoleto: Centro di Studi sul alto Medio Evo. 1994), pp.
25-47.
4' Michael Camille, The Gothic Idol. Ideology and Image-Making in Medieval Art
(Cambridge: Cambridge University Press, 1989).
heresy, and finally during the Protestant Reformation) have always been of
limited effect. But then the Latin churchmen never developed a ‘theology of
the image’ comparable in depth and subtlety to Greek icon theology. On the
other hand, despite the heavy influence of religious tradition, the diversity,
creativity, and sheer inventiveness of Western iconography contrast sharply,
and utterly, with Orthodox formalism.4647This made it easier to recognize the
artist’s autonomy, and from an early date.
The Romanian art historian Daniel Barbu seems to me to have perfectly
defined the formal differences between Western and Eastern art, and their dual
effect, both on the social level (regarding the status of the artist) and on the
political level (concerning in each case the relationship between image and
religious or secular authority). Taking as his point of departure the state of the
West in the Carolingian period, he writes:
The Libri Carolini lay down, even if only on a theoretical level, a certain degree
of autonomy for the creator of religious images, vis-à-vis the authority of the
Church. It is this autonomy which is perhaps responsible for the inventiveness
which distinguished Western artists from Byzantine painters, and also for the
highly diverse and restless treatment of forms which was to lend such vigour to
the art of the Christian West throughout the Middle Ages. The Church failed right
from the start to include the visual domain within the purlieu of its authority.
Since it had not been considered a res sacra, the image acquired a secular and
professional status which was always prohibited in Byzantium, where religious
painting remained essentially the responsibility of the clergy. In the West, because
the art of image-making was not duty-bound to follow prototypes controlled and
‘managed’ by the ecclesiastical hierarchy, it became almost free-thinking once in
the presence of its own criterion of truth: beauty. This was a sensitive concept of
beauty, no longer to be justified purely by its being the reflection of something
intelligible. It is at this very precise point that might be traced the very obvious
divergence of the Byzantine tradition and the flourishing art of the West.'1'
Going on to point to the status of the Byzantine image in the context of ritual
icon veneration, Barbu goes on to state: ‘The control of the Byzantine church
over the visual arts does not always conceal its close links with the operation of
Imperial authority’.48
The comparative approach should not merely set out to study the formal
make-up of images, or style, or the actual chronology of artistic traditions. As the
above references have shown, it must also bear in mind the religious, liturgical,
46 Jean-Claudc Schmitt, ‘L'Occident, Nicée II et les images du viif au xin1" siècle’, in

Nicée //, 787-1987. Douze siècles d ’images religieuses, ed. François Boespflug and
Nicolas Lossky (Paris: Cerf. 1987). pp. 271-301.
47Daniel Barbu, ‘L’image byzantine’: production et usage’. Annales.Histoire. Sciences
sociales, 51 (1996), 71-92 (74-75).
48 Ibid., p. 79.
and political functions of images, and even more generally the social and
ideological contexts of their creation and reception. Byzantine icons and Western
images draw on very different modes of social governance. In the East, the ‘war
on images’ was in all respects a political matter. Iconoclasm was decreed in the
eighth century by the Emperor, for whom it was an opportunity to assert his
prerogatives (to an unprecedented degree) in the spiritual domain. Yet it is a fact
that it also marked the beginning of the waning of his power, as the clerics began
to prepare their revenge. For them the only legitimate priest-king, the new
Melchisedech, was Christ, who had said, ‘My kingdom is not of this world’. In
opposition to the Old Testament models put forward by the Imperial party of
iconoclasts, the Patriarch and those monks in favour of icons cited the inspiring
example of the New Testament and the legitimacy of the Church of Christ and its
sacred images, the cult of which was renewed with unprecedented pomp.49
Nothing equivalent ever took place in the West, even when Charlemagne had the
‘Libri Carolini’ drawn up against Pope Hadrian I’s desire to extend the cult of
images into the Frankish Empire. Constantine’s dream of a mingling of powers
did not survive an early separation of regnimi and sacerdotium. The figure of
Melchisedech priest-king of Salem (Genesis 14. 18-20) did not become
dominant in Western political ideology, with the exception of some supporters of
the German Emperor: his presence was restricted to the typological symbolism
of Christ and the Eucharist.50 The ‘separation of powers’, which was in the long
term to remain a characteristic feature of all Western culture, was very formally
codified under Gregory vu. Admittedly, at this precise period, from the eleventh
century onwards, some Western images (called ‘objects of worship’ or ‘devo
tional’) were recognized as enshrining a sacrality and a miraculous power
unthinkable two hundred and fifty years earlier, at the time of the ‘Libri
Carolini’. But this combined power and sacrality—and one thinks, for example,
of the cult of the Veronica, promoted by the papacy throughout Western
Christendom at the beginning of the thirteenth century—only affected the
institutional Church and religious attitudes, or at the very most the religious
apanage and protocol of the political power, without the latter’s legitimacy ever
becoming dependent upon it. Neither the nature nor the exercising of imperial or
royal power was ever bound up with the denial or affirmation of the power of
holy images. Heretics and (later on) Reformers would indeed set about religious
49 Gilbert Dagron, Empereur et prêtre. Étude sur le 'césaro-papisme’ byzantin (Paris:

Gallimard, 1996).
50 See ‘Melchisedech’, in Lexikon der christlichen Ikonographie, ill (Freiburg im
Breisgau: Herder, 1971), cols 241—42. with particular reference to Nicholas of Verdun’s
retable at Klosterneuburg: Melchisedech, wearing the Imperial crown, is offering up on an
altar the sacrifice of Host and Chalice. For opposing examples of resistance to the
political use of the figure, in Western ‘Mirrors of Princes’, see Jacques Le Goff, Saint
Louis (Paris: Gallimard, 1996), pp. 127, 405, 578.
images:51 in so doing, they would certainly shake the Church hierarchy to its
foundations and disturb the public weal, but were never able to threaten the
legitimacy of political authority.
Western images, with the most diverse forms and functions, were able to
flourish in the Middle Ages in two distinct domains of society and culture. This
separation proved the guarantee of their relative freedom and of their un
paralleled development. In Western Christendom, religious images were con
strained neither by any ‘theology of the icon’ nor by the power, which in the
Eastern Empire a holy monarch might wield over icons, either to destroy them or
to venerate them. As for profane images (ranging from the frescoes in local
Italian courtrooms to portraits of fifteenth-century princes or merchants), these
very early on bore the stamp of a degree of autonomy, from which crucible
modern art and modern artists would finally emerge, fully emancipated.
51 Horst Bredekamp, Kunst als Medium sozialer Konflikte. Bilderkämpfe von der
Spätantike bis zur Hussitenrevolution (Frankfurt am Main: Suhrkamp. 1975); Olivier
Christin, Une révolution symbolique. L 'iconoclasme huguenot et la reconstruction
catholique (Paris: Minuit, 1991).
Nostalgia for the real: the troubled relation
o f art history to visual culture
K e it h M o x e y
‘When the real is no longer what it used to be, nostalgia assumes its full
meaning’.1
ne of the persistent features of art theoretical writing in the last twenty
O years has been the concern of a number of critics to insist on the ‘end of
art’, the ‘end of art history’, and the ‘end of aesthetics’. Motivated by a
variety of critical perspectives, such critics all address the grand narrative o f ‘art’
afforded by Hegelian philosophy. Some believe that we live at the end of the
process Hegel described, according to which art is to be viewed as the
materialization of the ‘spirit’, the principle of enlightenment, as it makes its way
through time. Following Hegel, Arthur Danto, for example, argues that the age
of art is over because the ‘spirit’ has transcended its materialization in artistic
production and has assumed wholly intellectual form as philosophy.2 While
acknowledging that artistic creation continues unabated, Danto insists that art is
no longer the place where the transhistorical drive toward self-consciousness is
realized. The teleological thrust of artistic modernism has ended. The narrative
whereby the various modernist movements bore a dialectical relationship to one
1Jean Baudrillard. ‘The Precession of Simulacra’, in Simulations, trans, by Paul Foss,

Paul Patton and Philip Bcitschman, Semiotext(e) 12 (New York: Semiotext(e), 1983).
2 Arthur Danto, ‘The End of Art’, in The Philosophical Disenfranchisement o f Art
(New York: Columbia University Press. 1986). pp. 81-115: idem. ‘Three Decades After
the End of Art', in After the End o f Art: Contemporary Art and the Pale o f History
(Princeton: Princeton University Press, 1997), pp. 21-39.
46 K eith M oxey
another, so that in contradicting its predecessor each new initiative sublated that
which came before into its own nature, has reached its conclusion. The unveiling
of Andy Warhol’s ‘Brillo Box’ in 1964 introduced a note of irony that rendered
the process complete:
For the past century, art has been drawing toward a philosophical self-
consciousness, and this has been tacitly understood to mean that artists must
produce art that embodies the philosophical essence of art. We now can see that
this was a wrong understanding, and with a clearer understanding comes the
recognition that there is no further direction for the history of art to take. It can be
anything artists and patrons want it to be. ’
Other critics, such as Hans Belting, also follow Hegel in the conviction that
not only art but also art history has ended.345Just as artistic production has lost its
internal necessity, so the principle of form, or style, on which a history of art
could be distinguished from other forms of historical interpretation, no longer
serves as a guarantee of its autonomy. The collapse of the idea of style as the
unifying concern of the history of art has led to a proliferation of different
histories of art, each pursuing a different agenda:
The more the inner unity of the history of art understood as an autonomous
discipline failed, the more it dispersed itself into the surrounding context of
culture and society in which it was thought to be included. The struggle about
method lost its sharpness, and the histories replaced one compelling art history
with several, indeed many, art histories, which exist peacefully alongside each
other like the different directions of contemporary art.’
For Douglas Crimp, it is the rise of the museum that sealed the fate of
universalist aesthetics. Associating Hegel’s notion of the ‘end of art' with the
historical rise of the museum, Crimp argues that art’s institutionalization, its
confinement in the museum under the aegis of an idealist theory of aesthetics,
served to remove it from the context of everyday life.6 The idea of art’s
autonomy, its alleged universal value, is effectively responsible for its demise.
While art continues to be produced, its autonomous status ensures that it cannot
be an agent in everyday cultural and social interactions. The birth of the museum
deprived art of its social significance, guaranteeing that it could only occupy a
marginal position in the life of culture.
3 Danto (as in note 2), p. 36.

4 Hans Belting, The End o f Art History, trans, by Christopher Wood (Chicago:
University of Chicago Press, 1987); idem, Das Ende der Kunstgeschichte: Ein Revision
nach zehn Jahren (Munich: Prestei, 1994).
5 Belting (as in note 4), p. 22 (my translation).
6 Douglas Crimp, On the Museum 's Ruins (Cambridge, MA: MIT Press, 1993).
Nostalgia for the reai 47
Once materialized within the museum, idealist aesthetics could be expected to

neutralize the possibility of art as revolutionary praxis or resistance. The effective
removal of art from its direct engagement in social life, the creation of an
‘autonomous’ realm for art, became the museum’s mission, and it was against this
that radical forms of modemist theory and practice were directed.7*
Finally, in a context far removed from art theory, the entire Western conception
of art and history has been challenged by a new awareness of the way in which it is
compromised by the existence of different cultures which either have no cultural
equivalents or whose potential equivalents are so radically different as to be
incommensurable. In his book Time and The Other, Johannes Fabian points out
that in the context of colonialism, Western modernism, with its obsession with
teleology and evolution, effectively discredited and marginalized notions of time
held by non-Westem societies. Post-colonial politics reminds us of the Eurocentric
bias of modernist ideology. Modernist claims to universality can no longer be
substantiated in an age that recognizes the value of cultural otherness. Just as
modernist, or historicist, time once invalidated the time of other cultures, so our
awareness of the legitimacy of different systems serves to undermine the Western
project.** In these circumstances, it is clear that there is nothing self-evident about
the concepts, ‘art’, ‘art history’, and ‘aesthetics’. Far from belonging to the order of
things, they have been shown to be cultural constructions that are bound to the
times and places of their creation.
It is no coincidence that these various hypotheses should have been advanced
in the context of the poststructuralist challenge to the signifying power of
language, and to the master narratives on which the myth of modernism was
based. The death of art, art history, and aesthetics can be viewed as a metaphor
of the broader epistemological crisis inaugurated by deconstruction. In his essay.
The Parergon, Derrida argued that in ‘The Critique of Judgement’ Kant’s
attempt to distinguish art from other cultural artefacts on the basis of a posited
universal human response depended upon the construction of aesthetic value
rather than upon its discovery.9 Derrida suggested that, say, in distinguishing a
picture from its frame, ascribing aesthetic significance to one and not to the
other, Kant discerned universal values in a distinction that was in fact culturally
ordained and therefore particular and local.
7 Ibid., p. 303.
NJohannes Fabian, Time and The Other: How Anthropology Makes Its Object (New
York: Columbia University Press, 1983).
9 See Jacques Derrida, The Trulli in Painting, trans, by Geoff Bennington and lan
McLeod (Chicago: University of Chicago Press, 1987), pp. 17-147; also David
Rodowick, ‘Impure Mimesis, or the End of the Aesthetic’, in Deconstruction and the
Spatial Arts: Art, Media, Architecture, ed. by Peter Brunette and David Wills
(Cambridge: Cambridge University Press, 1994), pp. 96-117.
4S K eith M oxey
A failure to distinguish the study of art from the study of other kinds of
images is often used as a criticism of the new field of academic inquiry known as
‘visual studies’ or ‘visual culture’. In this essay, I shall attempt to argue that the
animus directed against this new form of cultural analysis is misplaced. Rather
than view the rise of the study of varied and often popular forms of image
production—one in which art is included in a spectrum of other kinds of visual
products—as a potential threat to art as an institution, I would claim that the
value of these visual juxtapositions lies in comparing and contrasting the ways in
which the study of each genre, say painting, television, or advertising, makes use
of different theories and methods in the making of meaning. While this view
relativizes the study of art, suggesting that it is just one among many forms of
visual production, it recognizes the ethical and political concerns that motivate
its students. The study o f ‘visual culture’ insists that there is nothing ‘natural’ or
‘universal’ about claims to aesthetic value. The importance of the study of art, as
opposed to other kinds of images, lies in defining the particular significance
attached to the idea of aesthetic value in different cultures at different moments
in time.10
In viewing art from this perspective, in rendering it susceptible to comparison
with other forms of visual imagery and the ways in which they have been under
stood, the study of ‘visual culture' must acknowledge an agenda of its own. That
agenda, I would maintain, is inextricably linked to the history of poststructural
ism. Basing itself on a recognition of the role of subjectivity in the pursuit of
objectivity, a recognition that established fields of study are the record of an
‘archaeology’ of discourse, visual studies are often concerned with the way in
which the objects of study reveal the ethical and political commitments of those
who study them.
W. J. T. Mitchell, who is perhaps most responsible for the way the term
‘visual culture’ is currently employed, envisions visual culture as a field of
study, a focus for the varied interests of scholars working in many disciplines
from a number of different points of view. As a member of a faculty group at the
University of Chicago, Mitchell helped develop both a rationale and a syllabus
for a yearlong academic course in the subject." Regarding visual culture’s
relation to art history, Mitchell writes:*1
"’The continued use of the terni art in this essay, particularly in contrast to something
called non-art, should be regarded as a means by which to refer to a changing form of
cultural discourse that deals with artefacts for which philosophical, religious, and/or
political claims are advanced, which are considered part of the heritage of the nation
states, and which play a role in the capitalist commodities market.
11 W. J. T. Mitchell, ‘What is Visual Culture?’, in Meaning in the Visual Arts: Views
from the Outside: A Centennial Commemoration o f Erwin Panofsky (1892-1968), ed. by
Irving Lavin (Princeton: Institute for Advanced Study. 1995). pp. 207-17.
Nostalgia fo r lhe real 49
From the standpoint of a general field of visual culture, art history can no longer
rely on received notions of beauty or aesthetic significance to define its proper
object of study. The realm of the vernacular and popular imagery clearly has to be
reckoned with and the notion of aesthetic hierarchy, of masterpieces and the
genius of the artist have to be re-described as historical constructions specific to
various cultural place-times.'2
If the distinction between art and visual artefact is to be seen as a cultural con
struction rather than as a cultural ‘given’, then what kinds of imagery is visual
culture to address? On this point Mitchell is far less specific:
The point of a course in visual culture, in short, would be to provide students with a
set of critical tools for the investigation of human visuality, not to transmit a specific
body of information or values. If the questions and debates are posed frankly, as the
very substance of the course rather than as something already settled, then
information, values, and exposure to the finest productions of visual culture will
inevitably follow.1’
Such a definition appears to beg the question. The world of images is so vast
as to necessitate a frame. Having already dismissed ‘received notions of beauty
or aesthetic significance’, it is difficult to understand his invocation of the idea of
quality, his reference to The finest productions of visual culture’, as the grounds
for determining the content of the course. While Mitchell’s reluctance to suggest
a frame for the study of visual culture might be regarded as providing a useful
flexibility for a field of study that has yet to locate itself historically, that
reluctance must also be regarded as a theoretical weakness.12134
A more important drawback to Mitchell’s proposed course is its ambition to
provide students with ‘a set of critical tools for the investigation of human
visuality’. This sentiment, which echoes the goals of traditional art appreciation
courses that aspire to enable students to Team to look’, suggests that there might
be a simple set of concepts, a toolbox, with which the rich and complex world of
visual experience might be understood.15If visual culture is not to be reduced to the
12 Ibid., p. 210.
13 Ibid., pp. 207-17.
14 It is a weakness shared by certain other publications on the subject of "visual
culture’, which, in celebrating the heterogeneity of the potential subjects to be addressed,
fail to address the necessity to offer a focus, an agenda, or a rationale for the enterprise.
See. for example. Visual Culture, ed. by Chris Jcnks (London: Routledge, 1995). which
includes essays on topics as diverse as advertising, visual reporting, contemporary art. the
visual experience of the city, fascist imagery, television and so forth.
15 The formalist ideology of Joshua Taylor’s famous Learning to Look (Chicago:
University of Chicago Press. 1957) has been usefully contextualized by Linda Seidel and
Katherine Taylor in their account of the social and cultural agendas that have determined
the status and function of ‘art’ at the institution for which Taylor's course was designed.
50 K eith M oxey
study of art on the basis of a universal notion of aesthetic significance, why should
its potential be foreclosed by assuming that there is a universal epistemological
basis for the enterprise?
Rather than reduce the analysis of visual culture to a single set of principles, it
seems to me that the point of the academic study of images is the recognition of
their heterogeneity, the different circumstances of their production, and the variety
of cultural and social functions they serve. It is precisely because the interpretative
paraphernalia, the heuristic strategies, used in the elucidation of different traditions
of visual production are so radically different from one another that each form of
investigation has much to gain from a familiarity with the others. This, in fact, is
the underlying logic behind the structure of at least one of several university
programmes that have adopted ‘visual studies’ as their title. In the graduate prog
ramme instituted at the Univ. of Rochester in the late 1980s for instance, studies
in‘Visual and Cultural Studies’ consist of a variety of discipline-based require
ments.*16 Students can take courses in art history, film, or popular culture, without
sacrificing the specificity of the interpretive traditions associated with the study of
each form of cultural discourse. Their exposure to different methodological
traditions enables them to inform their own interpretations with an awareness of a
broad range of different points of view. For example, in the study of art history,
they are capable of drawing upon theories and methods familiar to them from their
study of, say, film or television.
More importantly perhaps, an awareness of the ideological diversity implicit in
the various fields that concern them, the sensitivity of students to the way in which
the production of knowledge manifests the circumstances in which it takes place,
as well as the interests of those involved, affords them insight into the changing
nature of their own cultural activity. In being asked to familiarize themselves with
the traditions that animate each discipline, students have an opportunity to analyse
and deconstruct the philosophical attitudes and cultural values that have informed
the frame constructed around those disciplines at different moments in time.
Recognition of the shifting and unstable nature of disciplinary discourses, the way
in which they have responded to changing historical circumstances, enables them
to evaluate the parameters within which the disciplines currently operate. The
awareness that disciplines are not frozen in time, that there is nothing foundational
about the methodological differences that distinguish them from one another,
empowers those who study visual culture to cross disciplinary boundaries with
creative and productive results. If there is nothing sacred about the way in which a
See Linda Seidel and Katherine Taylor, Looking to Learn: Visual Pedagogy at the
University o f Chicago (Chicago: University of Chicago Press, 1998).
16 For a discussion of the aims of this project, see Scott Heller, ‘Visual Images Replace
Text as Focal Point for Many Scholars’, Chronicle o f Higher Education (19 July 1996),
Appendix I. The Rochester programme has inspired a similar enterprise at the University
of California, Irvine.
Nostalgia for the reaI 51
subject has been studied in the past, then there is nothing to prevent it from being
viewed from a different perspective in the future.
Needless to say, the rise of visual culture as a subject of academic interest has
not gone unchallenged. Ironically enough, that challenge has come both from
scholars familiar with and sympathetic to poststructuralist thought, as well as
from those dedicated to the preservation of the status quo. While the tendency to
blur the distinction between art and non-art contains within it an implicit
rejection of the idea of universal aesthetic value that has been the traditional
basis of art historical studies, it also seems to have offended critics identified
with new approaches to interpretation in the visual arts, including those who
have explicitly rejected the idea of the humanist subject on which traditional
claims for aesthetics are made.
In 1996, the journal October circulated a questionnaire to a number of
academics who could be identified as broadly interested in the study of images,
including art historians, students of film, and of visuality in general, soliciting their
views regarding the visual culture initiative. The questionnaire betrayed the suspi
cion of its framers: namely, that the establishment of this new field was an unfortu
nate development, which catered to the exploitative interests of late capitalism.
Since such a claim is clearly not self-evident, it is worth examining in some detail.
Question three reads: ‘It has been suggested that the precondition for visual studies
is an interdisciplinary rubric in a newly wrought conception of the visual as
disembodied image, re-created in the virtual spaces of sign-exchange and
phantasmatic projection. Further, if this new paradigm of the image originally
developed in the intersection between psychoanalytic and media discourses, it has
now assumed a role independent of specific media. As a corollary, the suggestion
is that visual studies are helping, in its own modest, academic way, to produce
subjects for the next stage of globalized capital’.17
Who was it that made this suggestion? Who said what to whom, when, and
where, and what did they mean by it? The disembodied voice of the text affords
no clue to these queries. The potential respondent who seeks to react to these
assertions is placed at a disadvantage. What, for example, is meant by, ‘. . . the
precondition for visual studies is an interdisciplinary rubric in a newly wrought
conception of the visual as disembodied image’? If this is meant to suggest that
visual studies seek to separate the image from the rest of the cultural production
that surrounds it, nothing could be further from the truth. The point of visual
studies, as I see it, is to locate the image in the midst of the meaning-making
processes that constitute its cultural environment. Such a definition means, for
example, that visual studies deliberately ignores those guarantees that have
17October, 77 (1996), 25. For responses to the issues raised by this questionnaire, see
Douglas Crimp, ‘Getting the Warhol We Deserve: Cultural Studies and Queer Studies’,
Social Text (1999), and Irit Rogoff. ‘Studying Visual Culture’, in The Visual Culture
Reader, ed. by Nicholas Mirzoeff (London: Routledge, 1998), pp. 14-26.
52 K eith M oxey
traditionally been sought by scholars interested in preserving the autonomy of

art. One of the assumptions underlying visual studies, I believe, is that the image
is saturated with the word. Visual sign systems are inextricably bound up with
linguistic and other codes that characterize a particular culture at a particular
historical moment.
Question three’s suggestion, namely, that a poststructuralist approach to the
study of the image—one that implies that aesthetic value is culturally con
structed rather than foundational, is a manifestation of the economic characteris
tic of late capitalism—is hard to follow. Here the inspiration seems to be Fredde
Jameson’s influential argument concerning postmodernism.18 Jameson
subscribes to an evolutionary view of history according to which postmodernism
necessarily follows modernism because its culture is the product of the late
capitalist economy. While such an understanding views history as a teleological
movement through time, in Jameson’s case this is not a Hegelian model but a
Marxist one; the engine of history is not the passage of the spirit but the class
struggle. According to Jameson, theoretical critiques of the ‘grand narrative’ of
Western history offered by Marxist thought are to be viewed as symptoms of a
particular historical moment. Criticism of totalizing theories of history, in this
case the Marxist view, is considered a manifestation of the post-modern
condition, rather than as a challenge to a traditional conception of history. On
this view, the Marxist understanding of history is not to be questioned because it
is impossible to think outside it. History is regarded as the real circumstance of
human existence, rather than as a particular philosophical outlook. Jameson’s
Marxism is necessarily a metaphilosophy, one that must encompass all other
philosophical systems while refusing to recognize that it is one itself. This very
lack of self-reflexivity, however, this naturalization of the grounds of historical
thought (the foundation of modernism as a literary and artistic movement), is
what visual studies seek to challenge.
Based on a Derridean philosophy of language, one that insists that language is
the home of metaphysics and that meaning is only constructed out of its absence,
visual studies focuses first and foremost on the frame within which meaning is
produced. Rather than subscribe to a theory that reduces theoretical innovation to
the ideological ‘superstructure’ of a culture whose shape is determined by its
economic ‘base’, visual studies insists that despite the cultural forces that
necessarily condition consciousness, human action is far from predictable. In the
absence of the kind of totality posited by Jamesonian theory, it is the very
conflicts of cultural discourse, its heteroglossia, which ensures that the principle
of agency cannot be overlooked in cultural analysis. In the wake of
poststructuralism, a set of theories that emphasized the indeterminacy of
18 See Fredric Jameson, Postmodernism or The Cultural Logic o f Late Capitalism

(Durham, NC: Duke University Press, 1991), and more recently. The Cultural Turn:
Selected Workings on the Postmodern (London: Verso, 1998).
Nostalgia for the rea/ 53
meaning and our inability to know the world. Jameson supports one of the more
traditional of the grand narratives of history, namely Marx’s claim to understand
the relation of human beings to the real circumstances of their existence. In this
situation, not only his theory of culture but also the attack on visual studies that
is based upon it seems to constitute a kind of ‘nostalgia for the real’.
Hal Foster, who also contributed an essay to the same issue of October, took a
negative stance toward the idea of visual culture, arguing that it implied a move
from ‘'art to visual and from history to culture' P Taking the second set of
binaries first, Foster claims that visual culture participates in what he calls an
‘ethnographic turn’. By this he means that in recognizing that positionality
matters in the production of discourse—that knowledge is inflected by the
interests of those responsible for its articulation—visual culture is in danger of
compromising traditional (i.e., Marxist) conceptions of history. ‘For in the
ethnographic model one moves horizontally, from site to site across social space,
more than vertically, in a discourse inscribed with an historicity, a responsibility
of form, of its own. In this way, the shift from history to culture may promote, in
art as well as in criticism, a posthistorical reduction as often as a multihistorical
complication’.1920
Foster thus opposes diachrony to synchrony, arguing that the latter gains at
the expense of the former. Recognition of the spatiality of culture, of the multi
tude of subject positions that animate it, is regarded as inimical to chronology. If
the differences among subject positions are sufficiently radical, then it becomes
impossible to agree upon a single version of the historical events. Foster’s
misgivings, however, are misplaced. In dispensing with the grand narratives that
hitherto have animated our conception of history, visual culture does not propose
foregoing a philosophy of history. It is not time that is under attack by those who
wish to study visual culture, for it is the interaction of the principles of time and
space to which this field of investigation is dedicated. If the Hegelian and
Marxist narratives are abandoned, then it becomes possible to rethink history
from the perspective of cultural difference. Far from being a reductive
development, such a move opens up the study of history to make its narratives
potentially far richer and more interesting. In other words, ‘multihistorical
complication’, the possibility of telling an historical narrative from a variety of
points of view, cannot be identified with ‘posthistorical reduction’, the alleged
collapse of history attributed to the challenge of its grand narratives, without
appearing paradoxical.
In replacing art with the concept of the term visual, Foster claims that visual
culture foregoes the potential inherent in the concept of art’s autonomy. Invoking
‘subjection’, as autonomy’s ‘other’, he argues that without autonomy, art is
19Hal Foster, ‘The Archive Without Museums’, October, 77 (1996), 97-119 (p. 104).
20 Ibid., pp. 104-05.
54 K eith M oxey
capable of being ‘reduced’ to the merely visual. In an argument reminiscent of

Adorno’s view of art as a locus of resistance to capitalist culture, Foster writes:
Enlightenment thinkers like Kant proclaimed autonomy in order to wrest institutions
away from the ‘ancien régime’, art historians like Riegl to resist deterministic
accounts of art, modernists from Manet to Judd to challenge the priority of
iconographie texts, the necessity of illustrational meaning, the imperialism of the
mass media, the overburdening of art with voluntaristic politics, and so on. Like
essentialism, autonomy is a bad word, but it may not always be a bad strategy: call it
strategic autonomy.
It is no accident, of course, that Foster’s last phrase, ‘strategic autonomy’,
should echo Gayatri Spivak’s ‘strategic essentialism’.2122 Like Spivak’s conception
of essentialism as something that can be either relative or absolute according to
political circumstance, so the idea of ‘strategic autonomy’ both relativizes and
asserts the autonomy of art at the same time. In Foster’s case, the assertion of the
autonomy of art is a political gesture (one that is historically determined and
subjectively freighted), meant to stave off the threat to art that is posed by the
visual. For Foster, this is an either/or situation; either art is autonomous, or it
disappears into the visual.
Against Foster, I would argue that the binary oppositions ‘autonomy/
subjection’, ‘art/visual’, are, in fact, uncalled for. Rather than insist on the
primacy of art in relation to other forms of cultural imagery, why not simply
recognize its distinctiveness? Why not view art as one of the many kinds of
imagery that constitute visual culture? Rather than see art dissolved into other
forms of imagery that surround it, it is in the interest of visual culture to recog
nize it as a distinct form of visual creativity, one possessed of an historiographic
tradition of its own. Indeed, the point of visual culture would be not only to
recognize the different genres of image production that characterises a particular
culture, but also to insist that their unique qualities call for distinct approaches to
their interpretation.
If, as I have been arguing, the notion of art is not necessarily to be regarded as
opposed to the idea of ‘visual culture’, if it is possible to conceive of art as a
particular circumscribed form of cultural discourse, one distinguished or
bracketed off from other forms of cultural interaction, how then might we define
the breadth and scope of the field of imagery the new field of study might
address? One quick way to reduce the new enterprise to absurdity would be to
suggest that it should be open to a consideration of all forms of cultural imagery,
ranging from digital and electronic images to comic strips, without making
qualitative distinctions between them. The rationale for including or excluding
21 Ibid., pp. 118-19.

22 Gayatri Spivak, ‘Subaltern Studies: Deconstructing Historiography’, in In Other
Worlds: Essays in Cultural Politics (New York: Routledge, 1988), pp. 197-221.
Nostalgia fo r the real 55
different genres of image production under the rubric ‘visual culture’ will clearly
depend on historical circumstances, educational needs and political considera
tions.
In current circumstances, I believe that one of the most powerful reasons for
including the study of art along with non-art lies in the acknowledgement of the
collapse of foundationalist theories of knowledge. ‘Visual culture’ opens doors
to the production of knowledge from a variety of different subject positions. It
fosters forms of interpretation inspired by feminism, queer studies, and
postcolonialism. Rather than seek consensus, it benefits from the radical
disagreement among different styles of interpretation. The heteroglossia that
characterizes the operations of visual studies is regarded as a positive and
dynamic aspect of its activities. The pursuit of differing styles of interpretation
under the aegis of ‘visual culture’ means that the particular interests associated
with each form of knowledge production are hard to miss. Since it is no longer
possible to suggest that knowledge is ‘disinterested’, the alterity of the
juxtaposed voices inevitably will draw attention to the cultural agenda that
informs the enterprise. Each form of argumentation can afford to foreground
rather than conceal its political affiliations.
Current theoretical assumptions concerning knowledge production suggest
that it is in difference rather than in sameness that the strongest contributions to
the humanities are likely to be made. Each discipline has an opportunity to
expand its repertoire of interpretative methodologies in relation to those around
it, and it is in the interstices between disciplines that the greatest insights
regarding the structure of our knowledge, as well as our need to pursue it, are
likely to be derived.
Pourquoi élaborer des bases
de données d’image r
Propositions pour une iconographie sérielle
JÉRÔME BASCHE!
ujourd’hui, la nécessité de constituer des bases de données d’images est
A de plus en plus clairement ressentie par les chercheurs. Les initiatives et

réalisations se sont multipliées au cours de la décennie écoulée, promues
soit par des institutions de recherche, soit, souvent aussi, par des chercheurs
individuels, à la fois soucieux de valoriser les matériaux mis à jour par leurs
propres travaux et désireux de contribuer à la constitution des outils documen
taires dont ils ressentent le besoin.1Ce rôle important joué par les chercheurs, qui
détournent ainsi une part du temps normalement voué au travail scientifique
proprement dit, peut s’interpréter à la fois négativement, comme signe d’une
certaine inertie des institutions, et positivement, dans la mesure où la constitution
1 Parmi bien d’autres exemples, on citera pour le premier cas la base de données des
manuscrits enluminés des Bibliothèques publiques de France, entrepris par l’Institut de
Recherche et d’Histoire des Textes (C.N.R.S., Orléans), et pour le second le projet
PREALP, consacré aux peintures murales alpines et mené par Dominique Rigaux, ou
encore le projet Fons Sacra, consacré à l’iconographie des fonts baptismaux européens,
lancé par Harriet Sonne. Il existe évidemment de nombreuses situations intermédiaires (ne
serait-ce que parce qu’une initiative individuelle doit, pour réussir, trouver des soutiens
institutionnels). Parmi de nombreux exemples, on citera, la base de données consacrée
aux peintures murales danoises, dirigée par Axel Bolvig (cf. A. Bolvig, 'Danish Wall
Paintings—An Introduction’, Medium Aevum Quotidianum, 39 (1998), 7-8, numéro
consacré à cette base de données), ou encore les vidéodisques des manuscrits enluminés
de la Bibliothèque Vaticane, à la réalisation desquels j ’ai pris part (trois vidéodisques
réalisés en collaboration entre la Bibliothèque Vaticane, l’École française de Rome et
l’École des Hautes Études en Sciences Sociales, et disponibles depuis 1994).
60 JÉRÔME HASCHET
des outils documentaires s’articule ainsi directement aux besoins de la recherche.

Quoi qu’il en soit, on ne peut que se réjouir des progrès accumulés durant ces
dernières années, des réalisations déjà accomplies et des projets en cours, tout en
regrettant les retards pris et en remarquant que la situation actuelle est encore très
loin de correspondre aux possibilités ouvertes par les développements
techniques.2
Au-delà du bilan des années écoulées et du diagnostic sur la situation
présente, il convient de s’interroger plus attentivement sur les objectifs et les
perspectives scientifiques de tels investissements, sur leur nécessité et leurs
conséquences, sur les relations entre ces nouveaux moyens documentaires et la
modalités de la recherche dans le domaine des images médiévales. On ne
reviendra pas ici sur les principes qui, selon nous, devraient guider l’établisse
ment des bases de données,' et on s’efforcera seulement d'analyser leur effet
pour la recherche. A cet égard, il serait évidemment fort réducteur de limiter les
bienfaits des bases de données à un simple gain de temps, à la facilité plus
grande avec laquelle elles permettent de disposer des images nécessaires à une
enquête particulière. Faire valoir l’augmentation du nombre des oeuvres mises à
la disposition des chercheurs paraît également insuffisant. De même que la
photographie a modifié la pratique traditionnelle de l’histoire de l’art en facili
tant et multipliant les comparaisons possibles, on soutiendra qu’aujourd’hui la
possibilité d'accéder et de ‘manipuler’ des stocks d'images considérables (et à
terme la totalité des oeuvres conservées) doit entraîner dans la pratique de la
recherche des modifications non seulement quantitatives mais aussi qualitatives.
Ce sont précisément les transformations des méthodes d’analyse des oeuvres,
induites par les nouvelles techniques, qui nous semblent appeler réflexion et
discussion. On proposera ici une contribution à ce débat collectif, en défendant la
proposition d’une iconographie sérielle. Cette expression suggère d’emblée une
dimension quantitative (à la mesure des masses d’images sous lesquels les
historiens et les historiens d’art peuvent craindre d’être bientôt ensevelis . . .).
Pourtant, l’iconographie sérielle ne se préoccupe pas seulement de grands
2 L’un des principaux regrets tient à la sous-utilisation de l’Index of Christian Art de

Princeton, qui reste pourtant le plus ample outil de recherche documentaire concernant les
images médiévales. On peut ici incriminer la lenteur de son informatisation et la
restriction de ses implantations. Il est tout à fait anormal que cet instrument de travail ne
soit pas systématiquement utilisé par les chercheurs concernés par l'image médiévale.
Je m'en suis expliqué ailleurs, à la fois théoriquement (‘Les vidéodisques des
manuscrits de la Bibliothèque vaticane et la réalisation d’une base de données
iconographique’, Ane medievale, 1 (1992), 199-205, et ‘Remarques sur l'indexation
iconographique. L’expérience des vidéodisques des manuscrits de la Bibliothèque
Vaticane’, Mélanges de l'École française de Rome. Moyen Age, 106 (1994), 179-88), et
pratiquement, par la mise au point collective du Thesaurus des images médiévales, éds.
Groupe d'Anthropologic Historique de l’Occident Médiéval. CRH-EHESS (Paris, 1993).
Propositions pour une iconographie sérielle 61
nombres, qu’il suffirait de gérer par d’habiles recours statistiques. S'appuyant

sur une analyse de la nature et du mode de fonctionnement des images
médiévales, l’iconographie sérielle place au centre de la réflexion l’articulation
du traitement quantitatif et de l’analyse qualitative. Elle prend pour objet des
gammes de variations, incluant régularités massives et combinatoire des vari
antes, exploration du champ de possibilités figuratives et radicalité des coups
d’éclats. En faisant apparaître des séries de transformations, l’analyse sérielle
peut tout à la fois traiter des productions massives et restituer à chaque oeuvre sa
singularité. Ce faisant, non seulement elle contribue à une nouvelle perception de
l’iconographie médiévale, qu’on ne saurait analyser en termes de stéréotypes
codifiés ou de modèles passivement reproduits, mais elle prend part aussi à un
débat qui traverse l’ensemble du champ des sciences sociales, en quête de
méthodes permettant d’articuler l'individuel et le collectif, le particulier et le
général/
Inventivité de l ’art médiéval

Afin de fonder la pertinence d'une approche sérielle, il importe, en préalable, de
mettre en question le préjugé encore courant selon lequel les images médiévales
relèveraient d'un art stéréotypé, presque figé, et de surcroît obligé à reproduire
passivement la doctrine de l’Eglise. On soutiendra au contraire que le Moyen
Age occidental, surtout à partir du IXe et plus nettement encore du XIc siècle, est
une période de liberté particulièrement grande pour les images et de profusion
iconographique exceptionnellement intense.4 5 L’approche sérielle doit alors être
conçue en relation avec la reconnaissance de cette extrême inventivité. La
conception d’un art stéréotypé, dogmatique, épaulée par la vieille idée de l'image
comme Bible des illettrés, correspond particulièrement à la vision transmise par
Emile Mâle.6 Pour ce dernier, l’art, soumis à l’Eglise et au dogme, ‘traduit la
4 Le texte qui suit reprend (à quelques légères modifications près) la seconde partie de
‘Inventivité et sérialité des images médiévales. Pour une approche iconographique
élargie’. Annales. Histoire. Sciences sociales, 51 (1996), 93-133. En ce qui concerne
la notion d’iconographie, on renvoie à la première partie de cet article.
' On parlera de liberté de l’art (ou des images), plutôt que de liberté de l’artiste, pour
ne pas préjuger du statut de ces derniers, ni mésestimer le rôle des commanditaires. Ce
que l'on veut mettre en évidence concerne l’inventivité des oeuvres clles-mcmes. Enfin, il
est clair que cette liberté ne saurait être comparée aux conditions de création des artistes
des XIXe et xxe siècles. Voir R. Berliner, ‘The Freedom of Medieval Art’, Gazette des
Beaux-Arts, 28 (1945). 263-88.
6 Pour la critique de cette notion, je renvoie à mon article ‘Introduction: l'image-objet’,
dans L 'image. Fonctions et usages des images dans l'occident médiéval, éds. J. Baschct et
J.-C. Schmitt (Paris: Cahiers du Léopard d’Or, 1996), pp. 7-26.
62 Jérôme B aschet
pensée des docteurs’, reproduit ‘tout de que les théologiens, les encyclopédistes,
les interprètes de la Bible ont dit d’essentiel’.7 Expression d’une pensée
structurée, encyclopédique, l’art—en particulier celui du XlIIe siècle—constitue
un ensemble unifié; il repose sur l’usage d’un language codifié, normatif,
conventionnel.8 Il est donc possible d’en traiter comme d’une réalité homogène:
‘nous le considérons comme un tout vivant, un ensemble achevé’.9 L’art exprime
le discours de l’Eglise, et c’est pourquoi il est possible de tenir un discours sur
l’art de cette époque.
Dans L'image médiévale, Jean Wirth a reformulé de façon neuve et stimulante
le projet d’Emile Mâle d'une saisie globale de l’art et de l’iconographie du
Moyen Age.10*Concernant la question du caractère codifié de l’art médiéval, sa
position est ambiguë. D’un côté, il souligne que l’iconographie n’est pas le
décalque de la théologie, et peut même la contredire. Il traque avec prédilection
tout ce qui déborde des types iconographiques courants: les hybridations, les
traits originaux, et même l’inconvenance, et souligne ‘la prodigieuse inventivité
de l’art religieux’. Pourtant, l’hypothèse centrale du livre, attribuant un caractère
logique à l’image, réintroduit des caractères de codification et de systématicité
qui autorisent à parler de Vimage médiévale (au singulier)." Il est significatif que
Jean Wirth reprenne, pour souligner le caractère logique de l’iconographie
médiévale, l’exemple des nimbes, déjà invoqué par Emile Mâle.1213Sans doute en
complique-t-il la présentation; mais il demeure qu’il surestime le caractère
logique et systématique du fonctionnement du nimbe dans l’art médiéval (négli
gence de certaines formes, ni circulaires, ni carrées, ou encore des degrés
d’ornementation; non prise en compte du fonctionnement syntaxique du
nimbe).1’ Plus largement, tout en partageant l’idée d’une extrême inventivité de
7 Emile Mâle, L'ar! religieux du x u t siècle en France. Etude sur l ’iconographie du

Moyen Age et sur ses sources d ’inspiration (Paris, 1898 ; 8e éd., 1948), p. 13 et p. 351:
‘Ici, comme partout ailleurs, les artistes du XIIIe siècle furent les interprètes dociles des
théologiens’.
s L’art médiéval, ‘écriture sacrée’, se rapproche du fonctionnement du langage verbal:
ses signes ‘sont de véritables hiéroglyphes: l’art et l’écriture se confondent’, ibid., p. 31.
9 (cf. note 7), p. 15. Ceci se traduit dans l’étude concrète, par exemple lorsque E. Mâle
traite de l’Enfance du Christ, en faisant comme s’il existait un cycle type, reproduit à
l’identique dans toutes les oeuvres (p. 353).
10 J. Wirth, L ’image médiévale. Naissance et développements, v f- x d siècle (Paris:
Klincksieck, 1989), pp. 18-23.
1' Pour une critique de cette conception, et en particulier des références à la logique, cf.
J.-C. Bonne, ‘A la recherche des images médiévales’. Annales E.S.C., 2 (1991), 353-73.
12 Wirth (cf. note 10). pp. 23-25: Mâle (cf. note 7), p. 30.
13 Ainsi, l’aspect syntaxique conduit très souvent à réserver le nimbe pour le
personnage principal, comme le Christ, et à l’exclure pour les autres (les apôtres
l’art médiéval, on préfère attribuer à ce dernier un degré de systématicité

moindre, ce qui conduit à faire le constat d’une diversité plus considérable
encore que ne le suggère Jean Wirth.14
Mais qu’est-ce qui autorise cette ‘liberté’ de l’art, dès lors que l’on n’oublie
nullement l’intervention des commanditaires, ni non plus le rôle de l’Eglise qui
fait peser sur la société une domination idéologique puissante? Le plus important
ici est de souligner que l'Eglise médiévale n’est pas un corps homogène, mais
qu’elle est au contraire, comme la société tout entière, parcourue de tensions et
animée de contradictions parfois intenses. La doctrine n’est pas davantage une.
Elle évolue, fait l'objet de débats, donne lieu à des conflits y compris au sein de
l’orthodoxie (Immaculée Conception, Pauvreté . . .). Elle se module depuis la
haute spéculation des théologiens, jusqu’aux oeuvres de divulgation, en passant
par ses mises en actes liturgiques et théâtrales, ou encore ses expressions
dévotionnelles et mystiques. Les discours cléricaux intègrent des traditions
d’abord étrangères à la doctrine, qui peuvent ensuite en faire pleinement partie,
ou rester marginales (apocryphes, légendes, conceptions folkloriques ...) .
D’autre part, l’Occident médiéval—à la différence du domaine byzantin—se
caractérise par une faible intervention normative des clercs dans le domaine des
images. Du moins est-ce le cas après la fin de la Querelle des images,15 et à
l’exception de quelques tendances limitatives, par exemple chez saint Bernard.
Certes, les clercs rappellent, en particulier dans les traités liturgiques, les
fonctions des images et évoquent parfois certaines significations des thèmes
principaux.16 Mais rares sont les interventions visant à fixer, à corriger ou à
condamner les modes de représentations. De fait, on constate une mobilité
notamment), sans qu’il soit pour autant question de remettre en cause la sainteté de ces
derniers (exemples dans les mosaïques de Monreale). Dans certaines oeuvres, un même
personnage apparaissant plusieurs fois, peut être, selon les cas, nimbé ou non. Un
traitement exhaustif de la figuration des nimbes montrerait à quel point il est difficile de
réduire les images, même sur un point aussi élémentaire, à une codification stricte et
homogène.
14 L’analyse des représentations de l’âme fournit un autre exemple où le désir de
systématicité conduit à réduire l'iconographie à des codes beaucoup trop simplifiés; cf. J.
Wirth, ‘L’apparition du surnaturel dans l’art du Moyen Age’, dans L ’image et la
production du sacré, éds. F. Dunand, J.-M. Speser et J. Wirth (Paris, 1991), pp. 139-64,
et mes remarques dans 'Anima', dans Enciclopedia dell'arte medievale (Rome, 1991), pp.
804-15.
L' Y. Christe a souligné que les choix occidentaux inspirés par les milieux carolingiens
a permis de libérer l'art chrétien occidental 'de trop fortes contraintes doctrinales’, cf.
'L'émergence d'une théorie de l'image dans le prolongement de Rm 1.20 du IXe au XIIe
siècle en Occident', dans Nicée II. Douze siècles d ’images religieuses, éds. F. Boespflug
et N. Lossky (Paris, 1987), p. 304.
l6Baschet, ‘Introduction: l’image-objet’ (cf. note 6).
64 JÉRÔME B a SCHET
figurative, qui contraste de manière saisissante avec la stabilité beaucoup plus

grande des formules dans l’art byzantin. Cette ‘liberté’ de l'art est même admise
théoriquement, comme en témoignent les reprises du Didum Horatii, notamment
dans le Rationale divinorum officiorum de Guillaume Durand.17*
Enfin, les attitudes critiques de quelques clercs, condamnant certains types
iconographiques, illustrent a contrario l’existence de cette marge de manoeuvre de
la création figurative. Elles montrent que les images opèrent parfois sur les franges
de l’orthodoxie, créant des représentations qui tantôt apparaissent licites et tantôt
inadmissibles. Parmi d’autres exemples, on rappellera l’intervention d’Antonin de
Florence contre les Annonciations montrant l’enfant Jésus, déjà formé, descendant
vers la Vierge, ou encore celle de Gerson condamnant les statues des Vierges
ouvrantes.Is Mais, pour le Moyen Age, de telles critiques constituent des points de
vue, certes éminents, mais qui n'ont pas force de décision officielle, doctrinale ou
disciplinaire.19
Plus largement, c’est l’analyse des thèmes iconographiques qui doit fonder l'idée
d’une inventivité de l’art médiéval (comme on le verra plus loin). De nombreux
thèmes iconographiques n’ont pas à proprement parler d’équivalent textuel.
Ainsi, l’Arbre de Jessé a certes des sources scripturaires, qui légitiment cette
figuration (Is 11). Mais l’image constitue une élaboration originale, synthétique,
assurant une présence massive à une mise en jeu spécifique et visuellement
puissante d’un énoncé relatif à la filiation du Rédempteur, aux rapports entre
l’humain et le divin et au rôle qu’y tient l’Eglise.20 Autre exemple: la figuration
17 'Pictoribus atque poetis/ Quidlibet audendi semper fuit aequa potestas’, cf. A.
Chastel, ‘Le Dictum Horatii quidlibet audendi potestas et les artistes (x itf-x v f siècle)’,
repris dans Fables, formes, figures, 1 (Paris, 1978), pp. 363-76. Outre la fonnulc
horaticnne, Guillaume Durand indique que les peintres représentent les scènes bibliques
‘à leur convenance’ (Sed et diversae hystoriae tam novi quam veteris testamenti pro
voluntate pictorum depinguntur, cité par A. Chastel, p. 366).
Is On ne peut développer ici ni l’analyse de ces images ni celle des protestations
qu’elles suscitent. Sur le premier exemple, voir les remarques (qu'on ne partage pas
entièrement) de J. Wirth, ‘L’apparition du surnaturel’ (cf. note 14), pp. 148-52. Sur les
Vierges ouvrantes, cf. G. Radler. Die Schreinmadonna "Vierge ouvrante" (Francfort,
1990).
19 11 en va autrement à partir du Concile de Trente, lorsque le contrôle clérical sur les
images tend à se renforcer; et le traité de Molanus est l’expression de cette nécessité
nouvelle, Molanaus. Traité des saintes images, éds. François Boespflug, Olivier Christin
et Benoît Tassel (Paris, 1996). Voir aussi la condamnation pontificale de certains types
trinitaires, aux XVIIè et XVlllè siècles, analysée par F. Boespflug, Dieu dans l ’art (Paris,
1984).
20 Sur ce thème, c f A. Guerreau-Jalabert, ‘La Vierge, l’Arbre de Jessé et l’ordre
chrétien de la parenté’, dans Marie. Le culte de la Vierge dans la société médiévale, éds.
D. Iogna-Prat, E. Palazzo and D. Russo (Paris, 1996), pp. 137-70.
des types trinitaires, dans leur variété, illustre à quel point la complexité et les
contradictions de la doctrine chrétienne ouvrent largement le spectre des
possibilités figuratives.21 Ainsi, loin du constat d’un art homogène, décalque
passif de la doctrine de l’Eglise, la vitalité et le dynamisme de la pensée
chrétienne, au sein d'une société complexe, confèrent à la création figurative une
large marge de manoeuvre. L’analyse sérielle se doit de prendre en compte (et
même de rendre compte de) cette grande inventivité.
Les types de séries
L’analyse iconographique ne saurait se fonder sur des images isolées. S’il est
vrai que les images pensent, on doit ajouter qu” elles se pensent entre elles’.22
L’étude des séries est par conséquent une nécessité, ne serait-ce que parce
qu’elle est la condition de l’apprentissage du voir, de cette 'difficile conversion
de la vision’ dont parle Otto Pacht.23 La confrontation des images d’une même
série est indispensable à l’analyse de la signification de chacune d’elle, tant il est
vrai qu’il n’y a de sens que par différence, ou plutôt dans une dialectique des
régularités et des écarts.24 Enfin, une part importante de la signification d’une
oeuvre se trouve dans son rapport avec d’autres, ou pour le dire avec Hubert
Damisch, ‘ce qui importe est moins ce qu’une oeuvre d’art représente que ce
qu’elle transforme’.25
Trois espèces principales de séries peuvent être distinguées: l’ensemble
d'images produit par l’oeuvre elle-même; le corpus thématique construit par le
chercheur; les hyperthèmes, réseaux complexes également élaborés par le
chercheur. S’agissant du premier aspect, on rappelera que les images médiévales
sont rarement conçues de façon isolée (contrairement à ce que produit le
tableau). Elles s’inscrivent généralement dans un lieu ou sur un objet, au sein
duquel se forme un complexe d’images. Ainsi, dans la série des miniatures et des
lettres qui ornent un manuscrit, le sens ne peut se construire qu’à travers
21 Cf. F. Boespflug and Y. Zaluska, 'Le dogme trinitaire et l'essor de son iconographie
en Occident de l’cpoque carolingienne au IVe Concile de Latran'. Cahiers de civilisation
médiévale, 37 (1994), 181-240.
22 H. Damisch, 'La peinture prise au mot’. Critique, 370 (1978), 289 (reprennant ici la
formule de C. Lévi-Strauss concernant les mythes).
23 O. Pacht, Questions de méthode en histoire de I 'art, trad, ffse (Paris, 1994), p. 40.
24 Sur ce modèle de la connaissance, inspiré par la linguistique, cf. C. Lévi-Strauss,
lorsqu'il indique que l’objet de l’ethnologie est moins les sociétés elles-mêmes que 'les
écarts différentiels’ entre elles: Anthropologie structurale (Paris, 1957 : réimpr. 1974), p.
384.
25 H. Damisch, Le Jugement de Paris. Iconologie analytique I (Paris, 1993), p. 168.
l’analyse des constantes et des singularités, des échos entre les images, des
processus de transformation qui accompagnent la lecture du livre.26 Dans le cas
des cycles monumentaux, l’analyse doit faire intervenir d’autres paramètres
spécifiques, afin de rendre compte d’un ’lieu d’images’ (régularités narratives,
associations thématiques entre les scènes, relations entre les images et leurs lieux
d’inscription, perçus tant dans leur valeur symbolique que dans leur fonction
liturgique).27
L’importance fondamentale de cet aspect de la recherche ne saurait être mise
en doute, mais la nature du présent exposé conduit à insister surtout ici sur le
second type de séries. A côté des réseaux d’images constitués par l’oeuvre,
l’historien doit aussi construire des séries d’images appartenant à des objets
différents, dans des lieux et des temps délimités de façon précise. C’est du reste
une démarche constitutive de l’histoire de l’art classique, intéressée principale
ment par les phénomènes stylistiques: Otto Pächt a insisté sur le caractère
génétique de l’histoire de l’art, qui vise à inscrire chaque oeuvre dans le
processus du devenir des formes artistiques. Elle doit donc construire des ‘séries
généalogiques au sein desquelles chaque oeuvre trouve sa place et son sens.* 2"
Toutefois, on admettra que la recherche de ‘modèles’, largement pratiquée par
l’histoire de l’art, s’est souvent fourvoyée, en particulier lorsqu’elle pense la
production figurative de façon passive, en terme d” influence’. Cette démarche,
obnubilée par la recherche des similitudes, conduit généralement à négliger les
différences entre l’oeuvre et ses antécédents, alors que celles-ci sont au moins
aussi importantes que les points communs. Elle s’interdit ainsi de saisir le
processus de transformation, qui devrait pourtant être au coeur de l’approche
sérielle.
On ne saurait pour autant oublier que l’art médiéval, et plus largement la
culture médiévale, fonctionne sur la reconnaissance d’une forte valeur de
traditionnalité. Le prestige d’une oeuvre dépend souvent de sa révérence à
l’égard d’un prototype vénérable, ce qui n’empêche pas l’artiste de l’adapter, de
le transformer sous couvert de cette marque d’hommage. Herbert Kessler a bien
souligné que cette traditionnalité de l’art devait être pensé, non sous la catégorie
passive de modèle, mais à l’aide de la notion active de ‘citation’ (soit une forme
26 Cf. un cas de métamorphose réglée et de progression sérielle, analysé par J.

Hamburger, The Rothschild Canticles. Art and Mysticism in Flanders and the Rhineland
circa 1300 (New Haven: Yale U.P.. 1990). Dans une étude à venir. H. Kessler élabore,
à propos des miniatures de la Bible de Vivien, un modèle particulièrement subtil de
fonctionnement du manuscrit.
2' Je me permets de renvoyer sur ces aspects à mon livre Lieu sacré. Heu d ’images. Les
fresques de Bominaco (Abruzzes. 1263). Thèmes, parcours, fonctions (Paris-Rome: La
Découverte—Ecole française de Rome, 1991).
2ii Pacht. Questions de méthode (cf. note 23), notamment p. 163.
Plates 67
Plate 1: ‘Arnolfini arid his wife’, Jan van Eyck. 1434. OH on oak, 81.8 x 59.7 cm.
National Galleiy, London. NG 186.
68
Plate 2: ‘Lucca Madonna', Jan van Eyck. c. 1436. Oil on wood, 65.5 x 49.5 cm.
Städelsches Kunstinstitut, Frankfurt.
Piales 69
Plate 3: The mystical press (c. 1511): the image o f the pope echoing that o f God
the Father.
70
Plate 4: Heniy the Lion 's Gospels: Henry and Mathilda crowned by the hand oj
God. Wolfenbüttel, Herzog August Bibliothek, MS. Guelf. 105, Noviss. 2°, fo i
171".
Platea 71
Plate 5: Henry the Lion ’s Gospels: divine majesty and the creation o f the world.
Wolfenbüttel. Herzog August Bibliothek. MS. Guelf. 105, Noviss. 2°,fol. 172'.
72
Plate 6: Cornell box, default lighting.
Plate 7:Cornell box, raytraced rendering.

Plates 73
Plate 8: Cornell box, radiosity rendering.
Plate 9: Maya camera icon.

74
Plate IO: 'St Jerome in his study’, Antonello da Messina, c. 1450.

Piales 75
Plate 11: Wireframe.
Plate 12: Untextured three-dimensional form.

76
Plate 14: Jerome.

Piales 77
Plate 16: Jerome.

78
r u
Plate 18: Jerome.

Plates 79
Plate 19: Jerome.
Plate 20: Haus Altenkamp, räumliche Darstellung des Entrées überlagert mit
Schadenskartierung, Ansicht von Nordwesten.
80
Plate 21: Haus Altenkamp, Entrée, Wandmalerei-Ausschnitt der Südwand, Über

lagerung der Farbauszüge. Das Farbbild wurde aus den schwarz/weißen
Spektralbildern berechnet. Durch Kalibrierung des fotografischen Prozesses läßt
sich so eine langzeitstabile, rekonstruierbare colormetrische Abbildung reali
sieren.
Plate 22: Das Thermobild setzt sich aus 4 Einzelbildern zusammen. Vom Objekt
reflektierte Strahlungen im Bereich zwischen 8 und 14 mm wurden registriert.
Die Abbildung zeigt Kältebrücken an den Seiten der Wände (Anschluß zur
Außenwand bzw. zu unbeheizten Nachbarräumen). Der wärmere (grüne) Bereich
in Bildmitte weist darauf hin, daß durch einen offenen Kamin im Nachbarraum
Wärme abgesogen wird.
Plates 81
Plate 23: Haus Altenkamp, Entrée, Thermobild der Südwand, überlagert mit
Schadenskartierung.
Plate 24: Die Überlagerung korreliert die Schadenskartierung mit bauphysikal

ischen Aussagen.
82
Plate 25: Haus Altenkamp, Entrée, nach der Restaurierung.
Plate 26: Representation o f a view from the west balcony o f the first Jacobean
Banqueting House (c. 1607-1619) looking towards Inigo Jones ’ sceneiy for The
Vision of Delight (1617). Computer-generated model, Ross Parry, 1999. Details
from Inigo Jones, ‘A Street in Perspective’ and 'Proscenium and Hunt Scene’,
Devonshire Collection, Chatsworth.
d’intertextualité).29 Il importe en outre d’analyser l’enjeu de telles citations.30 la

référence à une tradition peut porter une valeur idéologique, constituer une
revendication politique (ainsi la reprise par les rois normands de modèles
byzantins impériaux),31 indiquer une dépendance institutionnelle (s’agissant de la
diffusion des modèles romains),32 ou un désir de Réforme spirituelle et
ecclésiologique (à travers la citation de modèles paléochrétiens.33 Enfin, on
suggérera que la tradition artistique est peut-être, paradoxalement, l’un des
facteurs qui contribuent à la liberté de l’image: le caractère vénérable attaché à
cette tradition, dont les hommes de l’art sont en partie les dépositaires, est un
paramètre important dans le processus conduisant à l’oeuvre finale, à côté de la
volonté du commanditaire et éventuellement des références textuelles obligées.
C onstruire une série
Bien que l’approche sérielle concerne également les approches stylistiques ou

formelles, c’est surtout aux séries iconographiques qu’on s’intéressera ici, et en
particulier aux séries quantitativement amples.34 Dans les conditions actuelles de
la recherche, construire des séries rassemblant la totalité des sources conservées
reste une tâche inaccessible. Aboutir au résultat le plus satisfaisant possible
demande encore un investissement en temps (et parfois en argent) considérable.
En dépit d’instruments précieux, au premier rang desquels l’indispensable Index
29 H. L. Kessler, ‘The State of Medieval Art History', The Art Bulletin, 70 (1988),
notamment p. 176.
j0 On tiendra compte de la gradation existant entre des oeuvres plus fortes, plus
innovatives, et d’autres plus imprégnées de formules établies, parfois produites de façon
répétitive (mais admettant néanmoins des variations infimes, et rarement analysables
comme simple reproduction).
31 E. Kitzinger. ‘The Gregorian Reform and the Visual Art: a Problem of Method’,
Transactions o f the Royal Historical Society, 22 (1972), 87-102.
'2 Kessler, ‘The State of Medieval Art History’ (cf. note 29). pp. 176-78.
” H. Toubert, ‘Le renouveau paléochrétien à Rome au début du X lf siècle’, repris dans
Un art dirigé. Réforme grégorienne et iconographie (Paris, 1990).
’4 D'autres séries, pour être plus restreintes, n'en sont pas moins dignes d’intérêt.
Ainsi, dans Le Jugement de Paris (cf. note 25), H. Damisch suit, à travers une série
d'oeuvres exceptionnelles, les métamorphoses d’un modèle de l’Antiquité jusqu'à
Picasso, en passant par Raphaël et Manet. Il y définit son objet comme ‘tresse’, plutôt que
comme série, pour en souligner le caractère imbriqué, excluant ‘tout principe
d’explication linéaire’ (p. 221). Il y a là un cas assez différent de ce que les images
médiévales nous donnent généralement à travailler: une série de chefs-d’oeuvre qui se
pensent les uns les autres, et non un ample corpus d'objets présentant entre eux une
gamme de relations beaucoup plus variée.
o f Christian Art de Princeton, l’état actuel des outils de recherches

documentaires (bases de données, iconothèques, catalogues des manuscrits
enluminés, corpus des oeuvres monumentales, banques d’images, CD-Rom,
serveurs internet . . .), reste très insatisfaisant en dépit des progrès accomplis, et
limite malheureusement la portée de la démarche sérielle. On ne peut donc
qu’insister sur la nécessité d'accélérer la réalisation d'outils dont la recherche a
besoin et que les techniques modernes permettent de produire. Outre leur vertu
première de recherche documentaire, ces outils, permettant une visualisation
rapide des images concernées, favorisent les rapprochements, les confrontations
et les croisements entre tous les fils de cet échevau que constitue la production
visuelle médiévale. Ils sont les vecteurs privilégiés de l’approche sérielle. Ils
doivent tout à la fois la servir et l’encourager.
Construire une série d’images suppose, dans une première étape, de
rassembler, par tous les moyens disponibles, un corpus aussi exhaustif que
possible. Une étude iconographique qui ne reposerait pas sur la prise en compte
de toutes les oeuvres pertinentes accessibles serait sinon illégitime, du moins
fragile. Mais construire une série, c'est ensuite, à rebours, dissocier, fragmenter
la masse des documents obtenus. Car le corpus rassemblé est constitué
d’éléments fortement hétérogènes, qui ne se laissent pas traiter au sein d’un
ensemble unifié.
La distinction par support (manuscrits enluminés, sculpture, peinture murale,
vitrail, orfèvrerie . . .) s’impose pour deux ordres de raison. Chaque technique
possède ses règles et ses procédés formels propres, jouant de la valeur et des
possibilités plastiques des matériaux utilisés, et des formats mis en oeuvre.
Surtout, la distinction par support impose de considérer les types d’objets ou les
lieux architecturaux auxquels sont liées les images. Elle fait intervenir les usages
des images-objets, les relations éventuelles entre thèmes représentés et fonctions
des objets, ainsi que les modalités de visibilité et de réception.35
La distinction selon les contextes thématiques est, dans certains cas au moins,
indispensable. En effet, l’analyse doit procéder différemment selon qu’elle a
pour objet un thème iconographique ou un motif iconographique. Un thème,
comme l’Annonciation ou la Remise des tables de la Loi à Moïse, constitue une
unité structurale élémentaire possédant une cohérence propre, pouvant faire
l’objet d'une scène spécifique, notamment dans une miniature, une lettre ornée
35 Enfin, ces distinctions permettent d'évaluer l’importance accordée au thème

concerné—laquelle se mesure moins aux dimensions et à la visibilité de l'oeuvre qu'au
degré de sacralité de l'objet ou du lieu auquel l’image est attachée. Cette dernière question
prend une acuité particulière pour les motifs négatifs, comme l'enfer ou le diable, dont
l’intégration dans un objet ou un lieu chargé de sacralité est hautement problématique: je
renvoie à mon article ‘L'enfer en son lieu: rôle fonctionnel et dynamisation de l'espace
cultuel', dans Luoghi sacri e spazi della santità. éds. S. Boesch-Gajano et L. Scaraffia
(Turin. 1990). pp. 551-63.
ou un cycle mural.34 En revanche, on appelle motif iconographique un élément

constituant d’un thème. Il en va ainsi de l'enfer qui, pour l'essentiel, apparaît à
l’intérieur d’ensembles plus complexes, comme le Jugement dernier, la Parabole
de Lazare ou la Chute des anges.3637 La distinction entre thèmes et motifs est
indispensable, môme si elle n’est pas rigoureusement étanche. Conformément au
principe structural de prééminence des relations sur les éléments constituants, on
ne saurait aborder un motif iconographique sans une analyse globale du thème
dans lesquels il s’insère. C’est le seul moyen de mesurer l’importance du motif,
de saisir sa fonction dans un ensemble cohérent, et par conséquent d’atteindre un
point pertinent de signification.
Les distinctions selon les temps et les lieux sont également importantes. Une
étude de la répartition chronologique (à la fois globale, et en tenant compte des
distinctions par thèmes et par supports) fait apparaître des phénomènes décisifs
pour l’étude historique. Elle permet de distinguer le nouveau du traditionnel,
d'identifier la période d’apparition du thème ou du motif étudié, de préciser ses
rythmes de développement, et éventuellement de déclin,38 et enfin de saisir les
évolutions et les variations géographiques de ses modes de représentations.
En outre, la fragmentation du corpus doit être poussée à l’extrême, dès lors
que chaque oeuvre singulière constitue un cas particulier, intégrant le thème
étudié dans une série d’images spécifique, en rapport avec un objet unique et
dans un contexte chaque fois particulier. Il ne saurait, en principe, y avoir
d’étude sérielle sans une approche structurale de chaque oeuvre singulière
appartenant à la série. Approche structurale et étude sérielle sont totalement et
intimement entremêlées. Chaque image doit même être située au croisement de
multiples séries (série dans l’oeuvre, environnement de celle-ci, séries
thématiques, traditions formelles . ..). Chaque série étant elle-même complexe et
non-linéaire, c’est une sorte de sphère de relations qu’il faut dessiner autour de
36 Ce qui n'cxclut pas que le thème soit intégré dans une série constituant un ensemble
structural plus ample. En outre, de même qu'un personnage peut synthétiser plusieurs
identités differentes, une unité thématique peut condenser différents thèmes, et
notamment plusieurs moments généralement dissociés.
37 Le fait qu’un motif puisse être parfois représenté de façon indépendante, devenant
ainsi un thème à lui seul, n’invalide pas la distinction proposée. Un tel processus
d’autonomisation doit au contraire être analysé comme un phénomène d'une grande
portée.
’s L’étude du développement quantitatif d’un thème devrait tenir compte non de
valeurs absolues, mais de valeurs relatives référées au nombre global d’images produites
(ou disons conservées). II faut donc prendre en considération la forte croissance de la
production d’images, entre xiè et xvè siècle. En ce sens, le développement régulier d'un
thème au cours de cette période est difficilement interprétable. En revanche, un déclin
quantitatif n'en prend que plus de relief.
Fig. 1. Evangéliaire, Bibliothèque vaticane, ms. Vat. lat. 39. fol. 58. ‘Lazare
dans le Sein d'Abraham' (troisième quart du XIIIe siècle). © Biblioteca
apostolica vaticana
Fig. 2. Psautier de Blanche de Castille. Paris. Arsenal, ms. 1186. fol. 171. 'le Sein
d'Abraham et l ’enfer’ (c. 1225). © Bibliothèque nationale de France. Paris
88 J é r ô m e B a s c ii e t
chaque image singulière. L’enjeu et la difficulté de l’étude iconographique

consistent à faire tenir ensemble l’approche sérielle et l’étude structurale des
oeuvres singulières; elle doit pratiquer d’incessants allers-et-retours entre ces
deux démarches.
Au total, le caractère fortement hétérogène des éléments constitutifs d’une
série iconographique aboutit à un fractionnement extrême de celle-ci. En matière
d’images, le sériel est un faux quantitatif. Il semble n'appeler que des traitements
mathématiques frustes. Il impose de compter les occurrences en fonction des
paramètres retenus, mais de façon nécessairement très approximative
(notamment en raison des incertitudes de datation), avec pour seul objectif
raisonnable d’évaluer et de comparer des ordres de grandeur. Les études
iconographiques doivent apprendre à compter plus qu’elle ne font généralement,
et à considérer l’importance quantitative des phénomènes étudiés. Pourtant, il
semble généralement prématuré de plaider pour un véritable traitement
statistique, en raison des marges d’incertitudes, des facteurs d'hétérogénéité et de
la faiblesse des masses concernées.39*
A n a lyse r une g a m m e sérielle
L’approche sérielle vise à cerner régularités, variations, singularités et exceptions.

Dans une première démarche, un classement typologique est souvent nécessaire et
permet de faire apparaître des constantes. On peut ainsi, par exemple, distinguer
parmi les représentations du sein d’Abraham trois types formels principaux, selon
que le patriarche rassemble les élus entre ses bras (fig. 1), dans un linge (fig. 2) ou
dans un pan de son vêtement (fig. 3).J0 Mais l’analyse sérielle ne peut s’en tenir à
39 Un véritable traitement statistique peut être envisagé dans quelques cas privilégiés,
concernant un motif simple, très fréquent, ou pour lequel les facteurs d’hétérogénéité
jouent un rôle moindre qu'à l'habitude. Les cas déjà évoqués des représentations de l’âme
ou du nimbe pourraient être de bons exemples de séries analysables statistiquement. Voir
aussi les traitement quantitatifs réalisés à partir de la base de donnée des peintures
murales danoises, J. J. Borrild, ‘Medieval Danish Wall Paintings. An Internet Database',
Medium aevum quotidianum (cf. note 1), pp. 21-36.
Jü Respectivement: Evangiles, Bibliothèque vaticane. Vat. lat. 39, f. 58 (Italic du Nord,
troisième quart XIIIe siècle); Psautier de Blanche de Castille, Paris, Arsenal MS. 1186, f.
171' (Paris, vers 1225); Légcndicr dominicain. Oxford, Keble College, MS. 49, f. 239
(Ratisbonne, vers 1270). Ces trois oeuvres relèvent de contextes thématiques différents:
Jugement dernier. Parabole de Lazare, sein d’Abraham isolé en rapport avec la fête des
morts. Sur le sein d'Abraham, figuration du lieu paradisiaque promis aux élus, je renvoie
à une première esquisse dans ‘Medieval Abraham: Between Fleshly Patriarch and Divine
Father’, Modern Language Notes, 108 (1993), 738-58, ainsi qu'à Le Sein du Père.
Abraham et la paternité dans l'Occident médiéval (Paris: Gallimard, 2000).
3. Légendier dominicain, Oxford, Keble College, MS. 49, fol. 239: ‘le Sein
d'Abraham' (c. 1270). © Keble College Library. Oxford
de telles classifications, qui semblent reproduire une notion de ‘type
iconographique’ rigidement fixé, qui doit être regardée avec méfiance. L’approche
sérielle montre au contraire qu’il n’existe que des gammes de réponses.*' Dans le
cas du sein d’Abraham, l'analyse des trois grands ensembles définis fait apparaître.41
41 On pense ici à la notion de multiplicité, précisée par Deleuze, à partir de l’oeuvre de

Foucault, G. Deleuze, Foucault (Paris. 1986), pp. 11-30; M. Foucault. L'archéologie du
savoir (Paris, 1969).
90 Jé r ô m e B a s c h e t
ici les variations des formes du linge, leur degré de concavité, les effets ainsi
produits, là la façon de tenir les élus, enfin les modalités et le degré plus ou moins
intense d’inclusion dans le manteau. A un premier regard, la typologie met en
évidence des constantes, tandis que l’analyse sérielle vise à intégrer, dans la
régularité d’une série, la singularité de chaque oeuvre, qui combine des réponses
particulières à un ensemble d’options possibles, et qui noue un réseau propre de
tensions et de contradictions. En articulant régularité et variations, à travers la
construction de gammes, on entend sortir de la dualité de l’original et du banal, du
singulier et du répétitif, critiquée par Michel Foucault.'12 On espère ainsi approcher
ce qu’il y a d’actif à l’intérieur de séries que l’on refuse de tenir pour des types
iconographiques constants.4 243
En conséquence, ce n’est pas nécessairement dans les chefs-d’oeuvre que
l’analyse iconographique trouve le plus de profit, même si ceux-ci sont parfois
d’une richesse qu’on aurait tort de mépriser. Se détourner des oeuvres majeures
ne signifie nullement que l’approche sérielle a pour vertu principale de faire
apparaître des constantes qui pourraient sembler à l’historien plus révélatrices
que les singularités. Car l’approche sérielle permet aussi de repérer, dans des
oeuvres apparemment assez banales, des variations qui, tout en étant discrètes,
s’avèrent pour l’analyse iconographique d’une grande portée. Plus largement
encore, la démarche sérielle montre que, au sein d’une gamme de variations, il
n’y a pas de régularités sans ondoiement des possibilités de transformations. Les
régularités sont ‘en alerte’; l’inventivité vient se nicher dans les micro
déplacements, dans les décalages infimes.
La série se donne comme une exploration quasi-systématique d’un champ de
possibilités. Ces champs de possibilités, objets spécifiques de l’approche sérielle,
ne sont pas infinis, mais ils sont d’une ampleur généralement surprenante. Ce
n’est pas l’effet d’une pure liberté du créateur ou d’une simple mécanique
combinatoire, mais plutôt la conséquence du caractère hétérogène des oeuvres
qui constituent la série. Les variations de matériaux, de localisation, de format,
d’usage, d’environnement, d’époque, de relation avec les traditions figuratives
(etc.) contribuent à faire varier la mise en oeuvre d’un ensemble thématique.
L'hétérogénéité des oeuvres, faisant jouer l’élément iconographique observé, est
comme le kaléidoscope qui permet à l’historien d’explorer les multiples facettes
de son objet. La gamme sérielle est le champ expérimental mettant à l’épreuve la
régularité et la variabilité de l’objet.
42 Foucault, L 'archéologie du savoir (cf. note 41 ), pp. 184-94.

43 Le banal lui-même est actif: ‘la régularité n’est pas moins opérante, n’est pas moins
efficace et active, dans une banalité que dans une formation insolite’ (cf. note 41), p. 189.
Dire que ‘tout le champ énonciatif est à la fois régulier et en alerte: il est sans sommeil’,
p. 191, c’est aussi reconnaître que la régularité des énoncés n’est jamais une pure identité,
une simple reproduction.
Fig. 4. Bible de Painpehtne. Amiens, Bibliothèque municipale, ms. 108, fol. 255',
‘le Sein d ’Abraham ’ (1197). © Bibliothèque municipale, Amiens
92 JÉRÔME B A S C H E !
Outre les gamines de variations, l’étude sérielle fait considération des cas les
plus radicaux, en ayant soin d’en distinguer plusieurs espèces. Certaines s’in
scrivent aux points extrêmes d’une gamme de variations. Dans le cas du sein
d’Abraham, il s’agit notamment des images qui poussent le plus loin l’inclusion
des élus dans le corps-vêtement du patriarche (fig. 4).44 Sujets de prédilection de
l’analyse, de telles images semblent révéler, plus clairement que les autres, la
dynamique à l’oeuvre dans la gamme sérielle. D’autres cas singuliers se placent
plutôt en marge des séries. Ainsi, les hybridations iconographiques constituent
une intersection entre deux séries. Lorsque, dans une image du Trône-de-Grâce,
le corps du crucifié disparaît à l’intérieur du manteau du Père (fig. 5),4546il s'opère
un croisement entre deux types trinitaires, le Trône-de-Grâce et la Paternité
divine, qui montre le Fils tenu in sinu patris (Jn 1,18, entendu ici au sens de
‘dans le pli du Père’; fig. 9).4f’ Il s’agit là d’une solution exceptionnelle, qui
relève de cette exploration systématique des possibles à laquelle l’étude sérielle
nous fait assister.47*
Enfin, on qualifie d” images-limites’ les oeuvres exceptionnelles qui se singu
larisent par leur radicalité au sein de leur série, et s’aventurent parfois jusqu'aux
marges de l’orthodoxie. Une des plus célebres est la Quinité de Winchester, qui a
jadis retenu l’attention de E. Kantorowicz.4S On évoquera ici l’exemple du
Jugement dernier du Camposanto de Pise.49 L’oeuvre est singulière, à la fois
44 Bible de Pampelune, Amiens, Bibliothèque municipale, ms. 108, f. 255 v

(Pampelune, 1197).
45 Psautier de Stephen de Derby. Oxford. Bodleian Lib., MS. Rawlinson G. 185, f. 97
(milieu xivc siècle).
46 Sur ce thème, cf. infra. Le terme d’hybridation utilisé par J. Wirth (cf. note 10, pp.
16-17), convient bien dans un tel cas, concernant la jonction de deux types
iconographiques. Mieux peut-être qu’au cas de la Judith-Salomé étudiée par J. Wirth: en
effet, une figure ambivalente constitue la somme des deux figures initiales (A + B); le
résultat d’une hybridation tient de A et de B, sans être entièrement ni A ni B.
47 Lorsqu’un élément ou une fonction caractéristique d’un personnage glisse vers une
autre figure, on parlera plutôt de transfert iconographique. C’est le cas au XVe siècle,
lorsque le linge contenant les élus n’est plus tenu par Abraham, mais par la figure
paternelle du Trône-de-Grâce: soit une image du sinus trinitatis. Un tel phénomène peut
révéler des relations, notamment d’homologie, entre les figures concernées, comme ici
entre Abraham et Dieu le Père. Sur ce type iconographique, voir (avec quelques réserves);
P. Sheingorn, ‘The Bosom of Abraham Trinity: A Late Medieval All Saints Image’, dans
England in the 15th Century. Proceedings o f the 1986 Harlaxton Symposium, éd. D.
Williams (Londres. 1987). pp. 273-85.
4S E. Kantorowicz, ‘The Quinity of Winchester’. The Art Bulletin. 29 (1947), 73-85; et
J. Kidd, ‘The Quinity of Winchester Revisited', Studies in Iconography, 7-8 (1981-82),
21-32.
49 Je renvoie à mon travail Les justices de l'au-delà. Les représentations de l'enfer en
Fig. 5. Psautier de Stephan de Derby, Oxford, Bodleian Libraiy, MS. Rawlinson

G. 185. fol. 97, 'le Crucifié dans le Sein du Père' (milieu xiv" siècle). © Bodl
eian Library, Oxford
parce qu'elle cristallise une mutation décisive dans l'histoire des représentations
infernales, mais aussi par certains aspects d'une telle radicalité qu’ils restent sans
postérité. Ainsi, la Vierge quitte sa position habituelle d’intercession pour
prendre place à côté du Christ, dans une mandorle identique à celle de son Fils.
La figuration d’un tel couple judiciaire est un phénomène extraordinaire, bien
éloigné des indications scripturaires et de toute la tradition théologique et
figurative relative au Jugement dernier. On doit en donner une interprétation
France et en Italie (xif-.xF siècles). Bibliothèque des Écoles Françaises d'Athènes et de

Rome, 279 (Rome, 1993).
94 J érôme Baschet
multiforme, par référence à la stucture exceptionnellement bipolaire de l’oeuvre

(Jugement d’un côté, enfer de l’autre), en rapport avec l’essor du culte de la
Vierge, ou encore en termes politiques (ralliement de Pise au camp guelfe dans
les années 1330, qui fait de la figure mariale un enjeu civico-ecclésial majeur).
On songe également à la dualité Justice (qui condamne)/Miséricorde (qui
pardonne), habituellement fondue dans la seule figure du Christ, mais qui
apparaît ici. de façon clivée, à travers le couple Vierge/Christ. A cet égard, cette
image révèle une tension, entre le désir de miséricorde qui anime une 'religion
d’amour’ et l’affirmation d’une exigence de châtiment, qui travaille en pro
fondeur le christianisme médiéval.50
Voilà bien l’enjeu de ces images-limites et de tous les types d’oeuvres excep
tionnelles dont il vient d’être question. Il s’agit de parvenir à articuler de façon
satisfaisante les régularités sérielles et les cas extrêmes. L’historien serait bien aise
de pouvoir attribuer aux images extrêmes une capacité à révéler des contradictions
que les oeuvres plus courantes tiennent généralement masquées. Elles seraient
ainsi comme les porte-voix de la majorité silencieuse des images. Mais la prudence
s’impose: à faire parler une oeuvre pour toutes les autres, on risque de prêter à la
série une homogénéité qu’elle ne possède pas. Or, il convient de souligner que
l’image extrême s’inscrit dans un moment précis et exprime les tensions d’un
contexte singulier (comme le revirement politique de Pise, lors de la réalisation des
fresques de Buffalmacco). C’est dans un second temps seulement, et en procédant
par recoupements, que l’on peut éventuellement attribuer à l’image extrême une
valeur plus large. La compréhension conjointe des gammes sérielles et des images
exceptionnelles impose d’articuler le local et le général. Il s’agit d’analyser, dans
ses dynamiques historiques, un champ de possibilités où jouent ensemble le
singulier et le régulier, l’extrême radicalité des coups d’éclats et la palpitation
tranquille des modestes variantes.
Vers l ’étude des hyperthèm es
On veut enfin évoquer l’étude des réseaux formés par un ensemble de thèmes
iconographiques—ce que l’on désignera sous le nom d"hyperthème’, c’est-à-
dire un réseau formé par un ensemble coordonné de plusieurs thème icono
graphiques. Au total, on aura marqué trois paliers dans l’analyse iconographique:
les motifs (constituants élémentaires d’une organisation thématique), les thèmes,
les hyperthèmes (jonction de plusieurs unités thématiques). Une fois encore, ces
distinctions ne doivent pas être appliquées de façon trop rigide, car l’étude peut
50 Les représentations de Satan s’aventurent également parfois jusqu'aux marges de

l’orthodoxie, en prennant un tour quasi dualiste; je renvoie à 'Satan, prince de l’enfer: le
développement de sa puissance dans l’iconographie italienne (xuf-xv0 siècle)’, dans
L autunno dei diavolo, éds. É. Corsini et E. Costa (Milan, 1990), pp. 383-96.
associer plusieurs niveaux, et certains hypcrthèmes peuvent être constitués à la

fois de thèmes et de motifs. D’autre part, on note que l’étude d’un motif—
comme l’enfer, la bouche, ou les anges—et celle d’un hyperthème, sont l’une et
l’autre transthématiques, obligeant à prendre en examen plusieurs thèmes.
Néanmoins, la triple distinction proposée n’est pas inutile: la difficulté dans
l’analyse d’un motif tient aux conditions d’extraction d’un élément commun
appartenant à des unités structurales distinctes, tandis que l’étude d’un
hyperthème vise à saisir les relations entre des unités structurales cohérentes.51
Le projet d’une approche hyperthématique évoque la démarche de Jean Wirth
qui, dans L'image médiévale, tente de mettre en évidence ‘un système icono
graphique’, ‘un réseau de relations qui produit du sens par sa logique propre’.52
Sans vouloir forcer la pensée de Jean Wirth, qui reconnaît les limites de
l’entreprise, il semble bien que sa visée (son désir) le pousse, en accord avec
l'hypothèse d’un fonctionnement logique de l’image, vers la recherche d’une
unité structurale globale de l’iconographie médiévale (du moins à chacune des
époques considérées). Or, une telle perspective risque de surestimer la
systématicité de l’objet étudié, et de laisser échapper une part de la spécificité
des images singulières (peut-être aussi parce qu’elle néglige parfois d’en con
struire les séries aussi complètement que possible). Pour le dire avec J.-C.
Bonne, en matière d’iconographie, ‘il y a du systématique mais certainement pas
un système, des réseaux thématiques et même des réseaux de réseaux (des
rhizomes) que les oeuvres tissent entre elles mais certainement pas un réseau des
réseaux qui puisse être suturé de bout en bout’.53 C’est pourquoi l’étude des
hypcrthèmes se place à un niveau de globalité inférieur à celui que vise Jean
Wirth. On se propose en effet d’inclure dans le projet iconographique l’étude de
réseaux de thèmes, sans pour autant supposer qu’ils forment des systèmes
parfaitement cohérents, logiques, et en excluant la visée d’un système global de
51 Ajoutons cependant que l’étude d’un motif peut tendre vers celle d’une formation
hyperthématique. En effet, comme on l’a dit, elle suppose nécessairement l'analyse de la
structure d'ensemble dans laquelle s’inscrit le motif, de sorte que l'ctude de ce dernier,
dans la totalité de scs contextes thématiques, revient d'une certaine façon—mais plus ou
moins nettement selon les cas—à mettre en relation tous les thèmes concernés.
52 L ’image (cf. note 10). p. 230 (‘Il existe en somme un système iconographique. . . .
On a essayé d’évaluer son degré d’autonomie et le jeu complexe des initiatives qui le
fondent. Il reste à voir jusqu’à quel point il se structure logiquement pour produire ses
propres significations’), et pp. 231-33 et 261.
53 ‘A la recheche’ (cf. note 11), pp. 371-72. Il en va de même que de la société selon
Lévi-Strauss: il y a du structurel, des structures, mais elle n’est pas une structure, ni même
une série de structures homologues. Elle est 'un ordre des ordres': un réseau de structures,
unies par des relations extrêmement complexes, et qui toutes ensemble ne suffisent encore
pas à rendre compte de la totalité d'une société (Anthropologie structurale (cf. note 24),
ch. XV—XVI, notamment pp. 390-92).
96 J !■ RÓ M E B a SCMET
Fig. 6. San/a Maria de 7'ahull, fresque absidiale (conservée an Musée d'Art de

Catalogne, Barcelone). ‘Vierge à l'enfant et adoration des mages' (xif siècle).
©Author
l’iconographie des périodes considérées.

Rappelons d’abord que les mises en relation de thèmes relèvent d’une
pratique omniprésente dans la culture chrétienne médiévale: correspondances
typologiques entre les deux Testaments, mises en relation d’événements
multiples par la liturgie. D'autre part, les cycles figuratifs, en particulier dans les
peintures murales, sont, à travers les échos, les parallèles ou les oppositions
qu’ils tissent, des outils à penser les relations entre les thèmes iconographiques.54
L’exégèse et l’iconographie se donnent elles-mêmes comme pensée de
l'intcrthématicité. L’étude des hyperthèmes peut s’appuyer sur ces caractères;
elle s’inscrit dans le prolongement de ce constat, mais pousse au-delà de ce que
les oeuvres singulières formulent explicitement.
54 Cf. Lieu sacré, lieu d ’images (cf. note 27). Aux exemples déjà évoqués, on ajoutera
le vis-à-vis fréquent (établi par exemple à la Martorana de Païenne) entre la Nativité
montrant le Christ enfant dans les bras de sa mère et la Dormition où Pâme-enfant de la
Vierge est cette fois dans les bras de son Fils. 11 y a là, explicitée par l’oeuvre elle-même,
un lien structural d’inversion entre deux thèmes iconographiques, qui produit une
signification nouvelle, inéductible à la somme de celle des deux thèmes isolés.
Fig. 7. Heures de Jeanne de Navarre. Paris. Bibliothèque nationale de France,

ms. nouv. acq. lat. 3145. fol. 3V, ‘reine d'Angleterre priant devant la Vierge à
l'enfant et le Trône-de-Grâce © Bibliothèque nationale de France, Paris
Pour illustrer ce qu’est un hyperthème,55 j ’évoquerai le cas dont je poursuis

actuellement l’étude, étant entendu que ce qu’on dira ici, de façon sommaire, ne
peut valoir comme démonstration quant aux significations mises en jeu, mais
seulement à titre de prétexte permettant d’esquisser quelques remarques
méthodologiques. L’hyperthème considéré concerne les représentations de la
parenté divine (c’est-à-dire l’ensemble des liens pensés en termes de con
sanguinité ou d’alliance, unissant les figures surnaturelles, ou celles-ci et les
hommes, ainsi que les liens de parenté spirituelle entre les hommes, qui sont en
rapport étroit avec les précédents). Sont ainsi englobées, entre autres, les
figurations de la Trinité, les relations entre Vierge et Trinité, entre Vierge et
Christ (filiation réciproque et alliance), ou les figurations de la Mater Ecclesia
(dans ses relations avec le Christ et avec les fidèles).
A l’intérieur de cette configuration, on peut délimiter un sous-ensemble, qui
constitue également un hyperthème et dont la présentation sera plus aisée. Cet
hyperthème est le paradigme dérivant de l’expression ‘dans le sein de’ (in sinu . .
.), qui s’avère être un opérateur de conjonctions particulièrement fécond. Il
regroupe des éléments se caractérisant à la fois par l’expression d’un lien de
filiation et par la représentation d’un personnage principal tenant une (ou des)
figure(s) dans son sein, c’est à dire entre ses bras, dans un linge ou dans son
vêtement, conformément au champ de significations ouvert par le terme sinus.56
Autour du sein d’Abraham déjà évoqué, on voit se dessiner une constellation de
thèmes, dont les principaux sont: le type trinitaire de la Paternité divine, la
Vierge à l’enfant, certaines figurations rares de l’Ecclesia, des représentations
généalogiques, ainsi que, dans une certaine mesure, la Vierge au Manteau.
L’analyse peut porter, d’abord, sur les relations des thèmes pris deux à deux
(qu’on ne peut toutes expliciter ici). Ceci est particulièrement difficile si l’on
veut tenir compte de tous les facteurs, notamment chronologique et quantitatif, et
éviter de réduire chaque thème à un type simple, alors qu’il constitue au
contraire une (ou plusieurs) gamme(s) de variantes. Partons de la Vierge à
l’enfant, le thème quantitativement le plus développé de notre paradigme—et
plus largement de tout l’art médiéval (avec la Crucifixion). La similitude
formelle est forte (fig. 6), tant avec la Paternité divine qu’avec le Sein
d’Abraham (du moins avec l’un de ses trois types formels), au point qu’on peut
attribuer à la Vierge à l’enfant un statut de référent, ayant probablement joué un
rôle dans le développement des deux thèmes mentionnés, autour de l’an mil.
55 Autre exemple: les Repas du Christ (huit thèmes, outre la Cène), étudiés par D.
Rigaux, dans une perspective sérielle, parfois vigoureusement quantitative, et soulignant
l’intérêt d'une approche comparée de ces thèmes; D. Rigaux, A la table du Seigneur.
L'Eucharistie chez les Primitifs italiens. 1250-1497 (Paris, 1989).
56 L'hyperthème est construit sur la base d’un motif commun, associé à des figures
différentes. C’est bien toutefois le réseau des thèmes, pris dans leur globalité, qu'il s'agit
d’étudier.
Mais, loin d'être la reprise passive d’un modèle, le transfert d’un schéma formel,
qui ne peut se produire que s’il est pertinent dans son nouveau contexte, est
hautement signifiant. La figuration d’un même mode de relation à l’égard du
Christ, mais glissant d’un emploi concernant sa Mère à un autre, mettant en
scène son Père céleste, constitue l’indice d’un relation forte à l’intérieur de
l’hyperthème étudié. Au reste, d’autres éléments confirment l’homologie, de plus
en plus marquée au cours du Moyen Age, entre la relation Père/Christ et la
relation Vierge/Christ. Cette équivalence constitue une relation structurale forte,
sensible par exemple dans les dispositifs visuels qui mettent en parallèle Trône-
de-Grâce et Vierge à l’enfant (fig. 7).57
Entre certaines représentations du sein d’Abraham et de la Paternité divine,58
les ressemblances vont parfois jusqu’au décalque formel, bien que, prises
globalement, les deux séries soient loin de se recouvrir strictement (fig. 8-9).59
Le rapprochement est aussi thématique: il s’agit dans les deux cas d’exprimer un
rapport de paternité, d’une nature assez proche (spirituelle ou divine), ce qui est
d’autant plus net que ces deux images paternelles, Dieu et Abraham, sont parfois
fondues en une figure ambivalente. La relation entre ces deux thèmes permet de
saisir un jeu complexe de rapprochements et de différenciations. L’écart entre les
deux thèmes est mis en évidence par l’analyse globale des deux séries, ainsi que
par l’examen des cas où ils se trouvent associés au sein d’une même oeuvre,
comme au tympan de Soria (fig. 8). L’enjeu de cette différenciation est
fondamental: il s’agit de souligner l’écart entre la filiation intratrinitaire du Père
au Fils (une filiation égalitaire, non-hiérarchique et sans rapport de génération,
qui inverse radicalement les données de la filiation chamelle) et la filiation
surnaturelle des chrétiens (à l’égard d’Abraham et de Dieu). La conjonction des
similitudes et des écarts souligne que ces deux relations sont en partie
homologiques (d’où le lien de germanité entre le Christ et les chrétiens), mais en
Heures de Jeanne de Navarre, Paris, BnF, ms. n. acq. fr. 3145, f. 3V (miniature
anglaise, ajoutée au début XVe siècle). Même dispositif dans l'Evangéliaire d’Italie du
Nord, déjà cité, Bibliothèque vaticane. Vat. lat. 39, f. 171v—172. C’est surtout à partir du
XIIIe siècle que les théologiens soulignent l’équivalence du sacrifice de la Mère avec celui
du Père (Bonaventure, dans ce contexte, indique ‘Mater per omnia conformis est Patri', In
Lib. Sententiarum, 1, 48, 2; cité, avec d’autres textes, par T. Dobrzeniecki, ‘Medieval
Sources of the Pietà’, Bulletin du Musée National de Varsovie, 8 (1967), 5-24).
58 Sur ce thème, voir en dernier lieu, F. Boespfiug et Y. Zaluska, ‘Le dogme trinitaire’
(cf. note 21), pp. 197-202.
59 Tympan de l’église Santo Domingo, à Soria, Castille, fin XIIe siècle; Décrétales,
Bibliothèque Vaticane, Vat. lat. 1389, f. 4 (Bologne, première moitié xrvc siècle).
Certaines formes d'inclusion sont utilisées pour l'un des thèmes mais exclues pour l’autre.
Ainsi, l’usage du linge, dominant dans le cas du sein d’Abraham, n’apparaît jamais dans
la Paternité divine, signe probable d’une volonté de différencier les deux thèmes.
Fig. 8. Soria. Santo Domingo, portail, au tympan: 'le Fils clans le Sein du Père ’.
au centre de la seconde voussure: les élus dans le Sein des Patriarches (vers
1200). ©Author
Fig. 9. Décrétales. Bibliothèque vaticane, Vat. lat. 1389, fo i 4. ‘le Fils dans le
Sein dit Père ' (première moitié du XI Vè siècle). © Biblioteca apostolica vaticana
partie seulement (les théologiens concevant, dans ce contexte, les hommes
comme fils adoptifs de Dieu, tandis que le Christ seul est Fils véritable). On a ici
un cas de relation complexe associant, de façon nécessaire, homologie et écart.
La relation entre Sein d'Abraham et Vierge au Manteau permet d'illustrer un
autre type de rapport (fig. IO).60 Cette fois, le rapprochement formel est limité,
puisque les fidèles sont, dans ce dernier thème, rassembles, non in sinu mais sous
le manteau de la Vierge (remarquons toutefois que la différence d’échelle entre
le personnage principal et ceux qu’il rassemble contre lui est de même nature).
Pourtant, le lien thématique est puissant: sein d'Abraham et Vierge au manteau
expriment tous deux, quoique dans des contextes généralement différents,
l’appartenance des fidèles à la communauté ecclésiale. Dans les deux cas, ce lien
60 Fribourg, cathédrale, contrefort Sud, XIVe siècle. Sur ce thème, cf. P. Perdrizet, La
Vierge de miséricorde (Paris, 1908); V. Sussman, ‘Maria mit dem Schutzmantel’,
Marburger Jahrbuch fü r Kunstwissenschaft, 5 (1929), 285-351.
est figuré par la réunion à une figure parentale unique (Abraham pater omnium
credentium; Maria mater omnium), par la fusion dans un groupe et par
Finclusion dans un corps-vêtement. Abraham dans son rôle de réceptacle céleste
des justes et Marie dans sa fonction de protectrice des vivants et des morts,
symbolisant l’un et l’autre la communauté ecclésiale, se trouvent donc en
position équivalente. Au reste, cette relation se trouve confirmée par des
convergences exceptionnelles. C’est le cas lorsque la Vierge au Manteau apparaît
dans le Jugement dernier, ce qui nous rapproche du contexte paradisiaque
caractéristique du sein d’Abraham (fig. 11);61 inversement, il peut arriver que ce
dernier prenne un forme si singulière que l’on soit tenté de le qualifier
d” Abraham au Manteau’.62 Là encore, les cas les plus singuliers de la gamme
sérielle favorisent les connexions interthématiques.
Mais un décalage chronologique existe entre les deux thèmes: la Vierge au
Manteau apparaît dans la seconde moitié du xmè siècle, pour se développer
seulement au siècle suivant, lorsque le temps fort de figuration du sein
d’Abraham s’achève et fait place à un progressif déclin. Sans conclure à un
remplacement d’un thème par l’autre (car leurs fonctions ne sont pas strictement
assimilables), on peut cependant avancer l’idée d’un relai objectif (et progressif)
de l’un par l’autre. On a donc à faire cette fois à une équivalence forte, mais
visuellement peu explicitée, entre deux thèmes se succédant chronologiquement.
Par ailleurs, l’étude hyperthématique doit prendre en compte les relations
entre les éléments considérés et ceux qui fonctionnent à leur égard dans un
rapport d’opposition. Dans le cas du sein d’Abraham, deux éléments anti
thétiques se présentent. En tant que figuration paradisiaque, le sein d'Abraham
ne peut être compris sans considérer en regard les représentations infernales,
d’autant que les échos formels soit avec la gueule d’enfer, soit avec la marmite
où cuisent les damnés sont souvent très explicites (fig. 2). L’expression d’une
incorporation, d’une inclusion dans une forme suggérant une intériorité
rassurante, caractéristique du sein d’Abraham, prend plus de force encore
lorsqu’elle est confrontée à des images de dévoration qui se fixent sur le seuil,
sur l’ouverture corporelle, dans son aspect le plus angoissant. D’autre part, on
peut établir un lien structural entre sein d’Abraham et sacrifice d’Abraham.63
Cette association révèle les deux faces de la figure du patriarche: d’un côté, le
61 Heures de Marie de Bohun, Copenhague, B.R., Thott. 547, 4°, f. 32' (Angleterre, c.
1380-95).
62 Psautier, Munich, Staatsbibl., Cim 2641 (plat de reliure, Ratisbonne, c. 1250): dans
ce Jugement dernier, le manteau d’Abraham, figuré en buste, s’élargit vers le bas et
recouvre les élus (on peut s’en faire une idée en isolant mentalement la partie supérieure
de la miniature de la figure 3, réalisée dans le même milieu).
63 Ce lien est établi par plusieurs oeuvres, qui placent ces deux éléments en vis-à-vis,
par rapport à l’axe médian (par exemple tympan roman de Saint-Michcl-dc-Lescure. près
d'Albi, ou autel d’or de Lisbjerg, conservé au Musée national de Copenhague).
Fig. 10. Fribourg. Cathédrale, contrefort, ‘la Vierge au manteau '. xive siècle.
©Author
i 04 JÉRÔME B a SCIIET
père protecteur qui rassemble avec bienveillance sa progéniture contre lui; de

l'autre, le père terrible, manifestant la cruauté de la Loi divine et prêt à trancher
la gorge de son fils. Outre cette évocation clivée de la figure paternelle, on
remarquera la logique de cette mise en relation. C'est en effet par l'épreuve du
sacrifice qu’Abraham devient définitivement père d’une multitude (Gn 22),
c’est-à-dire, dans la relecture chrétienne, père de tous les élus qu’il est précisé
ment destiné à réunir au paradis, dans son sein. L'analyse d'un hyperthème ne
saurait donc négliger ces ‘antithèmes’, ou plus largement les relations
d’opposition qui donnent leur plein relief aux motifs étudiés.64
Enfin, au-delà des relations entre les thèmes pris deux à deux, l’analyse de
l'hypcrthème peut tenter d'articuler l’ensemble de ces liens. Sans s’y engager ici,
on indiquera que cette démarche a permis de mieux cerner la position à la fois
stratégique et ambiguë du sein d'Abraham. Admettons, dans les monothéismes,
une organisation selon deux axes: l'un paternel, centré sur le sacrifice (Père-Fils,
Abraham-lsaac), l’autre maternel (l’Eglise), définissant la communauté.65 Il est
remarquable que l’étude hyperthématique fasse apparaître le sein d’Abraham, à
travers ses liens avec la Paternité divine et avec la Vierge au Manteau,
précisément à l’intersection de ces deux axes. En effet. Abraham est en position
d’équivalence structurale tant avec la Vierge-Eglise qu’avec Dieu le Père. On
peut ainsi mieux rendre compte de l'articulation de la fonction maternelle et de
la fonction paternelle dans l'image du sein d’Abraham, placé à la jonction de la
conception d'un Dieu-Père et d’une communauté-Mère.
En dépit du caractère encore expérimental de la démarche évoquée, on peut
dégager de cet exemple quelques indications générales. D’abord, on se gardera
de préjuger du dégré de systématicité de l'hyperthèmc étudié. Celui-ci doit au
contraire demeurer toujours en question, du fait de la complexité de chaque
thème, mise en évidence par l’analyse sérielle, et de la diversité des relations
susceptibles d’unir les thèmes entre eux. Ainsi, on a pu faire apparaître des types
de rapports de nature différente: transfert de modèles formels, possibilités de
comparaisons partielles mais aidant à saisir la spécificité de chaque thème,
cohérences structurales, soit par opposition ou inversion, soit par équivalence (en
distinguant coexistence et substitution dans la chronologie), soit par un jeu
complexe de rapprochements et d’écarts. Encore ces relations ne sont-elles le
plus souvent que partielles, s’appliquant rarement à la totalité des séries
thématiques considérées. Les relations mises en évidence sont fluctuantes,
indécises et hétérogènes quant à leur nature et à leur intensité (tout comme les
64 D’autres formes de relations antithétiques peuvent être analysées: dans le cas

Nativité du Christ/Dormition de la Vierge (qui se réfère également au paradigme de
sinus), il s’agit plutôt d’un effet de chiasme (cf. note 54).
65 Voir les remarques de G. Rosolato, ‘La communauté constituée en tant que fonction
maternelle’, dans son recueil d'articles Pour une psychanalyse exploratrice clans la
culture (Paris: Presses universitaires de France, 1993). pp. 287-301.
Propositions pour nue iconographie sérielle 105
Fig. 11. Heures de Marie de Bohun. Copenhague, Bibliothèque royale, Thon.

547, 4°, fol. 32', ‘la Vierge au manteau protégeant les élus lors du Jugement
dernier’ (c. ¡380-95). © Kongelige Bibliotek. Copenhagen
thèmes eux-mêmes, notamment par leur poids quantitatif). Ainsi, s’il peut être
possible de faire apparaître, dans un hyperthème, certaines relations structurales,
on ne saurait dire, du moins a priori, que l’hyperthème lui-même constitue une
structure. Il s’agit plutôt d’un réseau de relations complexes, chacune devant être
pensée dans sa singularité. Cependant, en dégageant des connexions, des
glissements, des homologies et des écarts, l'hyperthème apparaît comme l’un des
cadres pertinents de l’approche iconographique. Il permet de sortir du fractionne
ment dominant de la recherche actuelle, sans pour autant postuler une
systématicité incompatible avec le fonctionnement des gammes sérielles.
C onclusion
Les études iconographiques, au sens défini ici, ont de beaux jours devant elles.
L’approche des hyperthèmes peut ouvrir des perspectives nouvelles; celle des
motifs est presque sans limite, à la mesure de l'élargissement du questionnement
des disciplines historiques. Même dans le domaine des thèmes iconographiques,
pourtant plus traditionnel, beaucoup reste à faire, et on est surpris du nombre
considérable de thèmes qui n’ont pas fait l’objet d’une étude satisfaisante. Ainsi,
le renouvellement des problématiques et des méthodes découvre, pour
l’approche iconographique, un champ immense à explorer. C’est à cette aventure
que doivent contribuer les nouveaux moyens documentaires que, tous, nous
appelons de nos voeux et à la constitution desquels nous nous efforçons de
contribuer.
On insistera, pour finir, sur la nécessité d'articuler l’inventivité et la sérialité
des images médiévales. La reconnaissance de l'inventivité et de la liberté des
images (dans les limites qu’il convient d’assigner à ce terme) donne un sens à
l’approche historique. Celle-ci n’a de portée qu’à condition de reconnaître que
les images sont un lieu spécifique de mise en jeu des rapports sociaux et
d’expérimentation des représentations du monde. De plus, par l’exploration

systématique des thèmes, mise en évidence par l'approche sérielle, elles révèlent
certains aspects d’énoncés fondamentaux, qui, ailleurs, apparaissent sous
d’autres angles et sous d'autres lumières. L’approche sérielle est indispensable
pour faire apparaître l’épaisseur historique et la complexité des phénomènes, dès
lors qu’on la conçoit, non comme le repérage de types fixes dans un corpus
homogène, mais comme la production de gammes de variations, au sein
d’environnements complexes et mouvants. L’approche sérielle permet alors
d’articuler le singulier et le général, l’inventivité et la traditionnalité. C’est dans
cette perspective, croit-on. que l’iconographie peut contribuer à une approche
historique—mais problématiquement historique—des images médiévales. Et
c’est à cela aussi que peuvent contribuer les bases de données d’images, conçues
non seulement comme instruments documentaires, mais aussi comme supports
matériels d’un renouveau des méthodes d’analyse.
Image and word: systematic research into
the relations between image and w ord
in D utch culture (1500-1800)
JÖRGEN VAN DEN BERG, HANS BRANDHORST,

and P eter van H uisstede
aking a repertory of Dutch printers’ and publishers’ devices has been a
M long-term effort in an era marked by the short-term financing of

academic projects. It started in 1983 and, working on it intermittently,
it took until 1999 to reach a stage where its results can be published in the form
of a catalogue + CD-ROM. The printers’ devices project is essentially a spin-off
or ‘by-product’ of the Short-Title Catalogue Netherlands (STCN); its collection
is thus destined to grow as more material is gathered while STCN progresses.
Over time the original purpose of the project has gradually changed, from
preparing a mere bibliographical tool to compiling a corpus of images that
presents a valuable source for the history of visual culture.
In order to enable users to study the cultural and historical context of printers’
devices, not only did we make very detailed descriptions of the images with the
iconclass system but we also studied these images and the mottoes against the
background of other sources, such as emblem books, Erasmus’s Adagiorum
chiliades, and illustrated title pages. Several of these sources and essays on the
context they share with devices will be published separately.
In this essay we discuss some of the features of the catalogue, published both
as a book and as a CD-ROM.' After a brief introduction to the phenomenon of
1P. van Huisstede and J. P. J. Brandhorst, Dutch Primers' Devices I5tli-I7th Century,
a Catalogue with CD-ROM, 3 vols. (Nieuwkoop: De Graaf Publishers, 1999).
108 van den B erg. B randhorst, van H uisstede
printers’ devices, we describe a few editorial and technical issues facing editors
who answer the challenge to publish this amount of data both electronically and
in print.
R ecurrent U sage a n d the A rra n g em en t o f D evices
For many years after 1657, when Joannes Maire ceased to print and publish
books in Leiden, several variants of his well-known device—a man digging with
a spade in a landscape below the motto Fac et spera—continued to be used by
other printers and publishers. One of these variants, number 0223 (fig. 1), was
used by Maire in 1642 and 1652, and again in 1654 in a book printed for Lowijs
Elzevier. However, it also adorned the title pages of at least eight editions
available from the shop of another Leiden printer and bookseller, Pieter van der
Aa, in the years between 1686 and 1690. It was furthermore found on eight
editions published by Frederik Haaring in the period 1688-98. Moreover,
between 1668 and 1697 the same variant was associated with at least six other
names: Johannes van Bilderbeeck, Anthony Schoutcn, Cornelis Driehuysen,
Abraham Gogat, Jacobus Moukee, and Leonardus Strick. Although it could
perhaps be established who actually owned this small wood-block after 1657, we
can safely predict that the history of its ownership will be a complicated one.
From the late sixteenth century onwards, a device showing a personification
of the Christian, in particular the Calvinist, religion, was used throughout Europe
by many Protestant printers and booksellers. In the northern Netherlands during
the seventeenth century alone, some twenty variants of this device were pro
duced, ultimately deriving from an illustrated poem and emblem by Theodore de
Bèze. On title pages it occurs in combination with the names of at least twenty
different printers and booksellers. The oldest Dutch variant, number 0742
(fig. 2), was used in Delft by Bruyn Harmensz. Schinckel in 1593. Without its
typeset motto, Religio Christiana, it appears in several Schinckel editions from
1601 to 1603. After Schinckel, the block was associated with Jacob Lenaertsz.
Meyn, Jan Jacobs Palensteyn, and, finally, in 1634, with Volchard Jansz.
Camerlingh, all of whom were active in Enkhuizen.
Another variant of the same device, number 0387 (fig. 3), was first used by
Gerrit Jansz. Arensteyn in 1634. More than half a century later we still come
across the same block: in 1697 Jacob Claus had it printed on the title page of his
edition of the works of Hans Engelbrecht van Brunswijk. No matter what
historiographical issues this varied pattem of usage may suggest, it first of
all alerts an editor to the problem of how to arrange his material. On the more
than 16,000 title pages in our photocopy archives we have identified the prints of
some 2,000 different blocks and plates. So every block was printed an average of
eight times. On average too, every block has been associated with at least two
different printers or booksellers, leaving us with well over 4,000 instances of a
Relations between image and ward in Dutch culture (1500-1ROO) 109
im o a u c tttr.
Fig. 1. Device number 0223. ©Institute o f Information and Computing Sciences,

Utrecht University
Fig. 2. Device number 0742. © Institute o f Information and Computing Sciences,

Utrecht University
Fig. 3. Device number 0387. © Institute o f Information and Computing Sciences,

Utrecht University
lio van den B erg, B randiiorst, van H uisstedk
block linked to a name or a combination of names. The association of devices

with different names naturally poses some problems, for a catalogue—like this
one—is arranged alphabetically according to the names of printers and
publishers. Most essentially it brings out the problems of redundancy (a device
may be used repeatedly by different printers, which leads to multiple entries for
the same device) and of attribution (when two names are associated with a
device, where should one put it in the catalogue?).
Even though a device may seem to begin its life as a personal invention,
appearances can be deceptive, due to lack of information or historical coincidence.
In Dordrecht, in 1622, the printer Dirck van Vreeswijck used for the first time the
device number 0284 (fig. 4), which conceals a pun on his last name. The picture
shows Joseph fleeing Potiphar’s wife. Its motto Quantum quisque timet, tantum
fugit contains Latin words for both halves of Vreeswijck’s name: timere can be
translated in Dutch as vrezen [to fear], while fugere [to flee] can be translated as
wijken. However, already in 1621, and also in Dordrecht, this device had been used
for van Haemstede’s martyrology published by Joris Waters, Zacharias Jochimsz.
and François Boels. To be sure, the appearance of the device in this publication
could very well indicate that the actual printer was Dirck van Vreeswijck, but our
prime source, the book at hand, is silent on this question.
How careful we must be when inferring the ownership of a device from
internal evidence like an obvious pun on someone’s name can be shown with
many examples. An instructive one is that of device 0453 (fig. 5). If the only
occurrence of this block was in a book published by Philippus Bonus in Leiden
in 1673, we might be tempted to conclude that this device was originally
designed as a pun on his latinized surname. Bonus, then, would be not a
translation of the Dutch name ‘de Goede’ but a rather more trivial latinisation of
the name ‘Boon’. The word boon is the Dutch equivalent of the Latin fa ha [bean]
. . . said to have flourished in the motto of 0453: Dum stabant pisae sic viguere
fabae. Although this could very well be the pun Bonus intended, his device was
actually an objet trouvé rather than an inventio, because the block originally was
the device of the chamber of rhetoric ‘De Jonge Batavieren’. Jan Philipsz. van
Steenwegen used it for them on a play by P. Nootmans in 1629. In the same year
Jan Pietersz. Waelpot printed another play by Nootmans with the same device,
and Abraham Waelpot used it in 1652 and 1659.
It will be clear, even from these few examples, that reconstructing the
ownership or usage sequence of a device can be a puzzling affair. Internal
indications may be misleading, while the consecutive, or even simultaneous,
association of different names with a device can hide a Gordian knot of historical
realities. To cut the knot we decided on the separate inclusion of every instance
of a device’s use. In practice this means that a separate section of the catalogue is
devoted to every owner—or rather user—of a block. We also state for every user
in which years he or she actually used it, and how often it was used per year. By
thus mirroring the recurrent use of a block, the catalogue provides the reader
Relations between image and word in Dutch culture (1500-1ROO) 111
Fig. 4. Device number 0284. © Institute o f Information and Computing Sciences.

Utrecht University
Fig. 5. Device number 0453. © Institute o f Information and Computing Sciences.

Utrecht University
with a better opportunity for critical assessment of our data than a more polished
state of our attributions would have done. This strategy has produced catalogue
entries that will occasionally raise the eyebrows of some bibliographers, because
we may deliberately ignore what seems so obvious, as with Vreeswijck's device.
This will be most evident in cases in which a bookseller’s name is joined to a
device that is actually the device of the printer, while the printer’s name is not
mentioned anywhere in the book; or cases in which, vice versa, the printer’s
name is combined with the bookseller’s device. In all instances of this kind the
device has been catalogued under the name found in the book. Still, even though
this approach is aimed at preventing premature attributions, we cannot altogether
avoid attributions. Nor did we want to avoid them. Whenever more than one
name is found in a book, be it on its title page or in a colophon, we had to decide
under which primary name the device would actually be presented. Some of
112 van den B erg. B randhorst, van H uisstedh
these decisions are completely arbitrary, but the usage of devices and internal
evidence about the names of printers, streets, and shops may provide information
on which to found an attribution. Whenever available, the STCN data on
addresses and shop signs are provided with the catalogue entries.
H o w D evices Were S e le cte d
Ex Hierarum studiis immortalitatem acquiri . . . ‘To reach immortality through

the study of letters’. While some wishful thinking and commercial interest are
mixed with this humanistic creed, printers and booksellers throughout Europe
sincerely believed in it and gave vent to their beliefs by means of their devices.
In the northern Netherlands we come across several variants of the device. The
motto may be abbreviated to Studiis immortalitatem acquirimus, paraphrased as
Literae immortalitatem pariunt or played with as in Vita sine litteris mors est.
The main connotations of the imagery accompanying these statements are
immortality and fame, visualised either by a phoenix, a personification of Fame,
a Triton trumpeting on a shell, or Minerva, alone or with Mercury. Such a
widespread sentiment cannot be traced to a single source. At best we may try to
establish which literary and/or visual expressions were comparatively successful
in disseminating it. It will hardly come as a surprise that, judging by the number
of devices inspired by its motto and its imagery, one of the popular expressions
of the theme was an emblem by Andreas Alciatus.
Alciatus’s book of emblems certainly was not the only book of moral
instructions, proverbial expressions and attractive imagery that printers and
booksellers explored in their search for an attractive device. Emblems and
devices invented by Claude Paradin, Paolo Giovio, Roemer Visscher, Hadrianus
Junius, Jacob Cats, to name but a few; adages collected by Erasmus; quotable
phrases from classical sources such as Vergil and Seneca; and, of course, biblical
citations—all could be exploited by book producers who wished to make a
statement about their religious convictions or to demonstrate the quality of their
education and their business.
Sometimes devices testify to a ready knowledge of these sources and a high
level of inventiveness in adapting a source to the particular situation of a printer
or bookseller. Devices of this type can be found in the catalogue under names
like Corverius, van Waesberge, van der Wouw, Wachtendonck, Cloppenburgh,
Ravesteyn, and Brandt. Sometimes, however, devices are simply ornamental.
They may also be a simple ‘Hausmarke’, or limit themselves to a decorative play
with their owner’s initials, or use the arms of an academic, cultural, or political
organisation. Devices of this kind can be found under Allardus Gauter, Gheraert
Leeu, Marten Gerbrantsz, or Pieter van der Slaart.
The simple fact that the arms of a city or a university, emblems from a
widespread emblem book, images from shop signs, or someone’s initials could
ReUilions between image and word in Dutch culture (1500-1ROO) 113
all be turned into a device shows that any effort at defining what a printers’
device is would be irrelevant, because it is the use that makes the device. This
explains why other visual elements—tail-pieces, title vignettes, marginal illustra
tions, or initials—were sometimes included in the catalogue as well. The
criterion of usage proved an acceptable guide in deciding on the inclusion of
devices in the present corpus. Perhaps users may find it was applied too liberally,
but even then the criterion remains unambiguous and clear.
Slightly less self-evident is the history of the corpus upon which this catalogue
is based. When data collecting started, some fifteen years ago. the ambition was
almost exclusively bibliographical, as the initiative came from the STCN. The
STCN is the Dutch bibliography covering the period from the first Dutch products
of the printing-press to the year 1800. In its database the user will eventually find
all books published to that date within the present borders of the Netherlands as
well as all books printed in Dutch abroad. Between 1982 and 1997 the Dutch
holdings 1540-1700 in the Koninklijke Bibliotheek (KB) and the university
libraries of Amsterdam and Leiden were described. In 1997 the collection 1540—
1800 of the university library in Utrecht was taken in hand, while work on the
eighteenth-century holdings of the KB has been in progress since 1995. Also,
Dutch editions from the period 1622-1700 in the British Library are being
processed both for the STCN and for a future volume in the British Library’s
catalogue series. At the moment the STCN amounts to circa 80,000 descriptions.
The estimate is that it will eventually contain some 300,000 titles.
The STCN is accessible from many Dutch libraries, but it can also be
consulted from home or office by users with Internet access to the KB or any
other institution providing a connection to PICA’s Online Retrieval System
(ORS) and Open Library Network (OBN). Recently, it has also become
accessible worldwide through the Hand Press Book Database (HPB) established
by the Consortium of European Research Libraries (CERL).
The STCN’s original assumption was that an informal photocopy archive of
the devices of printers and publishers with access to their iconography could be
useful as a tool for the identification of printers and booksellers. And it was.
However, the systematic collection of large quantities of pictorial material is
somewhat of a terra incognita even for most art historians. Obviously, it took
some time for the team of bibliographers to develop criteria for the selection of
the imagery found on title pages, since not every picture found there is a printer’s
or a publisher’s device. Once or twice, during that process, a device may have
escaped notice. Conversely, sometimes a picture has been archived, which—it
has been decided, on closer inspection—is not an actual device.
The presence of a printers’ or publisher's device is always mentioned in the
bibliographical descriptions of the STCN, and this information can be used for
retrieval when consulting the database through the aforementioned Online
Retrieval System or the Open Library Network. In early 1999 the STCN
database contained some 17,000 references to printers’ devices. Since this figure
includes references to material that has been left out of the present catalogue, for
example eighteenth-century editions, it closely matches our own count of circa
16,000 prints. So, even though a complete comparison of the two lists of
references is simply not feasible, we are confident that only rarely will a device
have completely slipped through the net. Despite incidental or, rather, accidental
divergences, the devices project is completely dependent on STCN for its
material, so all devices included in our catalogue and dated between 1540 and
1700 were found in the libraries whose collections were indexed by STCN: the
Koninklijke Bibliotheek, the university library of Amsterdam, and the university
library of Leiden. Devices from Utrecht’s university library and the British
Library—both of which are still incompletely catalogued—are not included here.
Nor are devices included of printers or publishers who started their business after
1700. Devices dated before 1540 were included on the basis of secondary
publications, particularly from those that provide reproductions of this material,
such as Nijhoffs L ’Art typographique1 and Juchhoffs Drucker- und
Verlegerzeichen des xv. Jahrhunderts? The information about the usage of the
devices from the period before 1540 is not based on personal inspection but is
derived from these older published sources, once or twice supplemented from the
STCN database.
The E lem ents o f an E ntry
The Utrecht printer Abraham van Herwijck associated himself with his Old
Testament namesake and chose as his device a picture of Abraham about to
sacrifice Isaac. Another Abraham, namely Abraham van Blancken, used a device
showing some moles burrowing into a hill. His choice, at first sight surprising, is at
least partly explained by his address, the Molsteegh—‘Mole Alley’—in Amster
dam. A third one, Abraham Biestkens, also of Amsterdam, referred to the ‘church
under the cross’ by way of his device, which shows a lily surrounded by thorny
shrubs. His choice of device is directly related to the name of his shop and the sign
with which it was adorned: ‘De Lelye onder de doom en'. All kinds of associations,
whether simple or very intricate, could stimulate the invention of a device.
The names of streets and shops can help us to understand the choice of device.
We are grateful for the Koninklijke Bibliotheek for permission to include the
2 W. Nijhoff, L'art typographique dans les Pays-bas pendant les années 1500 à 1540:
reproducidlo en facsimile des caractères typographiques, marques d’imprimeurs,
gravures sur bois et autres ornements employés pendant cette période, 3 vols. (The
Hague: Nijhoff, 1926-35).
R. Juchhoff, Drucker- und Verlegerzeichen des XK Jahrhunderts in den Nieder
landen. England, Spanien, Böhmen. Mähren und Polen (Munich: Verlag der münchner
Drucke, 1927).
Relations between image and word in Dutch culture (1500-1800) 115
information on location and shops, which has been collected while STCN built
its thesaurus of printers and publishers, their places and years of activity,
addresses, and shop signs. Together, these data are the core of each record
created for the devices used by individual printers. Sometimes a note, usually
about some form of co-operation or certain family relationships, has been added.
The STCN project is scheduled to continue for quite a number of years.
During the last four years the rate at which new material was added to the on-line
database has even accelerated. Our biographical information, though ultimately
deriving from this database, had to be ‘frozen’ at some point, and then adapted to
the format of our own files. This implies that it dates back to the start of the final
editing, the beginning of 1998. Divergencies between the data of this study and
the data the STCN is offering on-line can usually be explained by this time gap.4
Users are advised to consult the STCN on-line data if they want to make
absolutely sure that they have the fullest and most recent information.
The main elements (entries in the database) that have been detailed by the
study are:
Device number [Number]
The device numbers, although randomly assigned, are meaningful since they are
the unique identifiers of blocks and plates. They can be employed to retrieve all
instances of the use of a particular block. The user can do this either by
consulting the printed concordance—device number to printer—or by searching
the CD-ROM by device number.
Iconography 1: Free text description
A free text description has been created of every block. Although it may seem
redundant to classify the pictorial elements of iconographical interest with
Iconclass as well as to describe them in prose, these two components of an entry
are actually complementary. A reproduction needs to be accompanied by a
description of its subject. The systematic classification of subject matter is a
prerequisite for building a systematic index of the iconography of the devices.
Wherever necessary, the description explains the general relationship between
the distinct iconographie elements to which an Iconclass notation is assigned.
A feature of the textual descriptions, intended to serve those who do not read
Dutch, is the inclusion of the Dutch translations of those words that hint at the
4 The same holds, obviously, for any publication based on STCN-data of a more recent
date than ours. As a matter of fact STCN editors J. A. Gruys and Jan Bos are preparing a
printed publication of all the addresses and shop signs.
1 16 van den B erg. B randhorst, van H uisstede
name of a printer, or his address or shop sign. Often these puns are pretty
obvious, because the keywords are present in the motto and/or name, e.g.
Cloppenburgh's Sie/ ick sta aen die deure ende doppe illustrated with a man
knocking [kloppend] on the door of a castle [burchi). Sometimes it is helpful to
know the Dutch translation of an English word. When a printer named Bot uses a
picture of a flounder for his device it is useful to know that ‘bot’ is also a Dutch
word for flounder. On the CD-ROM the full text of the descriptions is made
searchable in the Keyword picklist, with its accompanying KWIC index.
Iconography 2: Iconclass notations [ICO]
This article is not a suitable place to explain at length what is Iconclass. A more
extensive explanation of the Iconclass system is to be found in the CD-ROM’s
helpfile on Iconclass. Information is also made available on-line, on the Iconclass
website (http://iconclass/let.uu.nl) at Utrecht University. Moreover, a substantial
number of catalogues using Iconclass, both on paper and on CD-ROM, has been
published during the last decade or so, exposing a wide audience to the system.
However, since vocabularies are generally used according to rules that are
specific to a project, a few remarks are in order about the way we have applied
Iconclass to the printers’ devices. This holds even more strongly as the research
of our Image & Word project has considerably affected our views on the
iconography of the devices. In this project several types of sources—selected
emblem books, printers’ devices, occasional engravings, title pages, Erasmian
texts—are digitized and connected through hyperlinks; their text is transcribed
and SGML-tagged'’ to facilitate the hyperlinking; and their iconography is
indexed with Iconclass. And because historiographical studies are written in the
same electronic format, we can hyperlink comments and source texts. Thanks to
this project, which has embedded our research on printers' devices for a decade
now, we have gained a better grip on the contents of a number of important
emblem books and texts such as Erasmus’s Adagiorum chiiiades.
Being better informed about what we like to call ‘the emblematic game’, we
have included much detail in our iconographie encoding of the devices and also
added many Iconclass notations on the basis of our interpretation of the mottoes.
A few examples should clarify this approach and thus give an idea about the kind
of iconographie data that have been recorded.
The first device mentioned in this introduction was number 0223, originally
used by Joannes Maire. In spite of its apparent simplicity, any description of its
subject matter will of course depend on the purpose and the frame of mind of the
indexer. However, there will not be much debate about the fact that it shows a
' Standard Generalized Markup Language.

man digging with a spade in a hilly landscape. Over his head we read Fac et
spera as well as the name of God in the shape of a Jahweh tetragram. Men
digging with spades occur in more than fifty different printers’ devices, while the
Latin roots ‘spe-’ and ‘labor-’ are to be found in more than seventy. Whatever
the implications of these numbers may be, they make it unlikely that a man
digging in a landscape with a text such as ‘work and hope’ above him, is just a
realistic depiction of the hard labour of a seventeenth-century peasant. Without
speculating about the wider context or deeper meaning, we feel justified in
assuming that anyone researching the imagery of Labour or Hope could be
interested in a picture such as this. Therefore our description includes these more
abstract concepts.
Thus Maire’s device is indexed with the following Iconclass notations:
46A170 Labour;
5(+12) Abstract Ideas and Concepts—an abstract concept represented
by a male figure; and
56Dl(+3) Hope; ‘Speranza’, ‘Speranza delle fatiche’ (Ripa)—sym
bolical representation of concept.
Of course it is also to be found with those notations that describe the man
digging with a spade, the landscape with hills, the olive-and palm-branch
framing the scene, or the tetragram from which a radiance emanates:
11C l3 tetragram (in Roman or Hebrew script)—symbol of God the
Father;
22C31 radiance emanating from persons or things;
25G3(OLIVE-TREE)(+22) olive-tree: branch, stick;
25G3(PALM-TREE)(+22) palm-tree: branch, stick;
25H11 mountains; and
471122 digging with a spade—soil cultivation.
On the CD-ROM it is explained—and demonstrated in action—that the
hierarchical organisation of Iconclass also ensures retrieval of this picture through
keywords such as ‘God’. ‘Christian religion’, ‘symbol’, ‘supernatural light’, ‘ray of
light’, ‘soil’, ‘farmer’, ‘cultivation’, ‘agriculture’, ‘spade’, and many more.
A small detail of the picture that could easily go unnoticed is the bird on a
branch facing the man (fig. 6). While this could be seen as a naturalistic addition—
c.g. a small bird waiting for the seeds that will be sown shortly—it can also be read
as the visual counterpart of the motto’s exhortation to hope. Exploration of the
emblem books that we have indexed in much the same way as these devices,
informs us that the crow can be the attribute of Hope—corvix, avis tributa spei. As
such it is included in the various picturae of Alciatus’s emblem In simulacrum Spei
[the imagery of Hope], and commented on by humanists such as Claude Mignault.
It is the bird’s call—eras, eras—that triggers the association, because in Latin this
call means ‘tomorrow, tomorrow’. . . .
Fig. 7. Device number 1312. © Institute o f Information and Computing sciences.

Utrecht University
Needless to say, it remains a question whether the inventor of the device or its
contemporary observers would have connected it to this little piece of ornitho
logical lore, dating back to antiquity. These reservations, however, do not
diminish the potential usefulness of our encoding of the picture with the afore
mentioned notation 56Dl(+3) and with the concept:
25F32(CROW)(+l) song-birds (with NAME)—animals used
symbolically.
Our interpretation of the bird’s presence, which occurs in 25 out of 28 versions
of the device found to date, may be wrong. Nonetheless, given the modest
position of our description in the historiographical argument about the picture’s
meaning—it is nothing but a first step—no harm is done.
In much the same manner, the picture of a turtle with wings (fig. 7) accom
panied by the motto Festina lente [make haste slowly] is encoded with the
notations for the opposing speeds of motion—slowness and haste—which are at
the heart of the device’s meaning:
51Ml(+3) Fast Motion—symbolical representation of concept; and

51MMl(+3) Slow Motion—symbolical representation of concept.
Moreover, we include references to emblem books, the Adages of Erasmus,
and to other literature as well, wherever we think this would constitute a helpful
hint to the user. Thus the following concepts are added to the description of the
winged turtle:
83(ERASMUS, Adagiorum chiliades II i 1) specific works of
literature (with AUTHOR, Title); and
86(FESTINA LENTE) proverbs, sayings, etc. (with TEXT).
Two things should be clearly understood about this enrichment of the iconogra
phie description. References to literature such as the Adagiorum chiliades or an
emblem book do not imply that we think we have identified the source of a device.
In some cases devices are indeed straightforward copies of emblems. More often,
however, the ideas expressed by an emblem, and the visual and literary means used
in the process, are adapted to the situation of a new user and to the message he or
she wants to communicate. The cultural substratum on which the emblematic game
is played is far too complex and the information about the genesis of specific
devices too sparse to tolerate a reductivist approach. As a matter of fact, of the
possible relations between devices and emblems, the straightforward 'model-copy'
relation is not the most instructive.
In much the same vein, a word of warning is in order about the symbolic
interpretations, of which the identification of a crow as the attribute of Hope is an
example. The potential for conveying meaning, even of ‘simple’ iconographie
elements such as crows and turtles, is by no means exhausted once we have
identified one of their aspects. The corpus of printers’ devices itself provides the
means of countering such simplistic assumptions, as it shows that turtles could also
be used to express independence: carrying their house on their back, they are self-
sufficient animals that are overal thuys [anywhere at home]. Furthermore, their
Dutch name ‘schildpad’ could stimulate the imagination in a direct way, as is
illustrated by the device of Amelis Jansz. van Paddenburgh, which shows a turtle
in the foreground with a castle [burchi] in the distance. As lack of speed also
characterises snails, it is not surprising that they too could advertise a printers’
claim that his work did not suffer from error-producing haste.
The gist of what has been said here about the status of our references to
abstract ideas and emblem literature through Iconclass codes also applies to the
part of the codes identifying biblical sources. There too, mottoes play an
important role when we have to decide the usefulness and accuracy of a
reference to a particular text. One example should suffice to illustrate this: a
device that exists in more than a dozen variants shows an open book encircled by
a quotation from Joshua 1:8: Laet hel Boeck deser Wet van uwen monde niet
comen [this book of the law must ever be on your lips]. Whether or not the open
120 van den B iïrg. B randhorst, van H uisstede
book of the device should be identified with ‘this book of the law’ is not the
main issue here. What is important is that applying the most appropriate
leonclass concept ‘71E51 Joshua communicating with God (in general)’ could
suggest that the biblical hero himself is depicted rather than the content of his
communication with God. The reason we decided to assign possibly confusing
notations to a device like this is that, properly warned in this introduction, users
may still benefit from these references.
It may seem an odd thing to say, after having assigned more than 20.000
leonclass notations to the corpus, but given our limited systematic access to the
iconographie and textual details of sources such as emblem books, there still is
much room for improvement and expansion. Our descriptions, it cannot be
emphasised enough, may suffice to awaken the historical opponent, but not to
put him to rest.
Mottoes [MOT]
Mottoes of printers’ devices are not only interesting in their own right but also
important for other reasons. They help us understand the iconography of a device.
But a motto may also clarify why a particular theme was chosen by a particular
person. When Cleopas and Peter realise they have been talking to Jesus on the way
to Emmaus, they say, according to Luke: nonne cor nostrum ardens erat in nobis
dum loqueretur in via [did we not feel our hearts on fire as he talked with us on the
road]. The Dutch translation of the phrase ardens erat contains the root ‘brand’,
which gave Marten Jansz. Brandt a reason to choose this theme for his device. This
would seem to be confirmed by the fact that Jacob de Meester, who used the same
subject a few years before Brandt, quoted Matthew 18:20—waerder twee o f drie in
miin naem vergadert ziin—instead of Luke.
Mottoes may, furthermore, help us to distinguish variants of a device, for
example because of variant spellings. Whenever it was considered appropriate or
useful, the motto of a device would be exploited to improve or expand the
iconographie description, as we explained above. Literal transcription of the
mottoes, preserving orthographical variations and upper and lower case, are
included to help identify blocks. Normalised versions of the mottoes were
included too on the CD-ROM. to ensure easy retrieval of basically identical text.
Initials and monograms [INI]
Under this heading we have gathered the initials of printers, booksellers,

draughtsmen, and woodcutters. Whenever it seemed possible to assign initials or
a monogram to a bookseller or printer, we have done so, taking into account that
our chances of correctly identifying the letters hidden in an ornate monogram
depend on the number of times it is used by the same printer. In such monograms
you often see what you know.
Systematic research to identify the artists’ signatures was beyond the scope of
this catalogue. Suggestions of that nature are mostly traditional and intended to
allow the easy retrieval of pictures with identical initials or monograms.6*
Notwithstanding the moderate status of devices, they will supplement what is
known about the oeuvre of artists such as the van Sichems and the Serwouters.
Years and frequency of use [RAN]
An important addition to the types of information brought together in the pilot

study is the survey of a device’s use by a particular printer or bookseller. For
every user of a device we have recorded in which years he or she employed it
and how often it was used per year. These are the years that are included in the
picklists Stari date and Latest date of the CD-ROM’s central search facility.
Images
The illustrations of devices have been digitized with a flatbed scanner. All
scans—some of photographs, some of photocopies—were made at the same,
moderate, resolution of 200 dpi. In the printed catalogue the images are
presented at a fixed size. The full-size images can be found on the CD-ROM,
where one can check them against a fixed scale.
The C atalogue as a B o o k a n d as a C D -R O M
In the past century, important repertories have been compiled of the printers'
devices of France, England, Germany, Switzerland, Spain, Italy, and, most
recently, Belgium. No such repertory existed for the Netherlands, in spite of its
being one of Europe’s major centres of book production. To be sure, a survey of
devices from the Low Countries had been included in the fourth volume of the
Bibliotheca Belgica, and many devices were reproduced in Briels’s study about
the migration of printers and publishers from the southern to the northern
6 These will be found on the CD-ROM via the item Expert Search of the Search menu.
In Expert Search you should consult the picklist IN2 Initials (explained).
Bibliotheca Belgica. Bibliographie générale des Pays-Bas. Fondée par Ferdinand
van der Haeghen. rééditée sous la dir. de Marie-Thérèse Lenger. 7 vols. (Brussels:
Editions Culture et Civilisation, 1964-75). See especially vol. tv. pp. 60-208: ‘Marques
Typographiques'.
122 van den B f.rc;. B randiiorst. van H uisstede
Netherlands in the late sixteenth and early seventeenth century.8 The devices of
the Elzeviers were listed and reproduced by Rahir,9 and various essays were
devoted to the devices of specific printers or publishers. Occasionally, devices
were discussed in thematic studies such as that of Simoni10 about verbal and
visual puns in Dutch devices or that of Nielson11 about Latin mottoes on Dutch
devices. However, no attempt at a comprehensive survey of Dutch devices had
been made. Given the intimidating fact that by now ‘only’ 80,000 of an
estimated total of 300,000 editions have been catalogued by STCN, a complete
catalogue of Dutch devices cannot be expected for many years to come. Still,
enough material has been collected now to warrant a publication.
While its foreign predecessors differ among themselves in scope, method, and
ambition, our survey, a latecomer in the field, is characterised by a feature that
sets it apart from all of them. Publishing the repertory both as a book and in an
electronic form is symptomatic of the central role computerisation has played in
most stages of the research and of the editorial process. Not only is our selection
of material dependent on the computerised catalogue of the STCN project but
also the wish to hyperlink the sources we are bringing together in our Image &
Word project requires them to be transformed into electronic documents, and to
be enriched with some form of markup. We do think this process would deserve
a detailed treatment, but here we must limit ourselves to those aspects which
have had an immediate effect on the form and content of the present publication.
When we began to prepare our data for publication we decided to make the
results of our research available as a printed book and in an electronic format.
For technical and editorial reasons, CD-ROM was chosen as the medium. The
CD-ROM included with the book provides an environment tailored for fast and
flexible querying and offers quite a number of options to compare the results of
queries. However, we are well aware that the use of a CD-ROM is connected to
the life cycle of the operating system it is made for, and we can be sure that a
s J. G. C. A. Briels, Zuidnederlandse boekdrukkers en boekverkopers in de Republiek

der Verenigde Nederlanden omstreeks 1570-1630 (Nieuwkoop: B. de Graaf, 1974).
9 E. Rahir, Catalogue d ’une collection unique de volumes imprimés par les Elzevier et
divers typographes hollandais du I T siècle (Paris: Damascène Morgand, 1896). See
especially pp. 419-66: ‘Fleurons et marques typographiques des Elzevier de Leyde et des
Elzevier d’Amsterdam—Principaux fleurons et marques typographiques de diverses
officines de Hollande du xvif siècle dont les productions se joignent à la collection
Elzevirienne’.
10 A. E. C. Simoni, 'Bearwood, Tree, Flatfish & Co: some punning Dutch devices’, in
Festschrift Hellinga, Forty-three Studies in Bibliography presented to Prof. Dr. Wytze
Hellinga on the occasion o f his retirement from the chair of Neophilology in the University
o f Amsterdam at the end o f the year 1978 (Amsterdam: Nico Israel, 1980), pp. 447-66.
11 A. C. Nielson, Latijnse Zinspreuken op Nederlandse Boekmerken (Amsterdam:
Ellerman Harms, 1952).
Relations between image anti word in Dutch culture ( i500-1800) 123
printed book will outlive that. Moreover, computers, let alone CD-ROMs, still
are not used by every researcher potentially interested in the subject. In other
words, there are several arguments to add a printed catalogue of Dutch material
to other recent ones, such as the survey of Italian devices by Giusseppina
Zappella12 and that of southern Netherlandish devices by Frank Vandeweghe and
Bart Op de Beeck.13
A multi-purpose datastore
The concept of exploiting the same data for multiple purposes may sound rather
trivial. Is that not precisely what computers allow you to do? Still, when a file
should behave correctly in an SGML / XML environment but should also be
easily converted to HTML; when it should print as though it was typeset with a
desktop publishing program but do so without forcing the editor to arrange the
layout of some 1,600 pages with 3,500 illustrations page by page; when it should
be converted into a database with indices but remain open for corrections until
the evening before the final conversion-—then the concept becomes slightly more
ambitious and deserves some comment.
DEVICE
C onceptual D atam odel
The complex relationship of printers or booksellers with the devices they used to sign
and uniquely identify their book production can be described logically as a ‘many to
many’ relationship between the entity ‘device’ and the entity ‘printer’. This
relationship is actually even more complex since printers or booksellers worked
together and used in different ‘joint ventures’ the same or new devices. Sometimes
devices were even merged into one new presentation of symbols and concepts of the
original devices. So the entity ‘printer’ has a recursive relationship with itself. This
observation is crucial for designing the logical structure of the datastore that has to
hold all the data as well as for the interaction design of the software.
12 G. Zappella, Le marche dei tipografi e degli editori italiani del Cinquecento.

Repertorio di figure, simboli e sogetti e dei relativi motti, voi. 1: Testo, voi. 2: Tavole,
Editrice Bibliografica (Milan: Editrice Bibliografica, 1986).
13 F. Vandeweghe and B. Op de Beeck, Drukkersmerken uit de I5de en lòde eeuw
binnen de grenzen van hei huidige België (Nieuwkoop: B. de Graaf, 1993).
124 van den B erg, B randhorst, van H uisstede
SGML for structure and semantics
The CD-ROM, the printed catalogue as well as the hypertext system all derive
from a single file datastore that reflects the above-mentioned structure. It
consists of records that still preserve the data format but are now expressed in a
so-called ‘tagged format’ file, using SGML markup for its overall structure and
semantics. A (simplified) sample looks like this:
<PRINTER>
<THES>
<VNM> Frcdcrik Haaring</VNM>
<PL1> Leidcn</PL1>
<JA1> 1688—1712</J Al >
<PL2> Amsterdam</PL2>
<NJA2> 1710</NJA2>
</THES>
<STCN>
<OCC> bookseller, 1688-1700</OCC>
<ADR> Klokstceg (in de), 1692; Raapenburg, op de Hock van dc
Nonnesteeg nevens de Academie (‘t), 1694-1699</ADR>
</STCN>
<DM>
<NUM> 0017</NUM>
<TXT> A man digging with a spade, in a landscape; Jahweh-tetragram; a
laurel-branch and a palm-branch as frame</TXT>
<NOT> 11C 13; 22C31; 25F32(CROW)(+l); 25G3(OLIVE-TREE)(+22);
. . . </NOT>
<SP1> FAC ET SPERA</SP1>
<SP2> Fac et spera</SP2>
<RAN> 1690(1): 1693(1); 1695(1); 1696( 1) </RAN>
</DM>
<DM>
<NUM> 0182</NUM>
<TXT> Rocky coast with beacon-light, pillars of Hercules, the sun as a
face, ships, banderole with inscription</TXT>
<NOT> 24A6; 25H131; 25H23241; 46C24(+63);.. .</NOT>
<SPI> 1NTELL1GENT1BUS</SP1>
<SP2> lntelligcntibus</SP2>
<RAN> 1693(1) </RAN>
</DM>
</PRINTER>
(Indentation for readability only)
The SGML elements <THES>, <STCN>, and <DM> function as containers
for the information in the form of ‘tagged data’. They contain basic ‘chunks’ of
information about the printers and devices. These ‘chunks’ are hierarchically
subordinated to the <PRINTER> element. The contents of the <THES>,

<STCN>, and <DM> elements are as follows:
< THES>: preferred and variant name of the printer, places and dates of
activity; based on the thesaurus compiled by Grays and De Wolf;
<STCN>: professions, addresses, and shopsigns, if any; data derived
from the holdings of the STCN database; and
<DM>: information found on title page and in colophon, to which a
description of the device is added.
A small, dedicated document type definition (DTD) was written to be able to
check the file with an SGML parser. The strength of the SGML application lies
in the ease with which the separate ‘chunks’ of structured textual data can be
manipulated. This is especially convenient if new information, e.g. information
originating from the STCN database, necessitates moving the description of a
device to a different printer. Since <DM>-elements always contain the data we
saw on (the photocopy ot) title pages and colophons, attributions become
immediately identifiable as such: a title page may carry two or more names, but
the information is stored under only one of them.
For the book-production, the file was imported into an Inmagic database to
generate all indices and check its content. The original file, again, is filtered
through a pipe of small programs with the help of tools such as Sed, Awk, and
Python. At the end of the pipeline a full-blown SGML file is generated. This file
is transformed into a regular \LaTeX\ file using software based upon the model
of the linuxdoc project. \LaTeX\ is used to typeset the book because the core of
the catalogue consists of about 3,500 images, each with a corresponding
description. Typesetting this amount of material page by page would have been a
very cumbersome process, and time spent on it would have seriously reduced the
time available for checking and refining the data.
The issue here is precisely the one Leslie Lamport described in \LaTeX\ A
Document Preparation System: the time spent on typesetting material should be
infinitesimal compared to the time spent on collecting it. This resulted in a three-
volume, 1600 page catalogue organised into sections as follows:
Heading
• Name of the printer or bookseller
• Location: city (cities) and date(s) of activity
• Profession(s): professional specialization [printer, bookseller, etcetera]
126 van den B erg, B randhorst, van H uisstede
• Address(es), shop sign(s), and notes

Device entries:
• The number of the device
• A textual description of the device
• ICO: the ICONCLASS notations used for the systematic classification
of iconography;
• INI: initials that are visible on the device;
« MOT : the transcription of the motto of the device;
• RAN: the range of the use of the device by the printer or bookseller.
Format: a year
For the design of the CD-ROM the conceptual datamodel was used to logically
bind all the data of the different entities. For that purpose the basic SGML
datastore was converted to a multi-table physical datastructure that is optimized for
querying. This means that the data were finally organised in a way that allowed for
redundancy if it helped to speed up the process of selecting and browsing the data
or if it improved the ease of use of the CD-ROM application. The tabular data were
imported in a proprietary database manager that handles all the physical data-
manipulation. In the final application, the user is presented with indices on all the
attributes of the three conceptual entities: printer, device, and use.
Device attributes
NUM Unique ID number of devices (identical to printed edition)
TXT Free text description of device
NOT Iconclass notations
SPI Literal transcription of motto (case preserved)
SP2 Normalized transcription of motto [biblical location added]
SPL Language(s) of the motto
IN 1 Initials of printers, booksellers, artists
IN2 Initials interpreted (for artists: traditional interpretation)
OPM Notes
Printer & bookseller attributes
KEY Unique ID number of printers & booksellers
ASDN Preferred name, arranged by last name
VAR Variant orthography, pseudonyms, etc.
VNM Preferred name starting with first name
COMB Names (more than one on title page and/or in colophon)
DRU Preferred name, arranged by last name, no brackets
OCC Occupations and specific qualifications
UIT Shop signs
ADR Addresses, usually street names
PLC Name(s) of city/cities where shops were located
YEAR Period(s) during which shops were active
OPM Notes
Use attributes
KEY Unique ID number of printers & booksellers
NUM Unique ID number of devices (identical to printed edition)
RAN Years in which a device was used
The software design of the CD-ROM application (proprietary to our Institute for
Information Science) is conforming to the successful and well-proven concepts
that have been implemented for the DISKUS series (in June 1999 more than 16
different titles). The interaction design allows a gradual refinement of user
interactivity to access the overwhelming amount of data. It offers different styles
for novices and experts to formulate queries.
Query formulation
The interface of the Printers’ devices CD-ROM presents the user first with a
screen where a query can be formulated by choosing from different ‘TABs’ and
picklists. The different ‘TABs’ represent the entities Printers & Booksellers and
Devices and give access to the main attributes in associated picklists. In those
cases where single words were derived from phrases, a KW1C index (KeyWords
In Context) is available for even more detailed selection. Selection criteria can be
combined with Boolean logic for complex query construction.
In expert mode the user is given additional access to all original fields (more
than 20 different attributes). The user can browse through the systematic
classification scheme of the full ICONCLASS hierarchy (potentially 24,000
different concepts) and can search with ICONCLASS keywords (potentially
more than 18.000).
I2S van den B erg. B randiiorst. van Huisstede
Browsing the selection
The results are presented in different layouts that coincide with different ways of
scanning large amounts of data visually. One is a gallery with thumbnail images
(imagescan only), and the second is a short list of selected devices, containing a
thumbnail image and a basic description.
Single-entry browsing
A third way of looking at the results is by browsing single entries from the
selection presented in hyper-document style: thumbnail image + related
printer(s) + intermediate description and with hyperlinks to full data of printer(s),
device and image.
Web of full printer & device data
From the different presentations of the query results, one can easily jump within
the underlying complex relational web of printers and devices. This web has a
hyper-document style of presentation and interaction:
full text with hyperlinks, multiple fonts, hotspots, inline buttons and
icons;
multiple and resizable windows (the physical screensize of the monitor
is the only limitation); and
all data in detail presented.
What started as a query to limit the amount of results ends in a non-limitative
web of related documents about devices, printers, booksellers, woodcutters, et
cetera. The user certainly does not have restricted access to this domain.
B ack to the F uture
To ensure they can be read in the future and on different types of computers, the
data are kept in an ASCII file with SGML markup describing its structure and
content. But longevity and portability are not the only arguments for choosing an
SGML file as our primary method of storing information. A very important
additional reason is that as an SGML file it can also function as part of the larger
corpus of documents that is gradually built up as our Image & Word project
progresses.
This corpus consists of both primary sources and historiographical essays that
are to be made available electronically and that are interconnected through a
hypertext linking structure. Among the sources that so far have been transcribed
Relations between image and word in Dutch culture (1500-1 SOD) 129
in full, marked up, and indexed iconographically are editions of Andreas

Alciatus, Emblematum libellus, Hadrianus Junius, Emblemata, Paolo Giovio,
Dialogo dell’imprese militari et amorose, Gabriele Simeoni, Imprese heroiche et
morali, Claude Paradin, Devises heroiques, Johannes Sambucus, Emblemata,
Geoffrey Whitney. A Choice o f Emblemes and other devises, Joost van den
Vondel, Toonneel des menschelikken levens o f de vernieuwde Gulden Winkel,
and Jan Moerman, Microcosmos. In several cases more than one edition of a text
has been transcribed.
In addition to these emblem books a transcription was made of the 1730
edition of Friedrich Roth-Scholz, Thesaurus Symbolorum ac Emblematum, a
survey of devices of European printers and booksellers, with an extensive
commentary and also a short essay on printers’ devices by Claude Mignault.
In a sense the title of Roth-Scholz’s survey suggests a central topic we are
dealing with in our own essays: the printers’ device as a case of ‘applied
emblemàtics’. Central notions in the emblematic game are concepts that are also
pivotal in a key text such as Erasmus’s De duplici copia verborum ac rerum:
richness, association, adaptation, imitation, paraphrase, interpretation, emulation,
loci communes, etc. The rich and complex relations between texts, concepts, and
images, caused by their continuous ordering and re-ordering, irresistibly call to
mind the image of a web. Seen against the background of the Internet hype, this
metaphor may appear to be a platitude. Yet, the hierarchical organisation and
web-like linking structure of both Iconclass and SGML make them eminently
suitable for tracing the meandering structure of early modern thought and artistic
invention. Thus we consider electronic documents combining both instruments
to be the natural habitat for the essays we are presently preparing. The subjects
we shall deal with include: the relationship between emblems and devices, Dutch
devices in a wider European perspective, Rubens’s painting The Ship o f State in
the light of emblems and devices, the topography of bookselling and printing
using addresses and shop signs, the meaning of trees in printers’ devices, and the
imagery of Labour.
The idea that book production is a laborious process has been illustrated in
devices since the early stages of the printing business. One cannot help but
wonder why.
Six St Jeromes: notes on the technology and
uses o f com puter lighting simulations
S imon N iedenthal
would like to begin with a quotation from St Bonaventure: ‘Light is the
I substantial form of bodies; by their greater or lesser participation in light,

[bodies] acquire the truth and dignity of their being’.1 By ‘bodies’,
Bonaventure refers not to our physical being but, rather, to matter itself; his
argument takes place in the context of the great medieval metaphysics of light. In
a similar vein, Robert Grosseteste begins De Luce, his thirteenth-century treatise
on light, with the assertion that:
Light of its very nature diffuses itself in every direction in such a way that a point
of light will produce instantaneously a sphere of light, unless some opaque object
stands in the way. [...] Corporeity [that is, the capacity of matter to take form and
occupy space], therefore, is either light itself or the agent which perfonns the
aforementioned operation and introduces dimensions into matter in virtue of its
participation in light.2
Light is portrayed by both theologians as the mechanism by which the cosmic x,
y and z axes are defined, the first of a number of intriguing echoes of medieval
thought on light and optics that one encounters in digital image synthesis.
In a less metaphysical sense, an understanding of the power of light to define
things underlies foundational lighting curricula everywhere (including my own).
A quick survey of some notable photographs demonstrates the way in which
1 Umberto Eco, The Aesthetics of Thomas Aquinas (Cambridge, MA: Harvard

University Press, 1988), p. 110.
2 Ibid.
132 Simon N iedenthal
light and value are central to the perception of pictorial form, surface qualities,
space, and volume. Qualities of light, moreover, have the capacity to wield
emotional influence through association with primal visual needs. And as with
all visual experience, what we see and the way we think about light is informed
by cultural traditions. As sensory phenomenon and interpretative touchstone,
light yields to both scientific and cultural analysis, and serves as a useful starting
point for a consideration of new applications of digital technologies to the
subjects of art and architectural history.
In his seminal work The Reconfigured Eye, William J. Mitchell traces the
development of digital image synthesis from simplest to most complex
techniques, with reference to corresponding changes that occurred in painting
during the Italian Renaissance and afterwards. His opening argument establishes
a dialogue between photography and painting, two media with differing manners
of engaging value and representing light effects. ‘Synthetic-shading procedures’
Mitchell writes, ‘are used to develop perspective views into closer approxima
tions to—even simulations of—photographs. (Thus they are closely analogous to
the procedures Renaissance artists employed to convert line cartoons into tonal
and coloured paintings)’.34The process of creating a digital image based on a 3D
scene begins with the fashioning of a wire-frame model to express the physical
boundaries of surfaces and objects. In the early days of digital image generation,
the boundaries were denoted as a list of vertices in 3D Cartesian space, though
now we have become accustomed to manipulating wire-frame models through
interfaces as graphical representations. Once the objects and spaces of a scene
have been modelled, surface characteristics are associated with each formal
element, light qualities and positions are specified, and the scene is rendered.
The rendering program takes into account the geometry, surface qualities, and
lights of a scene and renders a two dimensional representation ofthat scene from
a given viewpoint.
Mitchell goes on to explore digital image rendering from simple flat shading
to the most complex radiosity calculations, referring along the way to artists as
diverse as Reynolds, Vermeer and Paul Gauguin. The easiest way to render a
surface in the computer is to simply specify a single colour for each point on the
surface, but over time rendering algorithms have been refined to allow the
calculation of shading and light effects of progressively greater complexity and
fidelity to visual experience. One conceptual breakthrough was the development
of ray-tracing. The rationale for ray-tracing is that while there are billions of
light photons bouncing around in any visual environment, the ones that really
matter are those that enter the eye.-' In the ray-tracing process a virtual picture
3 William J. Mitchell, The Reconfigured Eye (Cambridge, MA: MIT Press. 1992),
p. 137.
4 This insight is taken from the Lightscape Visualization System Getting Started,
pp. 1-7.
Six St Jeromes 133
plane composed of a grid of pixels is defined, and a ray is traced from an eye
point through each pixel into the scene until it reaches a surface. The light
environment informs the surface characteristics of each point, and the resulting
hue is used to colour the pixel. This process continues until all pixels in the grid
have been coloured. Ray-tracing is, as Mitchell points out, analogous to the way
in which a perspective is constructed, and he quotes Leonardo:
‘Obtain a piece of glass as large as a half sheet of royal folio paper and fasten this
securely in front of your eyes, that is, between your eye and the thing you want to
portray. Next, position yourself with your eye at a distance of two-thirds of a braccio
from the glass and fix your head with a device so that you cannot move at all. Then
close or cover one eye, and with the brush or a piece of finely ground red chalk mark
on the glass what you can see beyond if. Thus each ray [Mitchell continues] in the
viewing pyramid projects a colour from a point in the scene to a point on the picture
plane.. . .
Besides the similarities to perspective construction, the process of projecting
rays from the eye to a surface beyond is reminiscent of Euclid’s theory of vision,
which was considered viable at least until the optical work of Al-Hazan. though
popular long into the medieval period.5 6
Although ray-traced renderings allow the simulation of more sophisticated
visual phenomena, such as accurate reflections and refraction, the products of
this sort of rendering still lack a sense of the real presence of light. Specifically,
the interreflection and colour bleed from one surface to another are not
calculated. In a room like the one in which I write this sentence, we could say
that there are areas of direct illumination, where lights are shining directly on
surfaces, and other areas of indirect illumination that are lit by the light bouncing
from surfaces. The indirect illumination is not calculated in a ray-traced image.
Practically speaking, simulating accurate light presence in a ray-traced image
calls for a procedure that combines a photographic process—involving setting
lights within the software—with one in which areas of light and dark value are
applied directly to surfaces in a manner more analogous to painting. In the words
of the old cinematographer’s saw, ‘if you can’t paint with light, light with paint'.
So, following Mitchell’s lead, I have rendered a simple scene using tech
niques of progressively greater complexity (fig. 1, plate 6). This is a simulation
of a room with a skylight and several cubes in it, rendered under default lighting.
Default lighting in Alias is the computer graphics equivalent of Grosseteste’s
formless universe. Anyone who has looked at light for a while can play around
for a few minutes and produce something that demonstrates a more convincing
impression of light in the environment (fig. 2, plate 7). The Phong shading
5 Mitchell (as in note 3), p. 139.

6 David Park, The Fire Within the Eye: A Historical Essay on the Nature and Meaning
o f Light (Princeton: Princeton University Press, 1997), p. 78.
algorithm appears complex, yet it renders a fraction of the effects of light in a

real space. In fact, it is so limited a model that I had to paint many of the subtle
shadows and gradations in this figure directly on the surfaces, rather than rely
upon the algorithm’s inadequate simulation of surface-to-surface interreflection.
The advent of radiosity rendering—which models light behaviour much more
accurately than other rendering modes—signals a major advance in the
capabilities of computer simulation (fig. 3, plate 8). Radiosity rendering derives
originally from methods developed by thermal engineers in the 1960s to simulate
heat transfer between surfaces, as a means of designing jet engines, among other
things. In the mid-1980s, computer scientists at Cornell University began to
apply these techniques to the modelling of light with the computer. Radiosity
calculations are based upon the principle of conservation of energy. When light
is introduced into an environment, it illuminates some surfaces directly, and,
depending on surface characteristics, some of the light is absorbed and the rest
reflected back into the environment. Most current Tenderers do not calculate the
contribution of surface-to-surface interreflection to the illumination in a scene.
Radiosity software models interreflection by creating a mesh which stores the
illumination values associated with each surface. After distributing energy from
the light sources, the algorithm checks each surface and determines which has
the most energy to reflect back into the environment. The progressive distribu
tion of light energy continues until all of the energy in the environment has been
absorbed, or until the process is stopped.
Architectural historians eager to simulate unbuilt or destroyed buildings have
seized upon rendering technologies of greater sophistication, and examples
abound of reconstructions of everything from Frank Lloyd Wright buildings to
the castle of Mad King Ludwig of Bavaria.7 Notable works that have pioneered
the use of radiosity rendering technology in an architectural context include Kent
Larson’s 1993 simulation of Louis Kahn’s unbuilt Hurva synagogue.8 Larson’s
most recent work continues the exploration of light and surface in Kahn’s unbuilt
projects for Scripps at La Jolla.9 Also beautifully realised is MIT colleague
Takahiko Nagakura’s visualisation of Terragni’s unbuilt Danteum.10 These
projects are not the rule, however, for although computer-generated images have
become commonplace in recent years, just as commonplace is the absence of
rich and compelling illumination qualities and surface definition.
7 B. J. Novitsky, ‘Reconstructing Lost Architecture’, Computer Graphics World

(December 1998), 25. See also the April 1999 issue of CGW.
s Vincent Scully, ‘A Virtual Landmark’, Progressive Architecture (September 1993),
80-87.
9 Kent Larson, Louis I. Kahn: Unbuilt Masterworks (New York: Monacelli Press.
2000).
10 These images are taken from the 1998 Siggraph reel.
Six Si Jeromes 135
Mitchell’s contention that the development of techniques of digital imaging is

analogous to painting procedures is both encouraging and problematic: en
couraging because it suggests that processes we are currently engaged in can be
informed by art history; problematic because it fails to address significant
differences between digital and analogue. Making precise distinctions is central
to thinking critically about media, yet we continually witness the bleeding of art
terminology into the realm of the digital. Early on I was struck by the way in
which filmic and photographic concepts were embedded in software. Consider,
for example, the camera icon from Maya software (fig. 4, plate 9). But the
analogies in software go far beyond the photographic. In computer graphics we
refer to renderings, we employ software called Painter and Piranesi, and, while
waiting for an action to be performed, my laptop executes a little version of the
Mona Lisa instead of showing the usual clock face. Another amusing example of
the intersection of painting vocabulary and interface design is the creation icon
in 3D Studio Max, a hand with a spark at the fingertip that is a cross between Adam’s
creation in the Sistine Chapel and ET. But anyone who has performed a task as
elementary as ‘drawing’ a curve in a 3D package understands just how limited
are the analogies between digital and other media. I always think of defining a
c.v. curve as being more like an abstract type of wood bending, of applying
forces to a ductile material and editing the points until the desired profile has
been achieved. Any notion of the gestural component of drawing pretty much
goes out the window in 3D software.
The construction of painting space in a digital medium is one means of
exploring Mitchell’s painting/rendering analogy, and, in fact, the last few years
have seen a number of examples. Though the simulation of a painting in 3D
space is often seen as striking a blow against the ascendancy of photorealism in
computer rendering, some projects seek a wider scope, including ‘Rouen
Revisited’ by Paul Debevec at the University of California. The Rouen project
uses Monet’s 30 or so canvases of the cathedral as well as historical and
contemporary photos, to allow a viewer to reconstruct the front façade from any
number of possible viewpoints at any time of day or season. Debevec describes
his project as follows:
Fascinated by the play of light and atmosphere over the Gothic church, Monet
systematically painted the cathedral at different times of day, from slightly
different angles, and in varied weather conditions. Each painting, quickly
executed, offers a glimpse into a narrow slice of time and mood. We arc interested
in widening these slices...." 1
11 From Paul Dcbevec’s Home Page, alberti.cs.berkeley.edu/~debevec/. Another

project involving the 3D reconstruction of a painting space is the stomach-churning fly-
through of Van Gogh’s Yellow Room on the artmuseum.net web site:
www.artmuscum.net/vangogh/3dloginpage.asp
While this project undoubtedly advances the technology of image generation, it

raises questions about the uses of digital construction. Architecture at least aspires
to three dimensions in its built form. There are many reasons that the Kahn and
Terragni projects, for example, remained unbuilt—often financial ones—but there
is still something to be learned about the modulation of light that can be usefully
approximated by computer simulation. Construction of painted architectural space
is a bit more problematic. After all, a painting is quite content to remain in two
dimensions.
I would argue that 3D simulation of painting space is useful when it goes
beyond merely expanding the subject to another dimension and instead adds an
additional facet to our understanding of the work at hand. Applying photo
realistic simulations of light behaviour to a painting can serve, for example, as a
means of throwing the painter's value choices into relief. Thus the tasks of the
‘Six St Jeromes’ project: first, to reconstruct some of the intellectual context of a
painting, and, second, to rethink the dangers, uses, and legitimate pleasures of
this sort of simulation. The goal of this project as it was initially conceived was
to put Mitchell’s thesis to the test, to collide a fifteenth-century painting with the
most recent rendering technology, and to see what emerges.
Antonello da Messina’s ‘St Jerome in his Study’ (National Gallery, London)
dates from somewhere between 1450 and 1475 (fig. 5, plate 10); there seems to
be some uncertainty as to whether the painting represents Antonello’s early
work, or was produced during his later stay in Venice.1213The French novelist
Georges Perec, who kept a print of the painting on his wall, nicely grasped the
snug bookishness of the study. ‘The whole space' Perec writes,
is organized around the piece of furniture (and the whole of the piece of furniture
is organized around the book). The glacial architecture of the church (the bareness
of the tiling, the hostility of the piers) has been cancelled out. Its perspective and
its vertical lines have ceased to delimit the site simply of an ineffable faith: they
are there solely to lend scale to the piece of furniture, to enable it to be inscribed.
Surrounded by the uninhabitable, the study defines a domesticated space inhabited
with serenity by cats, books and men.1'
The St Jerome painting offers itself for several reasons. First, because da
Messina was among the first Italian artists to work with oil paints and glazes,
and the new materials allowed him a greater range of tonal and light effects. He
was also active in a period in which the changes documented by Mitchell were
becoming fully felt. Moreover, the painting responds well to conceptions of light
in the art theory of da Messina’s contemporaries like da Vinci.
12 Cf. N. Little, ‘A Note on the Date of London “St Jerome in his Study” by Antonello
da Messina', Arte Veneta (1976), 154-55.
13 GeorgcsPerec, Species of Spaces and Other Essays (New York: Penguin. 1997),
p. 86.
Six St Jeromes 137
The project began with the construction of a wireframe model of St Jerome’s

study, corresponding to the perspective of the painting (figs. 6 and 7, plates 11
and 12). A preliminary shaded version of the scene with default lighting and
basic texturing (fig. 8, plate 13) was brought into Lightscape software for
radiosity processing. Once the model was set up in an environment that roughly
corresponds to the illumination qualities in the painting, it became possible to
begin a quick series of ‘what if?’ renderings: what would it look like if sunlight
was coming from the other side? (fig. 9, plate 14). What if the light coming in
the doorway was very cool and flat? (fig. 10, plate 15). What if soft light was
coming in exclusively from the clerestory windows above? (fig. 11, plate 16).
What if Jerome decided to bum the midnight oil? (fig. 12, plate 17).
At this point it struck me that 1 had achieved only a sort of inverse
appreciation of the light of the painting. That is, none of my variations seemed as
satisfying as the light in the original. So in my fifth and sixth versions of St
Jerome I attempted to reconstruct the lighting environment suggested by the
painting (fig. 13, plate 18), and used whatever techniques necessary to replicate
the values on the surfaces (fig. 14, plate 19). In doing so, I learned that the
painting manifests a close relationship to theories on painting and light of near
contemporaries like da Vinci. The painting exhibits, of course, a wonderful
chiaroscuro, as defined by Leonardo:
The first intention of the painter is to make a flat surface display a body as if
separated from this [picture] plane, and he who most surpasses others in this skill
deserves most praise. This accomplishment, with which the science of painting is
crowned, arises from light and shade, or as we may say chiaroscuro.1415
Further, the treatment of the saint’s face echoes other writings of Leonardo on
the rendering of facial form:
The utmost grace in the shadows and the lights is added to the faces of those who
sit in the darkened doorways of their dwellings. Then the eye of the beholder
observes the shaded part of the face thrown into deeper shade by the shadows
from the aforesaid dwellings, and sees brightness added to the illuminated part of
the face by the radiance of the atmosphere. Because of such increases in the
shadows and light the face acquires great relief. .. .L’
Besides his 'St Jerome’, Antonello’s portraits also frequently demonstrate this
strategy. In his treatise on light in Renaissance theories of art, Moshe Baraseli
distinguishes two schools of thought on the concept of illumination. In the first, a
‘functional’ conception, the major task of light is to illuminate bodies and render
them visible, while the function of shadows is to produce relief and the
semblance of corporeality. The second approach emphasises the relation of light
14 Mitchell (as in note 3), p. 139.

15 Ibid. p. 143.
to contents and themes.16 ‘St Jerome in his Study’ responds to both of these
conceptions. Close scrutiny reveals careful delineation of the form of the desk,
apparent in the touches on the panel that do not reference the suggested light
environment—gradations, for example, where one surface meets another. And as
we have seen, the treatment of Jerome’s face echoes da Vinci’s advice on how to
make a body emerge from the picture plane. But the presence of light in the
painting also references iconographie themes. In her study of the painting, Penny
Howell Jolly notes the presence of Marian symbolism and the similarities of the
Jerome painting to an Annunciation. ‘Like Mary at the Annunciation’, Jolly
writes, ‘Jerome is bathed in a celestial light and visited by a divine presence’.17
Ultimately this project suggests possible uses for radiosity rendering as a tool
of art scholarship. One could use computer simulation and radiosity processing
to check a 2D representation against a 3D structure upon which it was based, to
help visualise the behaviour of light and simulate the space of unbuilt or lost
projects, to aid in the process of conservation or historic preservation, or to study
the illumination environment of works of art. This is especially useful in cases
where the conditions of illumination in a space have altered over time.
As a field of study in itself, however, computer-generated light has barely
been explored. One good critical touchstone to apply to the generation of new
digital projects is the question: could this project have been done if the computer
did not exist?18 Could it even have been thought of? One thing that sets computer
simulations apart from light in real space is the light algorithm itself, and gaining
an understanding of computer-generated light calls us to contemplate the
differences between the algorithm and real light behaviour. For example, one
unique quality of light produced by algorithm is that the algorithm can be
reversed: if the intensity of a source in most 3D packages is set to a negative
number, the source extracts light from the scene. ‘The light that shineth in
darkness’ becomes ‘the light that shineth darkness’, inverting the process with
which I opened this paper: ‘Light of its very nature diffuses itself in every
direction in such a way that a point of light will produce instantaneously a sphere
of light’. . . . The capacity to simulate the inverse of light behaviour—to create
spheres of darkness—is clearly one path to explore with algorithmic light, and
offers the possibility of new modes of spatial visualisation that philosophers like
Grosseteste could never have conceived.
16 Moshe Barasch, Light and Colour in the Italian Renaissance Theory of Art (New
York: New York University Press. 1978), pp. xi-xiii.
17 Penny Howell Jolly, "Antonello da Messina’s St Jerome in his Study', Art Bulletin,
65 (June 1983), 252. Jolly’s argument is based upon a reading of Jerome’s Epistle 22 to
Eustochium, in which Jerome counsels imitation of Mary as the basis for conduct.
18 I am indebted to an editorial by John Lansdown in Leonardo Magazine! 1998) for this
test.
The Lincoln C D -R O M Project:
history, theory, conservation, and images
P hillip L indley
he last two decades have seen paradigm shifts within Art History. Two
T are of particular relevance here. First is the macro-theoretical recon

figuration of the academic discipline of Art History, for the subject’s
intellectual boundaries have been challenged and redefined by a generation
impatient with inherited assumptions about the nature of Art and the tasks of the
art-historian.1Whilst art historians have sought to include wider fields of visual
culture within their study, so have other academic disciplines also begun to
deploy visual imagery as evidential material with increased sophistication; now,
art historians often need to understand and adopt their methodologies.
Inevitably, given the relatively small numbers of art historians and the in
creasingly wide spheres of their activity, this has led to some changes in the core
methodologies that define art history. Medievalist art-historians today sometimes
seem to have more in common theoretically with historians and archaeologists
working in the same period than they do with other art-historians working in
later chronological areas. The second transformation is technologically driven:
the digitization of texts and imagery has permitted new types of questions to be
posed (and answered), as texts and images can be readily accessed, analysed,
and manipulated. The project with which I am here concerned stands at the
intersection of these two paradigm shifts.
John Larson, the consultant conservator to the Dean and Chapter of Lincoln
E. Fernie, Art History and its Methods: A Critical Anthology (London: Phaidon,
1995). provides a helpful introduction to the vast literature on the subject. See also The
New Art History, ed. by A. L. Rees and F. Borzello (London: Camden Press. 1986).
140 Phillip L indley
Fig. L The west front. Lincoln Cathedral. © National Buildings Record. London
during work on the celebrated Romanesque sculptural frieze of the cathedral's
west façade (fig.l), first discussed with me the project of producing a
multimedia archive of the conservation policies and work on CD-ROM in 1991-
92. Two years later we were actually able to begin (January 1994). A CD-ROM
is a finished product, analogous to the book, and is imbricated within the
historical moment in which it was produced. This seemed important to us. given
The Lincoln CD-ROM projeci 14 1
that the aim was to produce a conservation archive that would inform future
debate about the conservation policies and practices which should be adopted at
Lincoln and elsewhere. The choice of CD-ROM rather than WWW can also be a
profound drawback because five-year-old designs look astonishingly outdated
and seem slow and crude, but the medium permits a firm navigational line to be
plotted, at the expense of the iterative de-privileged possibilities of the web.
Moreover, six years ago it was not clear to us how the web was going to develop.
The chief objective of the Lincoln project was, then, to document the
evolution of conservation policy and the progress of conservation work on the
Romanesque frieze sculptures. The conservation programme has been highly
contentious, with at times vitriolic debate about the methodologies to be
employed for cleaning the sculpture and—when it became clear that some
sculptures were in too fragile a condition to return to the façade—about the
nature of their replacements and where and how the removed panels should be
displayed. These issues were complicated by a weighty debate about whether
architectural sculpture should ever be removed from its context and about how
the meaning or meanings of the sculpture were and are affected by their
placement. It has been well said that ‘the history of conservation, which is rarely
written, is an integral part of conservation studies’, and the Lincoln CD is
intended to play an important role in future conservation debate.2
The sculpted frieze was, at the start of the current conservation campaign, in a
perilous state, a point first raised in 1976 by the doyen of Romanesque sculpture
studies in England. Professor George Zarnecki. Some major losses had already
occurred this century, and there is no doubt at all that without conservation there
would have been further, and potentially catastrophic, structural failures. A
detailed survey of the frieze took place in 1984. and the consulting architects
reported on several options in 1986. A great deal of work took place in the next
few years, and test cleaning started in 1987, with a major symposium convened
in 1988 to debate the future conservation strategies.' Work on cleaning and
conserving four panels began in 1989.
The CD documents the evolution of policy and the progress of conservation at
Lincoln, in minute detail, up until 1994. It charts the investigations into the
potential of modern technology—for instance the widely used ‘airbrasive’
machines—to clean the reliefs, to conserve them, and to replicate them. There
C. V. Horie. Materials for Conservation: Organic Consolidants, Adhesives and

Coatings (Oxford: Butterworth-Heincmann. 1987), p. vi. For an introduction to the
project as originally planned, see P. Lindley and D. Emmony. 'The Romanesque Frieze
of Lincoln Cathedral: Information Technology, Conservation and Archiving'. Local
Historian. 24 ( 1994). 83-90.
See The Romanesque Frieze and its Spectator, ed. by D. Kahn (London: Harvey
Miller. 1992).
Fig. 2. ‘The Death o f Lazarus and Dives and his companions in Hell' (panel 8,
according to Professor Zarnecki’s numbering system) after 1987 cleaning.
© The Dean & Chapter of Lincoln
was also considerable interest in the use of lasers to scan, clean and replicate
sculpture at Lincoln.4The CD should therefore be of immediate relevance to the
conservation, and specialist art-historical, communities. By making all this docu
mentation available, the Dean and Chapter have demonstrated a unique and
remarkable readiness to provide public access—for the first time for any English
Cathedral (indeed, probably for any historic building)—to the entire process of
conservation and policy formation.
A number of decisions have been taken at Lincoln, which will undoubtedly be
controversial once the scaffolding has been removed. The replacement of
original panels (fig.2) with modern carvings which are not ‘archaeologically
exact’ will certainly be contentious. Modern ‘copy’ carvings of the panels, based
on the removed originals, include reconstructions of lost details, to make the
meaning of the sculptures clearer; but the ‘recreation’ of lost sections of reliefs,
whether or not there is scholarly input into the appearance of the finished work,
is problematic when much of the rest of the frieze comprises damaged originals
1 For a convenient introduction to the subject, see now Laser Cleaning in

Conservation: An Introduction, ed. by M. Cooper (Oxford: Butterworth-Heinemann.
1998). For the debate about conservation methods in general, see Sculpture Conserva
tion: Preservation or Interference? ed. by P. Lindley (Aldershot: Scolar Press, 1997).
The Lincoln CD-ROM project 143
Fig. 3. ‘The Death o f Lazarus' block has been copied, and missing detail (e.g.
the soul of Lazarus) supplied by the sculptor John Roberts. The lower block of
‘Dives. . . in Hell’ is also a copy carving. © Author
(fig. 3).5 However, the current policy differs from the late eighteenth-century
one, which resulted in a replacement panel for one of the series of ‘Torments’: it
is more ‘scholarly’, and the damaged originals are being retained (fig. 4).
A research assistant, Mrs Ellen Pawley, began work in January 1994, her job
being to work systematically through all the works archives at Lincoln Cathedral
relating to the conservation of the Romanesque sculpture of the west front (there
was, at that time, no full-time archivist for the works department), to digitize all
the materia! and to arrange it in chronological order for publication on CD-
ROM. The assembling, transcription, and digitizing of the conservation archive
' This point is noted in the report by John Larson, dated January 1992 and accessed
from the CD's Conservation Archive, p. 55.
144 Phillip L indlby
and every aspect of the Information Technology design were her responsibility
in 1994 and 1995-96/’ There was to be no interpretation of the evidence, which
was to be assembled purely as an archive: in other words the principle of
separating records from analysis was to be respected. However, in 1997 the
conservation team at Lincoln imposed certain new requirements, the chief one
being that although the conservation archive is still arranged as a chronological
sequence (the work of Ellen Pawley) as originally planned, an introductory
summary of the documentary evidence has been provided at the ‘top level’ of
the archive. The basic point that interpretation should be distinguished from the
archival evidence has not thereby been entirely abrogated, for it will throughout
be possible to check my summaries against the contents of the documents
themselves. Paradoxically, perhaps, the process of bringing the project to
completion has imbued in me a certain scepticism about the relationship of
written archives to what actually took place. It has also become clear how many
decisions are made informally and are never officially recorded at all.
In order to make the conservation archive more accessible to non-specialist
users, there are four brief introductory ’tours'. It should be stressed, though, that
there is a profound difference between the conservation archive on the one hand
and the ‘tours’ on the other. The conservation archive is as close as we can make
it to an objective summary of all pertinent documents provided by the Dean and
Chapter's conservation team, together with the complete documents themselves.
The interpretative tours, on the other hand, necessarily contain my personal
views and will undoubtedly need to be modified in the light of the future
research. The tours comprise an introduction; a guide to the city, castle, and
minster (providing an introduction to the history of Lincoln from Roman times,
an account of the Castle and of the foundation of the diocese and building of the
cathedral); a description of the architectural location of the frieze in the west
façade of the minster; a descriptive analysis of the iconography of the frieze;
and, finally, a discussion of the date and style of the sculptures, examining
archaeological, structural, and stylistic matters. These tours cumulatively provide
an introduction to the conservation archive itself. The archive contains a 67-page
‘top-level’ guide to the contents and the substantial documentary material of all
kinds made available to us. It includes detailed reports by a variety of specialists
commissioned by the Dean and Chapter over the years. The process of writing an
introduction to these documents made it possible to compare the different
assumptions of these specialists, and it becomes evident that only a holistic
approach, combining the methodologies and employing all the studies of the
various specialists, will satisfactorily resolve some of the outstanding questions
about the frieze. In other words, the possible uses of the multimedia archive go
beyond its immediate objectives, to enable new questions to be asked and.
Although the design has been upgraded by Ms Kate Boardman in 1998-99. the
design is essentially that fixed in 1994-95.
Fig. 4. Eighteenth-century 'Torments' panel after recent conservation.

© Central Photographic Unit, University o f Leicester
sometimes, answered. One current debate, on the date of the frieze, may
illustrate this point.
Anyone writing on the Romanesque sculpture of Lincoln Cathedral is fortu
nate to be able to draw on the work of George Zarnecki, whose studies serve as
our starting-point. Professor Zarnecki first wrote a pamphlet on the sculpture in
1963, a work which he twice revised in the 1970s and extensively revised and
augmented in 1988, taking account of the work of others, most notably that of
Professor Eric Fernie and Dr Richard Gem.’ There has also been important work
G. Zarnecki. Romanesque Lincoln: The Sculpture of the Cathedral (Lincoln:

146 Ph illip L indley
by John Larson who was for a time responsible for the conservation programme
as consultant to the Dean and Chapter.*8*Recent archaeological investigations
have also led to radical new reappraisals of the frieze, together with equally
dramatic new interpretations of the architectural context in which they belong.’
The west front of Lincoln Cathedral is the work of several different periods
and styles. The central section of the west end was built by Bishop Remigius of
Fecamp, who was appointed by the Conqueror to the country’s largest see, at
Dorchester-on-Thames in 1067. A council at Windsor decreed the removal of
episcopal sees to large towns, and in 1072-73 Remigius moved his see to
Lincoln, the site of a Norman castle built by William the Conqueror in an area
enclosed by the old Roman town walls. The reasons for the removal of the see
from the extreme south of the diocese to the extreme north were partly strategic
(its military significance had already been grasped by the Romans), partly
economic (Lincoln was a wealthy and important trading city with excellent
waterborne connections with the east coast), and partly a rebuttal of the
Archbishop of York’s claim to a part of the diocese.
Henry of Huntingdon, who was a schoolboy in Lincoln at the end of the
twelfth century, records that the church had been established in a trading area, at
the top of the hill, next to the castle. Writing c 1129-33, he describes how
Bishop Remigius:
bought lands in the upper city itself, next to the castle which was distinguished by
its very strong towers, he constructed a strong church in that strong place, a
beautiful church in that beautiful place, dedicated to the Virgin of Virgins; it was
to be both agreeable to the servants of God and also, as suited the times, invin
cible to enemies.10
Honywood Press, 1988), p. 8 and note 1. The first booklet was entitled Romanesque
Sculpture at Lincoln Cathedral and was published in the Lincoln Minster pamphlets
series. See also E. Fernie. 'Alexander's Frieze on Lincoln Minster’. Lincolnshire History
and Archaeology, 12 (1977), 19-28 and R. Gem. ‘Lincoln Minster: Ecclesia Pulchra,
Ecclesia Fortis', in Medieval Art and Architecture at Lincoln Cathedral. BAACT, 8
(Leeds. 1986). pp. 9-28.
8 J. Larson, ‘The Lincoln Frieze: A Problem of Conservation and Historical
Investigation’, in Romanesque: Stone Sculpture from Medieval England, ed. by B.
Heywood (Leeds: Henry Moore Sculpture Trust, 1993), pp. 28-34.
’ J. Baily, 'Lincoln Cathedral. Progress on the Romanesque Frieze’. Church Building,
33 (1995). 16-21. Baily deploys evidence derived from Professor Lawrence Butler’s
unpublished report, now contained in the CD-ROM. A startling new interpretation of
Bishop Remigius’s 'façade' is contained in D. Stocker and A. Vince. "The Early Norman
Castle at Lincoln and a Re-evaluation of the Original West Tower of Lincoln Cathedral’,
Medieval Archaeology. 41(1997), 223-33.
10 Henry of Huntingdon. Historia Anglorum. London: Rolls Series, vi. 41, ed. by T.
Arnold, p. 212; the translation is that of Gem (as in note 7). p. 9. For Henry see also the
Henry ought to know what he was talking about: he had spent his early years in
the household of Bishop Robert Bloet. then became archdeacon of Huntingdon
and started to write his ‘Historia" at the request of Bishop Alexander the
Magnificent, a work that occupied him on and off till his death in 1155.
The recent publication of Stocker and Vince’s view that the west end of the
cathedral was conceived as a keep containing the bishop’s hall, rather than as a
church façade, mean that interpretation of the structure has become very
controversial. Whatever the original status of the western end of Remigius’s
cathedral (and even its dating is unclear now), all that apparently survives above
ground of his great cathedral is the remains of the west front, damaged and
palimpsested by later additions. It has a Triumphal Arch format: a major arch
flanked by two smaller recesses with flanking niches (the effect was changed
when the top of the central arch was removed in the mid-thirteenth century). The
west front was certainly wider than the aisles of Remigius’s cathedral, and it
may have marked a development towards the screen façade. It seems to have
been very plain, even austere in architectural design, the decoration being
confined essentially to the window openings, string course, and the capitals, a
very much simplified version of Corinthian. (The original entrance doorways
have been removed, and there is no evidence of what they originally looked
like.) It seems also to have been militarily defensible and was indeed used as a
military base in the 1140s."
Until the time of Bishop Alexander, the third Bishop of Lincoln, 1123-48,
there appear to have been no major changes to Remigius’s church. Professor
Zarnecki has argued, however, that under Bishop Alexander there were four
major operations on the west façade:
1. The heightening of the walls of the façade towards the west, south and
north and their decoration with intersecting arches (figs. 5 and 6).
2. The building of the western towers (these, in their upper parts, now
have fourteenth-century additions made by John de Welbourn, c. 1360—
80, who also inserted the Gallery of Kings in the façade).
The other two operations Bishop Alexander is credited with undertaking are
of particular relevance here. These are:
3. The insertion of three new doorways into the west façade.
4. The addition of a frieze of imagery below the string course.
Both of these works have been damaged by later changes to the façade. First, a
large and extremely inventive mid-thirteenth century window was inserted above
Dictionary of National Biography, s.n.

" The Historia Novella by William of Malmesbury, ed. and trans, by K. R. Potter
(London: Nelson. 1955), p. 48. For the context, see also Gem (as in note 7), p. 10.
Fig. 5. South side of west façade of Lincoln Cathedral in 1994. To the right of
the scaffolding in this photograph is bishop Remigius's work. Immediately above
it are the intersecting arches ascribed to Bishop Alexander. The massive
extension to the right and above dates from the thirteenth century. © Author.
Fig. 6. Detail o f the above. This shows Bishop Remigius’s work with the
intersecting arches ascribed to Bishop Alexander above. To the right one can
just see the beginning of the thirteenth-century extension to the façade.
© Author.
Fig. 7. Perpendicular-style window inserted by John de Welburn, in the

fourteenth century, into the framework of a thirteenth-century one. © Author
the central doorway, almost certainly in conjunction with other major works, the
massive extension to the Romanesque façade and, behind these extensions, the
additions of the Morning Chapel on the north and Consistory Court on the
south.12*Then, under John de Welbourn, probably in the 1370s, three new
windows were inserted, the central one sweeping away its Gothic predecessor
(fig. 7). The two lateral ones will have effaced the frieze panels at this level and.
even more damagingly, the splendid cycle of English monarchs was intruded
above the central doorway, completely destroying whatever Romanesque
sculpture was in this location. Deliberate iconoclasm at the Reformation and
later restorations have also affected the frieze.'5
It is an unhappy fact that there is no unequivocal evidence for the dating of
Alexander’s alterations to the west front of the church. The date and attribution
of this work has to be argued primarily from stylistic evidence. It is widely
agreed that Zamecki’s analysis of the doorways (fig. 8) proves their
indebtedness to Abbot Suger’s work at Saint-Denis.14All the stylistic evidence.
'■ G. Russell. ‘The Thirteenth-Century West Window of Lincoln Cathedral’, in

Medieval Art and Architecture at Lincoln Cathedral (as in note 7). pp. 83-89.
' T. Cocke. ‘The Architectural History of Lincoln Cathedral from the Dissolution to
the Twentieth-Century’, in ibid., pp. 148-57. See, too, idem, ‘James Essex. Cathedral
Restorer'. Architectural History, 18 (1975). 12-22.
14Zarnecki (as in note 7). p. 32.
Fig. 8. North side of central portal in May 1996. Much of the detailing belongs
to the eighteenth- and nineteenth-century restorations. © Author
Fig. 9. The panel showing the punishment o f Lust from the ‘Torments' sequence
on the north side of the façade, revealing the join between the panels of stone:
the large dark mark above the shoulder o f the right-hand figure is from the
grouting treatment, designed to reinforce the west front, and carried out in the
1920s (see R. S. Godfrey, Lincoln Cathedral , Lincoln 1930, for this process).
© Central Photographic Unit, University of Leicester.
Fig. 10. ‘The Deluge' and ‘Giants' (panels 16 and 17) in the Ringers' Chapel.
The sculptures are unfinished and out o f narrative sequence. They are carved on
three horizontal pieces o f stone each. © Author
particularly that which relates the portal detailing to Abbot Suger's Saint-Denis,
indicates a terminus post quem of 1140 for them.
One interesting feature of these lavish doorways is the lack of any sculpted
tympana, something at odds with their stylistically up-to-date French-influenced
features. A possible reason for this absence is that Alexander had decided to
confine the main iconographie thrust of the west front to a series of frieze panels
at a higher level. The general iconographie scheme on the south is devoted to
Old Testament scenes and on the north to New Testament parables and an
expanded Last Judgement (though not all the scenes read in the same direction in
the latter series). The narrative order is usually assumed to read out from the
centre to the sides, but the loss of the most important parts, over the central
doorways, makes any overall reconstruction tentative to say the least.15
Recent analysis of the stone types has revealed that Zarnecki was wrong in
assuming that the panels were carved from the same material as the surrounding
ashlar—in fact, those that have been analysed seem to have been usually carved
Ibid., chapters 5 and 6. For a challenge to his reading, see C. Frugoni, ‘Modène-
Lincoln: un voyage manqué'. Etudes et Lettres (1985), 25-52. Professor Zarnecki has
kindly sent me a copy of his yet unpublished paper delivered in 1996. in which he
convincingly rebuts her arguments. I had come to similar conclusions in my CD-ROM
text.
154 Ph iix ip L indley
on end-bedded Ancaster stone blocks quarried some 20 miles away'6—but he

was surely right in dating the frieze to the same period as the doorways. There
are detailed similarities: the foliage of the ‘Adam and Cain’ panel, for instance,
is very like that of the north doorway capitals, and a lion on a colonette now ex
situ is like those on the 'Daniel’ panel.11
Professor Zamecki argued that the panels were carved in a shop probably in the
cathedral precincts, where generally two or more stones were dressed to shape in
preparation for carving, the divisions within scenes generally being horizontal
(fig. 9).16178 This was not an invariable rule: the punishment of ‘Avarice’ (panel 3,
following Zamecki’s numbering system), for instance, is carved on a single panel:
exceptionally the panels in the Ringers’ chapel (the ‘Deluge’ (panel 16) and "Giants’
(panel 17)) are carved onto three horizontal stones (fig. 10). Very importantly, these
last panels are both unfinished and in the wrong narrative place, which indicates that
there was a major disturbance to the frieze before it was put in place.19 Professor
Zamecki has shown that the present placing of ‘Noah Building the Ark’ (panel 13)
(fig. 11) exposes the left edge of the two slabs on which the relief is carved and could
only have been the result of a change of plan. Moreover colonettes with capitals are
carved at the left side of ‘Noah building the Ark' (fig. 12) and on the right-hand side
of ‘God’s covenant with Noah’ (panel 15) showing that they were intended to open
and close the continuous stretch of Noah narratives. The scene of ‘God instructing
Noah to build the Ark’ (panel 12) was intended to be placed further back in the
recess, separated from the first in the continuous narrative series by the respond. Then
came ‘Noah building the Ark’ and a lost or unexecuted scene such as ‘Noah putting
the animals into the Ark’ (so long as it were a narrow panel). Then would have come
the ‘Deluge’ (panel 16) and the ‘Ark on the waters’. This is to assume that the
sculptors worked to a brief which required them to fill the space where there was an
opening in the front, which is now filled by the ‘Daniel’ panel (panel 14), placed
totally out of sequence. Zamecki argues that it was moved here in the 1240s when the
new Gothic window was inserted.20
16 Panel number 3, ‘Avalice’, seems to have been carved in Lincoln limestone, as is

the frame of the 'Daniel' relief. The unpublished petrotrographical report, by D. P.
Jefferson (1991), is in the CD-ROM. An earlier report (1988), by R. Sanderson, is also in
the CD-ROM.
17Zamecki (as in note 7). plates 37 and 38.
Is Ibid., pp. 36-37.
1’ Fernie (as in note 7), pp. 25-26.
‘° Zarnecki (as in note 7), pp. 55-58. He further argues that, originally, over scenes of
the ‘Creation' and 'Fall' was a large composition featuring a ‘Tree of Jesse' on one side
and ’Daniel' on the other, flanking a ‘Coronation of the Virgin'. Fragments of ‘Christ in
Majesty’ and ‘St. Paul' found in 1832. he believes, were originally associated with the
frieze above the north doorway.
The Lincoln CD-ROM projeci 155
Recently a good deal of criticism has been levelled at Zarnecki’s account of

the frieze. Almost all this evidence has resulted from the conservation campaign,
which thus far has been confined to the northern panels. Zarnecki argued that
five courses of the masonry from Remigius’s front were cut away for the
reception of the frieze.’1It has been pointed out that more of the scenes than he
had allowed for were partially cut onto the pre-existing ashlar of the front. This
tended to indicate they were sculpted partially, at least, in situ: this is particularly
clear in the case of ‘The Death of Lazarus' and ‘Dives in Hell’ (panel 8), where
the Hell Mouth is continued onto the ashlar (fig. 2). John Larson suggested that
the ‘Feast of Dives’ (panel 7) was a fragment of a larger panel, which had been
here reused.” He also argues that the ‘Harrowing of Hell’ (panel 4) and ‘Elect in
Heaven" (panel 5) are original to Remigius’s façade, and thus apparently favours
an eleventh-century date for them.25 He demonstrated that at the corner, at the
junction of the ‘Harrowing of Hell’ and the ‘Elect in Heaven’, the blocks of
stone overlapped, showing that they could only have been carved in situ (fig.
13).*4 In the case of the ‘Feast of Dives’, the scene is continued onto the drip-
course or ‘lintel’ (this is not the ashlar of the front but inserted Ancaster
material, according to Dr. Jefferson) and one block, at the bottom left, is part of
the ashlar. ‘Abraham’s Bosom’ (panel 6) (fig. 14) also has parts carved on
blocks of ashlar (the two on the lower right).
There are clearly enormous stylistic variations in the panels, and several
different hands can be detected. The stylistic distinctions between ‘Abraham’s
Bosom’ on the north side and ‘Adam and Cain’ (panel 10) (fig. 15) on the south,
for instance, are so great that it has been asked whether such differences can be
accommodated within a stylistic frame of just a few years.25 An important but
still unpublished report by Professor Lawrence Butler in July 1993 argued, on
the basis of his investigation of the mortar behind four panels which had been
removed and two which were still in situ though with side elevations exposed—
‘the eighteenth-century panel’ 3a; ‘Avarice’ (panel 3); ‘Sodomy’ (panel 2);
‘Harrowing of Hell" (panel 4); ‘Feast of Dives’ (panel 7); ‘Death of Lazarus’ and
‘Dives in Hell’ (panel 8)—that the panels, core, and surrounding stones were all
part of one building operation. One implication is either that the carvers, if they
were twelfth-century craftsmen, were working on pre-existing panels which had21*3
21 Ibid., p. 37.
" Larson (as in note 8), p. 3!.
23 Ibid., p. 33
14
This had already been noted by Zarnecki (as in note 7), p. 38.
■5Larson (as in note 8), p. 33, citing also earlier commentators who viewed the reliefs
as Anglo-Saxon in origin.
Fig. 11. View o f left-hand side of ‘Noah building the Ark’ (panel 15) (photo
taken in 1996). revealing the exposed side of the slab. © Author.
The Lincoln CO-ROM project 157
Fig. 12. Frontal view of ‘Noah building the Ark’ (panel 15). showing colonette
and capital on its left-hand side (photo taken in 1994). © Author
158 Ph 1L1.II' L indley
Fig. 13. Junction between ‘The Harrowing of Hell' and ‘The Elect in Heaven’
(panels 4 and 5), showing that figures from both scenes are carved onto the
same corner blocks. © Author
been placed in this location by Remigius’s workmen,26 or that the work of
inserting the panels was much more extensive than has been supposed. Although
Professor Butler did not pursue the argument further. Dr John Baily recently
suggested that the panels may even have been collected over a long period,
perhaps by Remigius as well as Alexander, ‘by sending carvers they discovered
on their travels back to Lincoln with a commission to cut some panels’.27
Much more investigative work must take place before such issues can be
resolved: does the composition and consistency of the mortar behind the reliefs
really indicate the latter were ali put in place in the same period that the façade
was built? Professor Butler’s report inadvertently provides evidence that mortar
changes between 1090 and 1140 cannot confidently be discriminated because he
assumes, following Professor Zarnecki. that the top, 'lintel', block of the ‘Feast
of Dives’ (panel 7), showing architectural canopies, was originally the string
course of Remigius’s façade (fig. 16). ‘There is no difference in the fawn mortar.
"6There is, of course, Anglo-Saxon precedent for architectural frieze sculpture: see M.
Biddle, ‘A Late Saxon Frieze from the Old Minster’, Antiquaries Journal. 46 (1966),
329-32: D. Kahn. ‘Anglo-Saxon and Early Romanesque Frieze Sculpture in England', in
The Romanesque Frieze and its Spectator (as in note 3), pp. 61-74.
27J. Baily, ‘Lincoln Cathedral: Progress on the Romanesque Frieze'. Church Building.
33 (1995), 20.
which attached [the lintel] to the core’, he comments, ‘from that which attached
the stones of the course above, the stones of the panel and the stones of the base
course below’. Jefferson's petrological analysis, also in the CD-ROM’s conser
vation archive, argues that this lintel is a replacement for a section of original
string course and he believes it was inserted at the same time as the panels below
it.38
The evidence of an interpretative mistake made by the sculptor responsible
for carving a single block of ashlar at the bottom left of panel 7 (fig. 16) also
suggests that the Ancaster panels were insertions. The sculptor carved draperies
of one of Dives’ companions when the block should have been carved as part of
the tablecloth: this suggests that the sculptor responsible for the rest of the relief
did not carve this ashlar block. The Ancaster panels were probably, then,
insertions, with the ashlar subsequently carved by another sculptor in situ. Had
all the relief been carved in situ, no such mistakes would have taken place. It
anyway seems rather unlikely that the large blocks of end-bedded Ancaster stone
on which the reliefs are carved could have been built into the façade during
Bishop Remigius’s construction of the west front rather than inserted during
Bishop Alexander’s episcopate (fig. 1). Modern architectural and archaeological
analyses of Remigius’s building have stressed its plainness and military air: it
seems improbable that it also featured, or was planned to feature, a sculptural
programme, the only possible reason for the placing of the Ancaster slabs.
Whatever the case, it is clear that further technical analysis is essential, on the
southern run of panels, to help resolve such questions. Updated versions of the
CD-ROM should incorporate this new research.
It is possible that the putative historical context of the carving of the reliefs
can explain their stylistic and iconographie anomalies. It seems unlikely that the
panels were put in place whilst Bishop Alexander was still alive, for it is difficult
to believe that the patron would tolerate such changes as were introduced when
the Ringers’ Chapel reliefs were left out of their proper narrative place.31The fact
that the latter sculptures were also left unfinished suggests that they were simply
inserted in a hurry, with little concern for narrative coherence. Historical
evidence provides a context both for Bishop Alexander’s commissioning of the
works and for the failure to install them in his lifetime. Between Bishop
Alexander’s appointment in 1123 and 1139, the bishop with his cousin, Nigel,
Bishop of Ely (1133-67), and his uncle, Roger. Bishop of Salisbury (1107-39),*
‘It appears more likely that a lintel made of oolite was inserted beneath the window
for the full width of the panel, prior to the removal of the ashlar where the panel was to
be inserted. This lintel could then have been carved in situ’.
*9 Cf. Zamecki (as in note 7), p. 58: ‘One is bound to say that the evident muddle in
the present order of the southern section of the frieze strongly suggests that the original
team of sculptors left before completing their work and that the reliefs were put up by
masons unequal to the task’: Femie (as in note 7), p. 27.
Fig. 14. 'Abraham's Bosom’ (panel 6). Note that part of the scene (lower right
hand side) is carved onto the ashlar o f the west front. © Author
The Lincoln CD-ROM projeci 16 1
ftp*
Fig. 15. 'Adam and Cain' (panel 10). ©Author

were enormously powerful figures at court. Stephen’s accession at the end of

1135 was greatly aided by Bishop Roger’s procurement of the treasury on his
behalf.30 The trio acquired considerable power, which collapsed when in 1139
they were all arrested. The contemporary author of the ‘Gesta Stephani’
portrayed Alexander and his kinsmen as bishops who ‘went back [. . .] humble
and downcast, stripped of their empty and ostentatious splendour, to hold their
church property in a manner which befits a churchman'. Before, Alexander had
been a great builder of castles, now he began to found religious houses, all in the
period 113 9-41.31
In the complicated period of the anarchy, when so many magnates changed
allegiance to protect or enhance their own interests, Alexander was soon given
an opportunity to re-establish himself in King Stephen’s favour. Around
Christmas 1140, the keep of Lincoln castle was seized by Ranulph Earl of
Chester and his half brother William of Roumare. Alexander reported the action
to Stephen. Stephen marched on Lincoln where he was received by the bishop
and invested the castle. According to William of Malmesbury the king ‘made a
castle of the church of the blessed Mother of God in Lincoln’ [Ecclesiam beate
Dei genetricis de Lindocolino incastelaverat] (a phrase used of other churches at
the time e.g. Bridlington), but Ranulph escaped, joined Stephen’s other
opponents, marched on Lincoln, took the king prisoner, and sacked the city on 2
February 1141,32 Lincoln was again the scene of warfare in 1144, though this
time, apparently, the cathedral was not used. It seems possible, then, that Bishop
Alexander’s work on embellishing the cathedral will postdate 1141 and might
even follow 1144. This would allow only four years until his death in 1148, and
given that he was abroad in 1144 and 1147, it is highly probable that the frieze
sculptures were not completed at the time of his death but were hastily inserted
into the façade with some parts carved in situ under his successor. Bishop Robert
Bloet. Thus the difficult historical circumstances of Bishop Alexander’s last
decade and his death perhaps before the reliefs were finished suggests that there
were already major discontinuities and disruptions to the original frieze
programme when it was first put in place: it was never finished in accordance
with the patron’s wishes. Such an hypothesis, of course, is wholly in accord with
Professor Zarnecki’s stylistic analysis of the reliefs. Although the middle of a
11A. G. Dyson, ‘The Monastic Patronage of Bishop Alexander of Lincoln’, Journal of

Ecclesiastical History, 26 (1975). 1-24 (p. 2).
31 Ibid., p. 3.
Bishop Alexander, who had celebrated mass for the royal army, followed the lead of
Bishop Henry of Blois and joined Matilda’s court at Winchester. In September 1141 the
king’s party exchanged Earl Robert of Gloucester, who had been captured by them, for
the king. See also P. Dalton. "In Neutro Latere: The armed neutrality of Ranulf II Earl of
Chester in King Stephen’s reign', in Anglo-Norman Studies, xtv. Proceedings of the
Battle Conference, ed. by M. Chibnall (Woodbridge: Boydell Press. 1992), pp. 38-59.
Fig. 16. ‘The Feast of Dives’ (panel 7) after conservation and removal from the
west façade, now on exhibition in the cathedral. The top block o f stone, showing
architectural detailing o f the canopy, is not the same ashlar as the façade
masonry, but is an Aneaste r stone of the same type that was used for the inserted
panels carved by the sculptors (the bottom lefi stone in this photograph is a copy
because the ashlar masonry of the façade onto which the detail was originally
carved was impossible to remove with the rest o f the scene). © Central
Photographic Unit, University of Leicester.
major campaign of research and conservation is perhaps not the best time to
advance theories about the date of the reliefs, it is to be hoped that the ready
accessibility of all the information on CD-ROM will soon facilitate many new
contributions to the debates about the Lincoln frieze.
The successful completion of the Lincoln CD raises some interesting general
issues for the future. It is, as the Lincoln project indicates, technically possible to
design a "shell' into which textual information and imagery can be integrated for
other major conservation projects. If the archive’s design is agreed in advance,
the difficulties of retrospective archiving experienced at Lincoln would be
obviated. When large volumes of state and private money are being devoted to
conservation schemes, should it not be an automatic requirement that all this
information is digitally archived? Any archive—whether issued on CD or on the
Web—should contain all pertinent pre-conservation reports and policy decisions
as well as an archive of the work which actually takes place, so that a relatively
complete archaeological record is available for the future. Not only will this
automatically democratize the often secretive decision-making of unelected
quangos but also it will make huge volumes of information available to
historians and will ensure that later treatments are easier to decide upon: past
failures and successes will have been carefully documented. Conservation
theory—which is far from sophisticated—will have the potential for major
advances on the basis of huge volumes of data. Now is the time to insist that it
should be state policy that every conservation project which receives major state
finances should be digitally archived as a prerequisite of such funding.
Innovative hybride graphische Systeme
zur Denkm alüberwachung und -Verwaltung am
Beispiel historischer W andmalereien—
Ein Erfahrungsbericht
Rolf -J ürgen G rote and A nnette H ornschuch
1. D ie D atenbank d e r m ittelalterlichen W andm alereien in

N iedersachsen— Servicem öglichkeit fü r D enkm alpflege und
K ulturw issenschaft
istorische Wand- und Deckenmalereien sind wichtige Belege unseres

Kulturerbes. Das schmückende Ausmalen von Räumen war in allen
Kulturen und Epochen ein wesentlicher Bestandteil der Raumgestaltung.
In anschaulicher Weise können sie kulturelle Werte und Normen vermitteln. Als
beredte Zeugen ihrer Entstehungszeit spiegeln sie kulturelle Entwicklungen
wider und geben Auskunft über Wandlungen oder Konstanten von Sitte, Moral,
Religion, Weltanschauung, Kunstempfinden und Geschmack im Laufe der
Zeiten. Außerdem können sie eine Chronologie von Handwerks- und Techno-
logiegeschichte aufzeigen.
Niedersachsen hat einen Bestand von mehr als 250 bedeutenden Raumaus
malungen allein des Mittelalters. Inzwischen konnte er im Rahmen einer
rechnergestützten Inventarisierung flächendeckend nach Verbreitung und Alters
schichtung, den wichtigsten Schäden und den kunsthistorischen Inhalten erfaßt
und ausgewertet werden.1 Die Endstufe wird ein Corpus-Werk mit einem ent-
' R.-J. Grote. H. Glashoff und D. Pandlowsky. 'Die Erfassung der mittelalterlichen
Wandmalereien in Niedersachsen—Ein Projekt zum Schutz einer gefährdeten Kunstgattung’,
166 Rolf-Jürgen G rote and A nnette H ornschuch
sprechenden Schadensatlas bilden.’

Um rasch zu ersten Ergebnissen zu kommen, wurde interaktiv zwischen
Dokumentären und Denkmalpflegern eine auf die wesentlichen Informationen
reduzierte Version einer praxisorientierten Datenbank entwickelt.’ Sie ist iterativ
ausbaufähig und setzt sich aus folgenden Bausteinen zusammen:
1. Gebäudebezogene Daten (Eigentümer, Ansprechpartner, Heizung, Mate
rial. Gebäudeschäden usw).
2. Stichwortartige Beschreibung der ikonographischen Zusammenhänge
(Auswertung der Darstellungen nach dem kunsthistorischen Klassifi
kationssystem ICONCLASS der Niederländischen Akademie der Wis
senschaften).
3. Lokalisierung der einzelnen Malereien im Gebäude mit grafischer
Orientierung durch Verteilungsskizzen.
4. Restauratorische Bestands- und Zustandserfassung der Wandmalereien
(Angaben zu Maltechnik, Konservierung, Restaurierung usw).
5. Arbeitsfotos nach normiertem Fotoplan als visuelle Hilfe.*1
Berichte zur Denkmalpflege in Niedersachsen, 14.2 (1994), ss. 81-83; R.-J. Grote. H. Glashoff
und D. Pandlowsky. 'Die EDV-gestützte Erfassung der mittelalterlichen Wandmalereien in
Niedersachsen—Ein Projekt zum Schutz einer gefährdeten Kunstgattung’, in Der Kaiserdom
in Königslutter. Ein Kulturdenkmal auf dem Prüfstand. Interdisziplinäre Service-Leistungen
der Denkmalpflege an einem national bedeutenden Kunstwerk, Arbeitshefte zur Denkmal
pflege in Niedersachsen. 14 (Hannover, 1996; im folgenden zitiert als: Kaiserdom in
Königslutter), ss. 98-101; vgl. außerdem in diesem Zusammenhang: K. Dobrat und U.
Paehlke, Mittelalterliche Wand- und Deckenmalerei in Niedersachsen—eine annotierte
Bibliographie (Dipl.-Arb.), Fachhochschule für Bibliothekswesen, Information und Dokumen
tation (Fh. BID) (Hannover, 1989); Konzept und Realisierung einer Objekt-Datenbank der
mittelalterlichen Wandmalereien Niedersachsens (Faltblatt Fh. BID. Objekt-Datenbank im
Bibliotheks-, Informations- und Dokumentationswesen) (Hannover, 1990); A. Seiberlich:
Probleme ikonographischer Klassifikation am Beispiel mittelalterlicher Wandmalereien
Niedersachsens (Dipl.-Arb.) (Fh. BID. Hannover, 1991); S. Weber. Mittelalterliche Wand
malerei: Literaturbericht zu Fragen der Maltechnik und des mikrobiellen Befalls (Dipl.-Arb.)
(Fh. BID. Hannover, 1990).
1 Für großzügige Unterstützung und Förderung des Projektes sei der Stiftung
Niedersachsen ganz herzlich gedankt. Als erster Einstieg in die Materie entstand
gemeinsam mit der Fh. BID die Diplomarbeit von C. Schladcnhaufen. Conception et
réalisation d ’un programme d’exploitation d’une base de données, développée sous
dBASE 111+ (Rapport de stage Univ. Claude Bernard Lyon I, 1991); außerdem: R.-J.
Grote, Wandmalereien—Probleme ihrer Bestands- und Zustandserfassung. Vortrag
gehalten anläßlich Alte Kirche Idensen—Internationales kunsthistorisches Symposium in
Wunstorf, am 21.11.1991.
’ A. Hilpüsch. Konzept und Realisierung einer Datenbank mittelalterlicher Wandmalereien
Niedersachsens (Dipl.-Arb.). Fh. IuK Hannover (Hannover. 1995) (inzwischen kontinuierlich
den aktuellen Erfordernissen der Denkmalpflegepraxis angepaßt).
Innovative hybride graphische Systeme 167
6. Literaturzitate nach DIN 1505 mit Annotationen zu gebäude- und gemäl

derelevanten Aussagen.
Über ein Recherche-Modul ist es möglich, differenzierte inhaltliche Fragestell
ungen für die ständige Denkmalpflegepraxis zu beantworten.4 Mögliche Such
fragen aus der Vielfalt kunstwissenschaftlicher und denkmalpflegerischer Prob
lemstellungen könnten beispielsweise folgendermaßen lauten:
1. In welchen Kulturdenkmalen des Landkreises Hildesheim existieren
mittelalterliche Wandmalereien?
2. Auf welchen niedersächsischen Wandmalereien des 14. Jahrhunderts
wird die Geburt Christi dargestellt?
3. Welche Wandmalereien im Braunschweiger Dom weisen Schäden durch
Algen oder Salzbelastung auf?
Selbstverständlich lassen sich mit den erläuterten Recherche-Möglichkeiten
nicht nur kunsthistorische Fragestellung beantworten. Bei voller Ausschöpfung
des Entwicklungspotentials lassen sich darüber hinaus eine Fülle aus der Denk
malpflegepraxis kommender Probleme lösen bis hin zu einem objektbezogenen
Monitoringsystem für Maßnahmen der Wartung und Qualitätskontrolle oder—
mit Blick auf den Gesamtbestand niedersächsischer Wandmalereien—einem
flächendeckenden Schadensatlas im Rahmen eines Denkmalerhaltungsplanes.
Das Modellprojekt wurde übrigens so angelegt, daß sich entsprechende
praxisorientierte Hilfen auch für andere wichtige Kulturdenkmale—wie z.B.
Altäre—daraus ableiten lassen könnten.
Wegen erheblicher Finanzierungsprobleme ist trotz des vorhandenen know
hows das große Entwicklungspotential dieses wichtigen Arbeitsinstrumentes zur
Zeit allerdings nur ansatzweise ausgeschöpft.
2. R echner gestützte D enkm alinform ationssystem e f ü r historische

W andm alereien
2.1 Problemstellung
Wegen immer knapper werdender Resourcen müssen auch bei der Betreuung des
einzelnen Kulturdenkmals Personal und Fördermittel noch effizienter eingesetzt
werden, ohne daß Service-Leistungen und Bearbeitungsgüte darunter leiden. Wie
dieses Ziel zu erreichen ist—und welche positiven Lösungsmöglichkeiten bereits
bestehen—, läßt sich an den Wandmalereien als besonders problematischer
4 F. Papendorf und R.-J. Grote. ‘Servicemöglichkeiten der Denkmalpflege—Datenbank

der mittelalterlichen Wandmalereien in Niedersachsen’. Berichte zur Denkmalpflege in
Niedersachsen, 17.2 (1997). ss. 97-100.
168 R olf -Jürgen G rote ano A nnette Hornschuch
Kunstgattung beispielhaft darstellen.56

Auch Kulturgüter in Innenräumen wie historische Wandmalereien sind viel
fältigsten Schadensfaktoren ausgesetzt.'’ Um die komplexen Wirkungsmechan-
ismen für Therapie- und Wartungskonzepte zu erfassen, zu interpretieren und zu
quantifizieren, ist eine methodisch abgesicherte Arbeits- und Dokumentations
systematik dringend erforderlich (wobei unter ’Dokumentation' die "planmäßige
Sammlung, Ordnung und Nutzbarmachung von Dokumenten und Daten' zu ver
stehen ist).7 Da die bisherigen Verfahren bei der Bearbeitung komplexer Frage
stellungen erkennbar an Grenzen gestoßen sind, bilden rechnergestützte Arbeits
hilfen inzwischen innovative Schwerpunkte auf internationaler Ebene.*
5 Vgl. z. B.: Historische Technologie mul Konservierung von Wandmalerei,

Vortragstexte der dritten Fach- und Fortbildungstagung der Fachklasse Konservierung
und Restaurierung (Schule für Gestaltung. Bern. 5. und 6. November 1984. Bern. 1985):
Schäden an Wandmalereien und ihre Ursachen. Ein Forschungsprojekt des Bundes
ministers für Forschung und Technologie. Aktuelle Vorberichte zu den ersten inter
disziplinären Befunden. Arbeitshefte zur Denkmalpflege in Niedersachsen. 8 (Hannover.
1990: im folgenden zitiert als: Schäden an Wandmalereien)'. Forschungsprojekt Wand
malerei-Schäden. Ein Förderprojekt des Bundesministers für Forschung und
Technologie. Schlußbericht zu den interdisziplinären Befunden. Arbeitshefte zur Denk
malpflege in Niedersachsen. 11 (Hannover. 1994: im folgenden zitiert als: Forschungs
projekt Wandmalereischäden): Kaiserdom in Königslutter.
6 Vgl. z. B.: S. Fitz, 'Metallkorrosion in Innenräumen', in Materialkorrosion durch
Luftverunreinigung, Tagung Köln. 17.-19.10.1984 (VDI-Kommission Reinhaltung der
Luft. Düsseldorf. 1985), ss. 83-94 (VDI-Berichte. 530): S. Fitz. E. Fitz-Ulrich. G.
Frcnzel. R. Krüger, H. Kühn und E. Schmidt, Die Einwirkungen von Luftverunreinig
ungen auf ausgewählte Kunstwerke mittelalterlicher Glasmalerei (Berlin, 1984): D.
Camuffo und A. Bernardi. 'Microclimate-induced weathering on the Sistine Chapel'.
European Cultural Heritage. Newsletter on Research 2.1 (Jan. 1988), ss. 36-42: D.
Camuffo und A. Bernardi, Analisi microclimatica al Cenacolo Vinciano. Bolletino
Geofisico 14.3 (Die. 1991): D. R. Fuchs und J. Leissner. Environmental monitoring using
Glass Sensors: Relevant Features for small-scale mapping at Monuments. 3rd Interna
tional Conference on Non Destructive Testing (Viterbo. 1992).
H. Glashoff. 'Möglichkeiten der Datenorganisation und Wissenschaftsdokumentation
am Beispiel des Pilotprojektes "Wandmalerei-Schäden und ihre Ursachen" '. Restaur
ierung von Kulturdenkmalen. Beispiele aus der niedersächsischen Denkmalpflege
(Berichte zur Denkmalpflege in Niedersachsen. Beili. 2) (Hameln, 1989), ss. 21.
* Inzwischen ist es gelungen, wichtige Einzelprobleme einer Lösung zuzuführen. Dies
hat z. B. das Kolloquium .Denkmalpflege und computerunterstützte Dokumentation und
Information' in Stuttgart. 1.-2.12.1992. beispielhaft gezeigt. Vgl. hierzu: Denkmalpflege
und computergestützte Dokumentation und Information. Dokumentation der Tagung des
Deutschen Nationalkomitees für Denkmalschutz in Zusammenarbeit mit der Vereinigung
der Landesdenkmalpfleger in der Bundesrepublik Deutschland und dem Informations
zentrum RAUM und BAU der Fraunhofer-Gesellschaft 1. und 2. Dezember 1992 in
Stuttgart (Schriftenreihe des Deutschen Nationalkomitees für Denkmalschutz. 44).
Bei historischen Kulturgütern verhält es sich wie in der Medizin: Auch bei
ihnen hängt die Therapie ganz entscheidend von einer sorgfältigen Anamnese
und Diagnose ab - irreparable Langzeitschäden sind die Folgeerscheinungen von
Maßnahmen, die mangelhaft eingeleitet und weitergeführt worden sind.
Historische Wandmalereien sind komplizierte Verbundsysteme. In der Regel
gestalten sich Konzipierung und Durchführung einer systematischen Maß
nahmenkette bei der Objektbetreuung daher außerordentlich schwierig. Es muß
das spezifische Schadenspotential des jeweiligen Kulturdenkmals, seiner Um
gebung und seiner Umwelteinflüsse berücksichtigt werden. Dabei sind sehr
heterogene Informationen zu erfassen, auszuwerten, zu vernetzen und nach
Interpretation in Therapiekonzepte umzusetzen.
Die wichtigsten Informationsträger sind:
1. visuelle Informationen: Fotos, Bestandspläne, Kartierungsunterlagen usw.
2. Textinformationen: Zum Bau, zu den Wandmalereien, zur Phänomen-*5
Anregende Impulse vermittelten die Fachtagungen 1993 und 1994 des Rheinischen Amts
für Denkmalpflege zum Thema: EDV in der Denkmalpflege (vgl. dazu auch: ‘Tagungs-
bericht. Pulheim—Abtei Brauweiler. Fachtagung: EDV in der Denkmalpflege'. Denk
malpflege im Rheinland Nr. 2. (1994). ss. 65-72. Wichtige Vorarbeiten leistete 1990 unter
anderem auch die Arbeitsgruppe .Bestandsaufnahme' des BMFT-Verbundprojektes
'Steinzerfair. Vgl. außerdem: Dokumentation in der Restaurierung. Akten der Vorträge
der Tagung in Bregenz 23.-25. November 1989 (Salzburg. 1994): Bestandserfassung und
Bestandsanalyse an Kulturdenkmalen. I. Fortbildungsveranstaltung für Restauratoren am
5. Matz 1993 in Hannover. Niedersächsisches Landesverwaltungsamt—Institut für
Denkmalpflege. Materialien zur Fort- und Weiterbildung.! (Hannover 1993). Für die
Bearbeitung von Kulturgut, vor allem historischer Bausubstanz, sind statistische
Erklärungsmodelle (Cluster- und Faktorenanalysen) eine wichtige Interpretationshilfe.
Führend auf diesem Gebiet ist das Laboratoire de Recherche des Monuments Historiques,
Paris. Als visuelle Hilfe bei der Bearbeitung der bedeutenden Wandmalereien in der
Franziskanerkirche von Arezzo (Kapellenausmalung von 1452-1466 durch Piero della
Francesca) wurde mit gutem Erfolg ein rcchnergestütztes Kartierungsverfahren mit einer
modifizierten Standardsoftware eingesetzt, das praxisoptimiert ist (den Herren Drs. Mauro
Matteini und Arcangelo Moles, die dem Verfasser eine Demonstration am Objekt
ermöglichten, sei an dieser Stelle nochmals herzlich gedankt). Vgl. auch St. Casciu. G.
Centauro und M. Chimenti, ‘Neue EDV-gestützte Verfahren im Praxiseinsatz—Arezzo
(Italien). Franziskanerkirche', in Kaiserdom in Königslutter, ss. 113-17: G. Centauro und
M. Maffioli. Un progretto per Piero della Francesca (Firenze, 1989). Außerdem: S.
Beck. C. Temme und H. Rademacher, ‘Facility Management als Werkzeug für die
Denkmalpflege-, in Kaiserdom in Königslutter, ss. 88-91: R.-J. Grote. ‘Objektdokumen-
tation? Innovative Arbeitshilfen für die Denkmalpflege—EDV und historische
Wandmalereien', in Kaiserdom in Königslutter, ss. 82-87; H. Gutscher und E. Favre-
Bulle. ‘EDV-gestützte Dokumentation eines interdisziplinären Restaurierungsprojektes:
Die Wandmalereien in der Cluniazenserkirche Romainmötier (Schweiz)’, in Kaiserdom in
Königslutter, ss. 106-12.
kartierung und zu Analysedaten.

3. Analysedaten.
Die Schwierigkeit besteht vor allem darin, daß zahlreiche Informationsträger
zueinander in Beziehung gesetzt und vernetzt werden müssen, wobei eine große
Zahl von Interaktionen möglich ist. die eigentlich nur EDV-gestützf’ oder mathe
matisch-statistisch (wie z.B. durch Cluster- und Faktorenanalysen)910ausgewertet
werden können:
1. Interaktion der Informationsträger zum Raum.
2. Exakte Zuordnung der Informationsträger zu der zeitlichen Komponente:
Die einzelnen Bearbeitungsstufen müssen—vor allem auch bezogen auf
frühere Maßnahmen—im Zeitkontinuum multitemporal vergleichbar sein,
denn die denkmalpflegerische Betreuung von Wandmalereien ist ein nie
endender zyklischer Prozeß. Dessen wichtigste Phasen (Anamnese—Diag
nose—Therapie—Monitoring—Qualitätskontrolle) münden schließlich
wieder in neue Erhaltungsmaßnahmen—sofern nicht primäre Schadens
ursachen ermittelt und ausgeschaltet worden sind.
3. Die jeweilige Dokumentation muß daher auf die gleiche Art gewonnen
werden, um die Vergleichbarkeit der Daten zu gewährleisten.
4. Schließlich müssen die Interdependenz der Schadensphänomene selbst
und deren dynamische Abläufe erfaßt und für Therapiekonzepte sinnvoll
miteinander in Verbindung gebracht werden.
2.2 Lösungsmöglichkeiten
Der Einsatz der in den letzten Jahren rasch weiterentwickelten innovativen
9 Erste Erfahrungsberichte über entsprechende Praxisanwendungen des Nieder

sächsischen Landesamtes für Denkmalpflege u. a. bei R.-J. Grote, V. Stelzer und G.
Weinzierl, ‘EDV-gestützte Zustandserfassung und Dokumentation'. Bautenschutz und
Bausanierung. 14.8 (1991), ss. 46-49: R.-J. Grote und V. Stelzer. 'Innovative Verfahren
bei der Restaurierung von Kulturdenkmalen’ in .Berichte über die Tätigkeit der Bau- und
Kunstdenkmalpflegc in den Jahren 1989-1990", Niedersächsische Denkmalpflege 14
(Hameln. 1992). ss. 36-44 (im folgenden zitiert als Niedersächsische Denkmalpflege)'. R.-
J. Grote, V. Stelzer und G. Weinzierl. ‘Zustandserfassung: Der Geschichte auf der Spur.
Ein EDV-gestütztes Verfahren für historische Baudenkmäler." Bausubstanz. 8.10 (1992).
ss. 30-33; H. Glashoff. 'Dokumentation und Datenmanagement', in Forschungsprojekt
Wandmalereischäden, ss. 34-36; R.-J. Grote. ‘EDV-gestützte Hilfen für die Be
standserfassung und Zustandsanalyse bei Schäden an Kulturdenkmalen’, ebd. ss. 37-40.
10 H. Glashoff. ‘Analytisch-statistische Interpretationshilfen zur Ursachenforschung
von Schäden an historischen Wandgemälden', in Niedersächsische Denkmalpflege. 14. ss.
45-53.
Techniken und Verfahren, wie z. B. Bilddatenverarbeitung, Multispektral

analyse, CAD und computergestützte Dalenvisualisierungs- und Verwaltungs
techniken, ermöglicht in vielen Belangen eine sinnvolle Anwendung in der
Denkmalpflegepraxis. Im Vordergrund müssen dabei die Möglichkeiten einer
straff geplanten, interaktiven Maßnahmenkette bei der Objektbearbeitung stehen.
Der Einsatz der Datenverarbeitung bietet die große Chance einer effektiven
Verknüpfung so verschiedenartiger Datenträger wie Kartierungen, Textinforma
tionen und Bilder. Messungen und Analysen, um die dringend erforderliche
Verbesserung der Zeitabläufe sowie der Bearbeitungsgüte zu erreichen.
Im Rahmen verschiedener Forschungsprojekte" sind im Niedersächsischen
Landesamt für Denkmalpflege (NLD) die Grundlagen für ein mehrdimensi
onales, bildbasierendes (rechnergestütztes) Informationssystem geschaffen wor
den, das gleichermaßen innovative Lösungen für die effiziente Bearbeitung des
einzelnen Kulturdenkmals und erste positive Ansätze für einen fläch
endeckenden Schadensatlas mit kunsthistorischen Beschreibungen für die multi
disziplinäre Auswertung enthält. Im Praxiseinsatz konnten inzwischen vielfältige
Erfahrungen und weiterführende Anregungen gesammelt werden. Dabei ergab
sich als wichtigstes Fazit, daß ein verbindliches Standardverfahren der Objekt
bearbeitung nicht realisierbar ist. sondern die Dokumentationsmethodik dem
spezifischen Anforderungsprofi] des jeweiligen Kulturdenkmals adaptiert wer
den muß.
2.2.1 Der Braunschweiger Dom und die Stiftskirche in Königslutter
Als exemplarische Belege können unter anderem die Bestands- und Zustandser
fassung der bedeutenden Wandmalereien in der Stiftskirche von Königslutter
und der Domkirche in Braunschweig gelten: Kulturdenkmale, deren außerord
entlich umfangreiche Ausmalungszyklen eine sehr unterschiedliche Schadens
problematik und Restaurierungsgeschichte aufweisen.
" Vgl. die entsprechenden Themenschwerpunkte in: Schäden an Wandmalereien.

sowie: Forschungsprojekt Wandmalereischäden: außerdem Anmerkung 9 und R.-J. Grote
und P. Königfeld. 'Dic Stiftskirche in Königslutter und ihre Ausmalung—Erhaltungs-
planungen für ein gefährdetes Kulturdenkmal’, Niedersächsische Denkmalpflege. 15. ss.
81-94. (Die Methodenplanung wurde während des DBU-Vorprojektes zum Modellvor
haben Stiftskirche Königslutter im Rahmen der innovativen Fortschreibung des Projekt
antrages inzwischen konzeptionell aktualisiert. Vgl. dort S. Beck und C. Temme.
Möglichkeiten der (EDV-gestiitzten) Kartierung als ein Teilbereich der Dokumentation
bei der Bearbeitung von Objekten. Report (Hannover: Nicdcrsächsisches Landesamt für
Denkmalpflege. 1996): D. Pandlowsky, Innovative Möglichkeiten der EDV-gestiitzten
Daten- und Bildspeicherung. Report (Hannover: Niedersächsisches Landesamt für
Denkmalpflege, 1996).
172 Rolf-Jürgen G rote ano A nnette H ornschuch
Als erster Einstieg wurde bei beiden eine generalisierende Schadenserfassung

gewählt. Die angewandten Verfahren unterscheiden sich wegen der objekt
spezifischen Gegebenheiten jedoch erheblich: In Braunschweig eine Foto
dokumentation auf CD-ROM mit rechnergestützter Suche nach restauratorisch
naturwissenschaftlichen und kunsthistorischen Fragestellungen112*— in Königs
lutter dagegen eine konventionelle “hündische" Phänomenkartierung in Ver
bindung mit einem (bilddaten-basierten) 3D-Raummodell. das nach entsprech
ender Praxisoptimierung vor allem der Visualisierung von Wandmalereischäden
in Korrelation mit den bauimmanenten Schadensagentien sowie möglicher
Umweltbelastungsfaktoren dienen soll.15
Die vom NLD konzipierte Arbeitssystematik ist konsequent auf die spezi
fische Problematik des Ausmalungsprogrammes im Braunschweiger Dom mit
dem Ziel maximaler Arbeitseffizienz zur Kostendämpfung abgestimmt. Bereits
im Vorfeld der Konzepterstellung gemeinsam mit dem zuständigen Stadt
kirchenbauamt durchgeführte Kosten-Nutzenanalysen erwiesen die gewählte
Lösung als optimal.
Wesentliches Kennzeichen ist der modulare Aufbau: Die Erfassungs
systematik kann als Ganzes oder auch nur in einzelnen Bausteinen zum Einsatz
12R.-J. Grote, H. Glashoff. G. Drescher und D. Pandlowsky. ‘Die Wandmalereien im

südlichen Querhaus des Braunschweiger Doms—aktuelle Befunde und ihre Dokumenta
tion'. Bendile zur Denkmalpflege in Niedersachsen, 15.3-4 (1995). ss. 109-15; R.-J.
Grote. H. Glashoff und D. Pandlowsky. ‘Arbeitsökonomie und Kosteneffizienz als
Grundlagen denkmalpflegerischer Zielplanung: Die Wandmalereien im Braunschweiger
Dom’, in Kaiserdom in Königslutter, ss. 92-97. Für großzügige Unterstützung und
Förderung seien der Richard Borek Stiftung, dem Braunschweigischen Vereinigten
Kloster- und Studienfonds sowie der Domstiftung gedankt, die in Verbindung mit
erheblichen Bundes- und Landesmitteln die Durchführung des Projekts ermöglichten.
Ganz herzlich bedanken möchte sich der Verfasser auch bei Herrn Dipl.-Ing Norbert
Koch sowie Frau Karin Kautzer-Lehmann vom Stadtkirchenbauamt. die über Finanzier
ungsplanung. örtliche Bauleitung und Koordination aller Beteiligten hinaus mit mannig
fachem fachlichen Rat und vielseitiger Hilfestellung ganz erheblich zum Gelingen des
Projektes beigetragen haben. Schließlich sei an dieser Stelle dem interdisziplinären Team
für die konstruktive Zusammenarbeit im Rahmen der Projektarbeit Dank gesagt.
11Ebd. S. Beck, E. Behrens. D. Pandlowsky und C. Temine. ‘Maßnahmenbegleitende
Dokumentation—Entwicklung eines Orientierungssystems'. Kaiserdom in Königslutter.
ss. 14-19: J. Heckes. A. Hornschuch und P. Haferland. ‘Möglichkeiten der Bilddaten
verarbeitung. Visualisierungen und Kartierungen auf der Grundlage von Photos mit
Standardsoftware'. Kaiserdom in Königslutter, ss. 118-21. Der Deutschen Bundesstiftung
Umwelt und dem Braunschweigischen Vereinigten Kloster- und Studienfonds sei an
dieser Stelle nochmals herzlich gedankt, die mit erheblichen Fördermitteln das
interdisziplinäre Projekt ermöglicht haben. Dank gilt aber auch der Kirchengemeinde und
allen übrigen Helfern und Team-Mitgliedern, die mit Rat und Unterstützung zum
Gelingen der Untersuchungen beigetragen haben.
kommen. Zur Verknüpfung der verschiedenen Datenträger wurde die Datenver

arbeitung so einfach wie möglich konzipiert und praxisoptimiert. Besonderer
Wert wurde darauf gelegt, daß der Zugriff auch auf bewährte "hündische”
Verfahren jederzeit möglich ist.
Im Braunschweiger Dom handelt es sich um eine restauratorische Schneller
fassung des sehr umfangreichen Wandmalereibestandes und seiner wichtigsten
Schadensparameier. Sie soll einer anschließenden interdisziplinären Bearbeitung
die Konzentration auf objektrelevante Schadensbereiche ermöglichen. Korre
spondierend mit der Befunderhebung durch szenenbezogene 'Checklisten' wurde
als visuelles Medium eine systematische Reihenfotografie gewählt, wobei es sich
um mehrere hundert Einzelszenen handelt.
Ihre Einzelsequenzen wurden auf Photo-CD überführt und können mit einer
einfachen Datenbank phänomen- und themenbezogen angesteuert werden (Shoe-
box Photo-Access für Windows/PhotoEdge für Windows). Parallel dazu wurde
ein "hündisches” Ablagesystem der "Ortho-Fotos” nach Einzelszenen geschaf
fen. Auch ihnen ist die entsprechende "Checkliste” (also die Erfassungsmaske)
des Restaurators jeweils zugeordnet.
In einer zweiten Bearbeitungspha.se wurden die Befunde auf Übersichts
zeichnungen (layer auf Transparentfolien) übertragen und zusammenfassend im
szenischen Kontext visualisiert. Damit ist es möglich, die Schadensverteilung
und andere objektrelevante Phänomene wie z. B. den Anteil der Originalsubstanz
an Wandmalerei mit dem Komplexitätsgrad der Schäden quantifizierend in
einem ersten Überblick zu erfassen und zu korrelieren.
Cluster- und Faktorenanalysen (Algorithmus nach WARD. Euklidische
Distanz, Software SYSTAT) dienten als statistische Auswerlungshilfen. um die
dominanten Schadensbereiche aus den sekundären herauszufiltern und damit
später den gezielten Einsatz eines interdisziplinären Untersuchungsteams zu
ermöglichen.
2.2.2 Forschungsprojekt ‘Wandmalerei-Schäden’
Auch im Rahmen des vom Bundesminister für Forschung und Technologie

finanzierten Forschungsprojektes "Wandmalerei-Schäden” wurde eine innova
tive Arbeits- und Dokumentationssystematik realisiert (die allerdings inzwischen
erheblich verfeinert und praxisoptimiert worden ist).14 Dabei konnten ver-
14 Siehe auch VDI-Richtlinie 3798 .Untersuchung und Behandlung von

immissionsgeschädigten Werkstoffen insbesondere bei kulturhistorischen Objekten'; P.
W. Mirwald. 'Erhebungen. Situationsstudien und Handlungsorganisation im Koopera
tionsverbund aller an Untersuchungen und Erhaltungsmaßnahmen am Baudenkmal
beteiligten Stellen'. Bautenschutz und Bausanierung, Sonderh (1989), ss. 16-20; H.
Glashoff, 'Möglichkeiten der Datenorganisation und Wissenschaftsdokumentation am
174 Rolf-Jürgen G rote and A nnetti; H ornschuch
schiedene EDV-gestützte Verfahren piloterprobt werden. Ziel war die Ent

wicklung eines ganzheitlichen modularen Denkmalinformationssystems. das den
jeweiligen objektspezifischen Gegebenheiten flexibel angepaßt werden kann.
Der Projektverlauf war geprägt von der Bewältigung einer übergroßen Fülle
heterogener Daten. Dies zeigte sich z.B. schon bei ganz normalen restaura
torischen oder naturwissenschaftlichen Phänomenkartierungen''—wobei die
"Kartierung” allerdings nur als Modul des umfassenderen Denkmalinformations
systems aufzufassen ist. Als wichtige visuelle interdisziplinäre Erkennungs- und
Entscheidungshilfe durch Schichtung von Transparentfolien in Kartierebenen
(layer) konzipiert, war bei komplexen Schadensbereichen die beabsichtigte
Ablesbarkeit mit herkömmlichen Verfahren in der Regel nicht mehr zu
bewältigen.
Es ergab sich daher zunehmend, daß die Entwicklung rechnergestützter
Verfahren zur Bewältigung der bei der Erfassung. Analyse und Dokumentation
der Wandmalereischäden progressiv wachsenden Datenfülle unabdingbar war.*15
Beispiel des Pilotprojektes ''Wandmalereischäden und ihre Ursachen” '. in Restaurierung

von Kulturdenkmalen (Berichte zur Denkmalpflege in Niedersachsen. Beih. 2: Hannover.
1989), ss. 21-24; S. Beck. U. Geruschkat und H. Glashoff. 'Foto und Raumkoordinate als
Hilfsmittel der Befunddokumentation', in Scheiden an Wandmalereien, ss. 16-18; H.
Glashoff. 'Dokumentation und Datenmanagement', in Forschungsprojekt Wandmalerei-
Schäden. ss. 35-36.
15 Speziell zum Kartieren und anderen visuellen Arbeitshilfen vgl. B. Fitzner und R.
Kownatzki. ‘Kartierung und empirische Klassifizierung der Verwitterungsformen und
Verwitterungsmerkmale von Natursteinen an geschädigten Bauwerkspartien'. Bautenschutz
und Bausanierung, Sonderh (1989), ss. 21-25; B. Fitzner und R. Kownatzki.
'Bauwerkskartierung’, Der Freiberufliche Restaurator. H. 4. 1990, ss. 25-40; 'Die
Dokumentation in der Bestandsaufnahme—Untersuchung, Bewertung und Restaurierung
dcnkmalpflegerischer Objekte'. Bericht der Arbeitsgruppe 'Bestandsaufnahme' des BMFT-
Projekts. Steinzerfall'. Bautenschutz und Bausanierung. Sonderh (1990); ferner: H. F.
Reichwald. ‘Möglichkeiten der zerstörungsfreien Voruntersuchung am Beispiel der
ottonischen Wandmalereien in St. Georg Reichenau-Oberzell', in Historische Technologie
und Konservierung von Wandmalereien (Bern. 1985), ss. 106-31: G. Drescher und M. K.
Jensen. ‘Schadenskartierungen. Fallbeispiel Ev.-ref. Kirche in Krummhöm-Eilsum'. in
Forschungsprojekt Wandmalerei-Schäden, ss. 88-92: vgl. auch H. P. Autenrieth und P.
Turek. ‘Modelle für eine grafische Inventarisation mittelalterlicher Wandmalereien’, in
Schäden an Wandmalereien, ss. 27-32; S. Beck, ‘Computer Aided-Design (CAD) als
Arbeitsmittel’. Niedersächsische Denkmalpflege, 14. ss. 53-62: U. Geruschkat und H.
Glashoff. 'Visuelle Objektdokumentation', in Schäden an Wandmalereien, s. 15: T. Trapp.
"Optische Methoden der Schadenserfassung', in Forschungsprojekt Wandmalereischäden.
ss. 45—48: S. Beck. 'Die Kartierung als Mittel der Bestandsaufnahme", in Forschungsprojekt
Wandmalerei-Schäden, ss. 41-44: D. Gadesmann und P. Königfeld. 'Möglichkeiten einer
"einfachen Phänomenkartierung" als Grundlage der Schadensanalyse-Schloßkapeiie in
Celle’, in Kaiserdom Königslutter, ss. 102-05: zum aktuellen Entwicklungsstand vgl. z. B.
den Beitrag von Jürgen Heckes et al. Anm. 13.
Das Denkmalinformationssystem wurde an ausgewählten Objekten bzw. Objekt

bereichen erfolgreich piloterprobt. Eine Einsatzoptimierung konnte allerdings
nicht mehr erfolgen, da inzwischen das Forschungsprojekt seinen Abschluß
gefunden hatte.
Da bei denkmalpflegerischen Maßnahmen viele Beobachtungen, Untersuch
ungen und Messungen durchzuführen sind, erwies sich die Einführung eines
definierten dreidimensionalen Koordinatensystems in Verbindung mit einer
relationalen Datenbank als sinnvoll. Die exakte Zuordnung der vielfältigen
interdisziplinären Aktivitäten zu Koordinaten ermöglichte eine gemeinsame
Basis für die spätere Auswertung, Dokumentation und Archivierung.
Die Erfassung des Baudenkmals erfolgte innen und außen dreidimensional
über Tachymetrie. Markante Meßpunkte, die das Baudenkmal in seiner
Geometrie beschreiben, wurden in Fotos eingetragen. Somit war eine eindeutige
Identifizierung der Punkte bei der nachfolgenden Rekonstruktion durch CAD
gegeben. Bei Bedarf können über das CAD-System Längs- und Querschnitte
sowie 3D-Gittermodelle erzeugt werden.
Als Grundlage für die anschließende Kartierung von Zuständen und
Schadensbereichen dienten herkömmliche Fotografien. Bei Wandmalereien ist es
dringend erforderlich, daß diese Einzelbilder zusammenhängende Gemälde
partien erfassen. Sie sollen möglichst frontal zur Wand aufgenommen werden,
um starke Verzerrungen zu vermeiden. Durch Überdeckung der Einzelbilder
wurde ein ‘Fotomosaik’ zusammengefügt. So war es möglich, alle Bereiche des
Baudenkmals lückenlos abzubilden.
Die Einzelbilder wurden sodann nach ihren Koordinaten erfaßt und den
dreidimensionalen Raumkoordinaten des Baudenkmals zugeordnet. Über foto
grammetrische Stereoaufnahmen der Wandpartien konnten die Ecken und Mitten
der Einzelbilder dreidimensional bestimmt werden. Dabei dienten die über die
Vermessung (Tachymetrie) aufgenommenen Punkte als Paßpunkte für die
Fotogrammetrie. Damit war der Bezug zwischen den zweidimensionalen
Bildkoordinaten und den dreidimensionalen Raumkoordinaten hergestellt und
jeder Punkt eindeutig lokalisierbar.
Für die Zustands- und Schadenserfassung wurden im Projektverlauf ver
schiedene Verfahren getestet. Im Vordergrund stand die Erarbeitung EDV-
gestützter Kartierungsunterlagen als Grundlage für weiterführende Auswertungen.
Zunächst wurden videografische Aufnahmen erstellt, digital aufbereitet und als
Einzelbilder auf einem Laptopbildschirm dargestellt. Eine entsprechende Software
sollte die Phänomenkartierung direkt vor der Wandmalerei auf den Hinter-
grundbildem des Laptops ermöglichen. Wegen der technischen Grenzen des
Laptops wurde die Realisierung dieses Vorhabens allerdings dann zurückgestellt.
Als Alternativlösung wurde versucht, ein eigenständiges Kartierungspro
gramm zu entwickeln, das auf kalibrierten, nicht digital vorliegenden Einzelfotos
basierte. Dabei fehlte allerdings der direkte Bezug von Bearbeitungsfläche und
Kartierung: Das Foto liegt separat auf einem Digitalisiertablett, während die
Kartierung isoliert und ohne bildlichen Bezug auf dem Bildschirm erscheint. Ein
weiterer Nachteil bestand in den hohen Entwicklungskosten einer speziellen
Kartierungssoftware, die menuegesteuert auch von Compulerlaien angewendet
werden kann.
Daher erschien es sinnvoll, bereits auf dem Markt erhältliche CAD-Standard-
software entsprechend den spezifischen Anforderungen zu modifizieren und
anzuwenden. Auch dabei wurde ein auf dem Digitalisiertablett liegendes,
kalibriertes—möglichst orthogonales—Einzelfoto als Kartierungsunterlage ver
wendet. Als Orientierungshilfen wurden Strichzeichnungen des entsprechenden
Fotos erstellt und auf den Bildschirm eingeblendet, um eine Lokalisierung der zu
kartierenden Phänomene zu ermöglichen. Als großer Vorteil erwies sich die
Flexibilität des Verfahrens: Die Kartierungen lassen sich entweder konventionell
vor Ort anfertigen und anschließend stationär weiterbearbeiten oder unmittelbar
am Objekt in den Laptop eingeben.
Die erfaßten Bildinformationen wurden über eine Schnittstelle in einer
relationalen Datenbank abgelegt und verwaltet. Die Datenbank korrelierte die
Eintragungen der Zustands- oder Schadensphänomenerfassung sowie der text
lichen Beschreibung miteinander, da sie über das Koordinatensystem mitein
ander verbunden sind. Fotos, Videoaufnahmen usw. können leicht über die
Datenbank diesen Eintragungen zugeordnet werden. Dadurch war sowohl ein
schnelles Auffinden aller gespeicherten Dokumente wie auch ein jederzeitiges
Reproduzieren der Informationen und eine erleichterte Kommunikation mit
anderen Projektteilnehmem, aber auch ein langfristiges, einheitliches und
sicheres Abspeichern der Informationen gegeben. Bei zusätzlichem Ablegen der
Fotos auf langzeitspeichernde Bildplatten könnten diese Fotoinformationen zu
jedem beliebigen Befund und zu jedem späteren Zeitpunkt wieder auf dem
Bildschirm dargestellt werden. Damit ließe sich in Form eines späteren
Monitoring eine Langzeit-Zustands-/Schadensbeobachtung durchführen.
2.2.3 Fotografische/visuelle Bestandserfassung in Haus Altenkamp
Problemstellung und Auswahl der Bearbeitungstechniken

Wegen der großen kunst- und kulturgeschichtlichen Bedeutung der Wand
malereien werden zusätzlich zu den konventionellen Verfahren der Bestands
und Zustandserfassung/Dokumentation weitere innovative Lösungen piloter
probt. die qualifiziertere Erkenntnisse und Entscheidungshilfen ermöglichen. Ein
Bearbeitungsschwerpunkt ist daher der Einsatz der Bilddatenverarbeitung zur
Schadensfeststellung und als Monitoringsystem in Verbindung mit multi
spektralen Aufnahme- und Auswertungstechniken. Angelehnt an Verfahren aus
der Fernerkundung, die bereits seit geraumer Zeit multispektral zerlegte Bild
daten von Satellitensystemen zur Flächendetektion, -Überwachung und kartie-
Fig. I. Haus Altenkamp, Ansicht von Westen. © Landesamt für Denkmalpflege,

Hannover
rung nutzt, konnten an Ausmalungsbeispielen (Stiftskirche Königslutter1617und
Herrenhaus Altenkamp [fig. 1, 10. plate 251'7) entsprechende Verfahren zur
Phänomenerkennung zielgerichtet modifiziert und eingesetzt werden. Eine
Kombination aus analogen fotografischen Aufnahmeverfahren und digitalen
Bearbeitungstechniken sollte an diesen Beispielen eine Arbeitsmethodik
aufzeigen, die einerseits optimierte Daten für eine manuelle Kartierung liefert,
darüber hinaus jedoch die Bilddaten in einem Monitoringsystem verwaltet. So
können zum Beispiel die multispektralen Bildfolgen eines Motives durch
Mustererkennungsalgorithmen ausgewertet und nach Schadensbildern abgefragt
werden. Wurden an einer Wandmalerei zum Beispiel in der Vergangenheit
Ausmalungsbereiche nachbearbeitet oder restauriert, so kann eine komplexe
Analyse der Aufnahmen dies dokumentieren, obwohl für das unbewaffnete
menschliche Auge keine Veränderungen feststellbar sind. Eine a priori Analyse
von Schäden zur Vorbereitung restauratorischer Maßnahmen wird so ermöglicht.
16Vgl. Heckes et al. Anm. 13.

17Herrn Jürgen Heckes. Deutsches Bergbau-Museum/Forschungsbereich Informations
systeme. sowie Herrn Maro Moskopp. Luftbild Eifel seien an dieser Stelle für
mannigfachen fachlichen Rat und vielseitige Hilfestellung ganz herzlich gedankt.
Einleitung
"Unser so beschäftigtes Jahrhundert hat selten Zeit zum Lesen, immer aber Zeit
zum Sehen” (Theophile Gautier, 1858).
Eventuell beeinflußt durch diese Erkenntnis hat das technologisierte 20.
Jahrhundert zahlreiche ‘Seh-Hilfen’ in Form von fotografischen Abbildungs
techniken geschaffen. Eine gezielte Auswahl dieser Techniken, abgestimmt auf
die Möglichkeiten der digitalen Datenverarbeitung, führt in einigen Wissen
schaftsbereichen sogar dazu, daß Computer gut genug sehen können, um den
Menschen ablesbare Antworten auf spezielle Fragestellungen geben zu können.
Als ein Beispiel seien die Geowissenschaften zu nennen, die seit geraumer
Zeit multispektral zerlegte Bilddaten von Satellitensystemen zur Flächen
detektion, -Überwachung und -kartierung nutzen.'8 Speziell entwickelte Bildver-
arbeitungsalgorithemen können hierbei durch zielorientierte Kombinationen der
Eingabe-Bildddaten neue leicht interpretierbare Ergebnisse schaffen.181920
Motiviert durch diese positiven Ergebnisse wurden neben der konventionellen
Dokumentation der Wandmalereien am Beispiel des Entrées des Hauses Altenkamp
innovative Ansätze einer erweiterten fotografischen Abbildungsserie zur besseren
Unterstützung und Rationalisierung der restauratorischen Arbeiten verfolgt.
Erweiterte fotografische Dokumentation der Wandmalereimotive
Vor allen weiteren Maßnahmen wurden die Wandmalereimotive flächendeckend
mit einer Mittelformat-Fachkamera auf SchwarzAVeiß Film abgebildet. Gemäß
einer im Vorfeld erstellten Einteilungsmatrix wurden Ausschnittvergrößerungen
sämtlicher Bildszenen als Basis für eine manuelle Kartierung angefertigt (fig. 2.
plate 20).
Zusätzliche photogrammetrische Meßbilder und eine geringe Anzahl von
geodätisch bestimmten Meßpunkten stellten eine geometrische Grundlage für die
Herstellung von maßstäblichen digitalen Kartierungsunterlagen dar.2“
Mit dem Ziel einer annähernd wertneutralen fotografischen Farbaufzeichnung
wurden die Wandmalereien durch RGB-Farbauszüge (Rot, Grün. Blau) abgebildet.
Diese werden durch Absorptionsfilter realisiert, die nur den Spektralanteil ihrer
Eigenfarbe passieren lassen. Die Farbauszüge werden sequentiell auf Schwarz/
Weiß Film belichtet und können durch additive Farbmischung der Primärfarben zu
Reproduktionszwecken zu einem Farbbild zusammengesetzt werden (fig. 3, plate
21). Diese, gegenüber der herkömmlichen Mehrschichtenfarbfilm-Belichtung auf
wendige, jedoch seriösere Art der Farbaufzeichnung erfordert die Einhaltung und
Kontrolle verschiedener Aufnahmeparameter, die ständig eine Anpassung auf den
jeweiligen Stand der Entwicklung im fotografisch-technischen Bereich verlangen.
18Manual of Remote Sensing, I & 2 (American Society of Photogrammetry. 1983).

19M. Kappas. Fernerkundung nah gebracht (Bonn, 1994).
20Vgl. Heckes et al. Anm. 13.
Der Hauptvorteil dieses Verfahrens besteht jedoch darin, daß unkontrollierte

Farbvermischungseffekte, die bei der konventionellen Farbabbildung unvermeid
bar sind, nahezu ausgeschlossen werden können.21
Zur Normierung und Kontrolle der Aufnahmeserien wurde jeweils ein durch
spektrale Messung erfaßtes Testobjekt aus 4 Graufeldern in jede Aufnahme
integriert.
Für ausgewählte Testflächen mit typischen Schadensphänomenen wurden für
eine erweiterte Differenzierung der spektralen Eigenschaften der Wandmalerei
Zwischenkanäle im sichtbaren Bereich (Cyan und Orange) und in den angren
zenden nicht sichtbaren Bereichen (Ultraviolett und Infrarot) installiert (fig. 4).
Diese Zusatzkanäle sind eine wertvolle und unverzichtbare Hilfe zur Aufdeck
ung minimaler unterschiedlicher spektraler Reize, die jedoch den gleichen Farb
eindruck auslösen. Multitemporale normierte Multispektral-Aufnahmen könnten
so z.B. das Verblassen der Farbe oder Retuschen deutlich sichtbar machen.
Eine Wiederholung der Aufnahmeserien im Haus Altenkamp ist nach Ab
schluß der Restaurierungsarbeiten vorgesehen.
Digitale Dokumentation
Als Basis für eine digitale Befundkartierung wurden die fotografischen analogen
Abbildungen durch einen kontrollierten Scanvorgang gerastert. Der Feinheits
grad der Digitalisierung wurde so gewählt, daß Flächen mit einer Auflösung von
1 mm2wiedergegeben wurden. Zur Schaffung eines einheitlichen Bezugsystemes
wurden die Bilder eines spektralen Bandes digital entzerrt und zu einem Bildplan
montiert. Auf diesen Plan wurden anschließend die weiteren Bänder pixelgenau
eingepaßt.
Gleichzeitig bildet dieser Bildplan die Grundlage für die Einarbeitung der
manuellen Kartierung, die auf Kopien der Bildpläne durch Umfahren der
Schadensmerkmale mit Farbstiften durchgeführt wurde. Die digitalen Kartier
ungen wurden nach Regeln, wie sie aus Geographischen Informationssystemen22
bekannt sind, geometrisch und thematisch strukturiert und mit einem geeigneten
CAD-Programm digitalisiert (fig. 5). Sämtliche Elemente der manuellen
Kartierungsfolien wurden in Form von Flächenelementen übertragen und durch
Definition der Umgrenzungslinien durch geschlossene Polylinien gezeichnet.
Die visuelle Differenzierung fest eingrenzbarer und nicht eingrenzbarer Bereiche
erfolgte über unterschiedliche Linientypen. Die Visualisierung der Befunde
innerhalb einer Kartierungsthematik wurde durch variierende Schraffuren
differenziert. Innerhalb des digitalen Informationssystemes werden die Bereiche
durch eindeutige numerische Identifikationen verwaltet, die, gesteuert durch
21 H. Frieser. Photographische Informationsaufzeichnung (München. Wien, London,

New York, 1975).
22 R. Bill und D. Fritsch, Grundlagen der Geo-Informationssysteme (Heidelberg,
1994).
180 Rolk -Jürgen G rote and A nnette H ornschuch
Fig. 4. Haus Altenkamp, Entrée, Wandmalerei-Ausschnitt der Südwand, UV-

Fluoreszenz-Aufnahme. © Landesamt für Denkmalpflege, Hannover
Fig. 5. Das Bild entsteht durch Beleuchtung mit 320-380 nm und durch Filterung
mit 440-660 nm. Die Verwendung unterschiedlicher Farben bei früheren
Restaurierungen der linken Figur wird deutlich sichtbar. © Landesamt für
Denkmalpflege. Hannover
Fig. 6. Haus Altenkamp, Entrée, Bildplan der Südwand, überlagert mit Thermo-
bild. © Landesamt für Denkmalpflege, Hannover
Datenbankabfragen, numerische und statistische Auswertungen über das
Schadenspotenial durchführen können.
Für eine integrale Objekterfassung, die auch räumliche Komponenten wie
z.B. Raumklima berücksichtigt, können die Daten auch 3-dimensional bearbeitet
werden (fig. 6).
Neben der Übertragung der Ergebnisse der konventionellen manuellen
Kartierung vor Ort wurde unter Einbeziehung der digital aufbereiteten multi
spektralen Bildserien die Möglichkeit der stationären digitalen Kartierung—einer
Kartierung ohne ständigem Zugang zum Objekt, ausschließlich anhand der
Bilddaten—erprobt.
Zusammenfassend ist nach Abschluß der interdisziplinären Arbeiten fest
zuhalten:
1. Die multispektrale Aufnahmetechnik ermöglicht eine gute Visualisi
erung vorhandener Schadensbilder. Alle kartierten Schadensbilder
konnten in den Bildern lokalisiert werden. Eine Vielzahl von Schäden
kann einfacher, genauer und schneller in Bildern kartiert werden als
manuell vor Ort (figs. 7, 8. 9, plates 22, 23, 24).
2. Die manuelle Kartierung kann nur eine erste Übersicht der Schäden
darstellen. Eine Detailkartierung ist mit Bildverarbeitungstechniken
wirtschaftlicher zu lösen.
3. Eine langfristige, platzsparende Archivierung der digitalen Kartierung ist
nach heutigem Stand der Technik möglich. Der Zugriff und die
182 Rolf -Jürgen G rote and A nnette Hornschuch
Reproduzierbarkeit der Ergebnisse bleibt dabei stets gewährleistet.

4. Im Rahmen eines langfristigen Wartungskonzeptes können bei ziel
gerichteter Weiterentwicklung Kartierungen ohne die “Handschrift”
wechselnder Bearbeiter realisiert werden.
Computergestützte automatisierte Dokumentation
Als Fernziel der multispektralen Datenerfassung und Auswertung wurde
versucht, ein typisches Schadensbild durch automatisches Bildverstehen zu
erkennen. Die Bilder werden hierbei aufgrund ihrer Rasterung als eine Vielzahl
von Flächenelementen aufgefaßt, wobei partiell unterschiedliche physikalische
Eigenschaften des abgebildeten Objektes durch variierende Reflexionen in den
spektralen Bändern aufgeschlüsselt werden. Somit ergeben sich pro Flächen
einheit mehrere Meßwerte, die zu einem n-dimensionalen Punkt zusammen
gefaßt werden. Durch mathematische Ähnlichkeitsbetrachtungen der N-Tupel
können so Punktmengen (Cluster) gebildet werden.
Eine Definition der Auswerteschlüssel und ihrer mathematischen Definition,
einhergehend mit der Interpretation der maschinell erkannten Flächen, wird sich
nur durch enge interdisziplinäre Zusammenarbeit von restauratorisch, fototech
nisch und naturwissenschaftlich geschulten Fachleuten erfolgreich realisieren
lassen.
Obwohl der Aufwand zur Objektabbildung durch multispektrale Aufnahme
serien zunächst nicht unerheblich ist, kann diese Vorgehensweise thematische
Kartierungen ohne menschliches Eingreifen ermöglichen. Die Objekt
informationen müssen dazu in relevanten spektralen Bereichen optimiert
erschlossen werden, wobei die technische Realisierung der spektralen Bänder
stets innerhalb eines vertretbaren technischen Aufwandes bleiben muß. Ein
großer Vorteil der Objektdokumentation durch die erweiterten fotografischen
Aufnahmen besteht bereits heute schon darin, daß durch Einzelbetrachtung, aber
besonders auch durch Synergieeffekte bei der digitalen Kombination von
spektralen Bändern Schadensphänomene deutlicher sichtbar werden.
Somit kann die Multispektralauswertung eine Methode der Zustandserfassung
werden, die in Teilen stationär durchgeführt werden kann und Kosten, z.B. für
Gerüstzeiten, minimieren könnte.
Ein weiterer Vorteil besteht in der Möglichkeit der direkten digitalen Kartier
ung ohne den Umweg des ‘Abzeichnens’ der manuellen Kartierung.
Ein bereits bestehendes Schadensglossar wird zu einer Dokumentations-
Vereinheitlichung führen; es ermöglicht, das Schadenspotential eines Einzel
objektes—aber auch das von verschiedenen Objekten—im zeitlichen Verlauf zu
vergleichen.
3. E rgebnisse
Die Modellversuche erbrachten zahlreiche Anregungen, die als Bestandteile

EDV-gestützter Verfahren bereits im Vorfeld wichtiger Restaurierungen in die
Praxis überführt werden sollten, ln allen Projekten des NLD hat sich nach
drücklich gezeigt, daß vor jeder wissenschaftlich fundierten Untersuchung zu
nächst eine generalisierende Analyse des Baues und aller auf ihn einwirkenden,
meist sehr heterogenen Schadensfaktoren unerläßlich ist (“integrale Objekt
erfassung“). Erst nach dieser langfristig zu terminierenden Vorlaufphase ist es
sinnvoll, ein interdisziplinäres Untersuchungsteam gezielt an das Baudenkmal
heranzuführen und die erheblichen personellen, apparativen und finanziellen
Bindungen effektiv zu strukturieren.
Folgende wichtige Bearbeitungsschritte wurden inzwischen realisiert bzw.
befinden sich im Projektstadium:
1. Vollständige geometrisch und farblich richtige Abbildung der Wand
malereien und ihrer Schäden als Grundlage für die händische
interdisziplinäre Schadenskartierung sowie als Basis für die digitale
rechnergestützte Weiterbearbeitung zu einem entsprechenden Denkmal
infonnationssystem.
2. Das Denkmalinformationssystem ist modular aufgebaut: Es kann geo
metrische Informationen, Bilder und Textinformationen speichern und
verarbeiten. Es wird für die aktuelle Bestands- und Zustandserfassung
sowie für alle weiteren Bearbeitungsschritte dringend benötigt. Die
Einarbeitung und das Abrufen der Daten erfolgt rechnergestützt unter
Einsatz von layer-Bildung, ihre Ablage über Fenstertechnik in einer
relationalen Datenbank.
3. Die Kartierung von Zustandsmerkmalen am Objekt erfolgt nicht mehr
wie bisher von Hand (z.B. mit Buntstiften auf Transparentpapier über
einem Foto), sondern direkt am Rechner, wobei ein beliebig wählbarer
Bildausschnitt der Wandmalerei als Hintergrundinformation auf dem
Bildschirm erscheint. Vor diesem Hintergrundbild können dann direkt
auf dem Bildschirm mit einer entsprechenden Software bestimmte Bild
inhalte interpretiert und markiert werden. Über Legenden und speziell zu
erstellende Menues sollen die erfaßten Zustandsmerkmale mit Ken
nungen versehen und in Ebenentechnik (layers) rechnerintern abge
speichert werden. Auf diese Weise können in einem nachfolgenden
Schritt verschiedene Zustandsmerkmale miteinander korreliert werden.
Es werden damit Interpretationshilfen für eine interdisziplinäre Ur
sachenforschung der verschiedenen beteiligten Fachleute und erste
Hinweise auf Schadensursachen gegeben. Außerdem lassen sich die
rechnergestützt erfaßten Kartierungsdaten über Raumkoordinaten einem
3D-ModeI! zuordnen und in Verbindung mit einer Datenbank langfristig
184 Rolf -Jürgen G rote and A nnette Hornschuch
und mit schnellem Zugriff abspeichern.

4. Das bisher entwickelte Verfahren der Zustandskartierung und Doku
mentation, das modulartig aufgebaut ist und dessen Komponenten—dem
objektspezifischen Anforderungsprofil entsprechend—austauschbar bzw.
entbehrlich sind, kann zu einem integralen System der Objekterfassung
ausgebaut werden, um im Rahmen einer Erstdiagnose generalisierende
Analysen des Objektes mit allen auf dieses einwirkenden, meist sehr
heterogenen Merkmalen und Schadensfaktoren durchführen zu können.
Dazu sollen als weitere Module generalisierende Detektionsverfahren zur
Erfassung der flächenhaften Verteilung z. B. von Mauerfeuchte, Salzen,
Temperaturparametem usw. am Objekt und im Raum (Thermografie,
Elektrowiderstands- und Kapazitätsmessungen. Elektromagnetische Re
flexion (EMR), Radar- und Seismikverfahren usw.) eingesetzt werden, um
damit ggf. aktive und inaktive Schadenszonen herauszuarbeiten.
5. Die wichtigsten Parameter der gewonnenen Daten könnten in ein
vereinfachtes, über CAD erstelltes 3D-Gittermodell oder über ein durch
Bilddatenbearbeitung generiertes 3D-Modell überführt werden. Damit
lassen sich anschaulich Bereiche mit Schadenshäufungen relativ genau
in ihrer räumlichen Verteilung erkennen und auch ggf. ‘Altschäden'
(durch Akten-, Archivauswertungen, Altfotos) mit den aktuellen Be
funden korrelieren, ln Verbindung mit dem 3D-Modell können vor allem
bei komplexen mehrräumigen Bauwerken sämtliche zu einem Objekt
gehörende Informationen raumweise strukturiert ('Raumbuch’) abgelegt
werden.
6. Nach einer ersten Schadenserfassung (‘Phänomenkanierung’) und flächen
haft angelegten interdisziplinären Tastuntersuchungen sollen an schadens
relevanten Referenzflächen analytisch-statistische Interpretationshilfen zur
Ursachenforschung von Schäden (Faktoren- und Clusteranalysen11) ange
wendet werden. Sie sollen eine Trennung von dominierenden und
zufälligen Symptomen an potentiellen Schadenstellen erreichen und
damit eine Entwicklung von Strategien einleiten, außerdem die
Beobachtungen und Analysen zunächst auf ein nötiges Minimum
reduzieren. Somit wäre es möglich, den analytischen Aufwand vor Ort
erheblich zu straffen und trotzdem erste Hinweise auf Schadensursachen
zu geben, typische Symptomkombinationen/Interaktionsketten modellhaft
aufzuzeigen und damit das Erarbeiten von Interpretationshilfen für die
interdisziplinäre Ursachenforschung zu erleichtern.
7. Insgesamt dient die systematisch strukturierte Maßnahmenkette einer
fundierten Grundlagenermittlung und damit der effektiven Ausführungs-
planung und -Überwachung bis hin zur Qualitätskontrolle.
11Val. Glashoff Anm. 10.

4. Z usam m enfassung
Insgesamt bilden die zum Teil innovativen rechnergestützten Verfahren des NLD
der Anamnese und Schadenserstdiagnose sowie Dokumentation ein integrales
System, dessen Komponenten—dem objektspezifischen Anforderungsprofil
entsprechend—austauschbar bzw. entbehrlich sind. Der Schwerpunkt bei
wichtigen Kulturdenkmalen sollte jedoch auf einer ganzheitlichen Anwendung
liegen.
Selbstverständlich können die im Verlauf der Forschungsprojekte entwickelten
oder geplanten rechnergestützten Verfahren der Visualisierung und Dokumentation
zunächst nur vorläufige Interpretationshilfen geben. Sie sind als Service-
Leistungen aufzufassen, die einem interdisziplinären Untersuchungsteam den
Einstieg in vertiefende Zustands- und Schadenserfassungen erleichtern und erste
Hinweise für Konservierungs- und Restaurierungsmaßnahmen ermöglichen, ln
diesem Zusammenhang könnten die entsprechenden Verfahren der Schadens
erstdiagnose—in einer zweiten Bearbeitungsstufe vor allem auf objektrelevante
Referenzflächen konzentriert—zunehmend an Bedeutung gewinnen. Dabei muß
jedoch immer im Vordergrund stehen: mit möglichst geringem finanziellen und
dokumentarischen Aufwand komplexe Datenstrukturen schnell und präzise zu
erfassen, zu korrelieren, zu einem Therapie- und Wartungskonzept auszuwerten
sowie—unter dem Gesichtspunkt multitemporaler Reproduzierbarkeit für
qualitätssichernde Monitoringmaßnahmen—langfristig abzuspeichern. Dabei
sollten nach Möglichkeit selbstverständlich auch die bewährten ‘händischen"
Verfahren mit einer schrittweisen Überführung in rechnergestützte Hilfen der Be
standserfassung und -Sicherung zum Einsatz kommen. Daß eine ständige Pflege
und Aktualisierung dieser hochentwickelten Arbeitsinstrumente durch gründlich
ausgebildetes, nach mehrjähriger Erfahrung mit den spezifischen Erfordernissen
der Denkmalpflegepraxis vertrautes Fachpersonal—wie diplomierte Dokumen
täre—eigentlich ein zwingendes Erfordernis ist, sei an dieser Stelle als selbst
verständliche Grundvoraussetzung nachdrücklich betont.
Cutting off the Icing’s head:
images and the (dis)location o f power
F rank C o lson , J ean C o lso n , Ross P a rry ,

and A ndrew S awyer
Im a g in in g the P a st
oday, it seems unquestionable that the privileging of text within our
T histories has denied us certain stories and certain views of the past.
Allowing text to be the stuff of substance, and image merely that of
illustration, has impoverished our models of distant historical places and
moments. Culturally conditioned and technologically limited to building the past
from and with words—with all the structural constraints therein—we have been
limited (or so we now believe) as to the shapes and patterns we can conceive and
demonstrate. It is arguable that the complexities of power in the historical
contexts we are engaged in analysing cannot be fully grasped when restricted to
one medium.
This is a multi-authored essay representing the work of four individuals, of
different formations, but all engaged by the proximity of images and power.
Moreover, the work exemplified in the case studies discussed here could not have
been done without an active engagement with digital technology—they have been
enabled by the use of Kleio Image Analysis System, an Open Hypermedia
environment, and Computer Aided Design. But our engagement is led by our
theoretical and historiographical perspective; and whilst that perspective has been
broadened and developed by these technologies, it was not shaped by them.
Therefore, outside the footnotes, technology as such will not be mentioned again.
188 Colson. C olson. Parry , and Sawyer
Im a g in g P ow er
The emergence of the novel and puissant Dutch Republic from the medieval
Low Countries in the last quarter of the sixteenth century, stands as our first
example: the textualization of its political theory never achieved the impact of
Machiavelli, Bodin or Hobbes—indeed its most elegant theorist, Johannes
Althusius (1557-1638)—and the terms and concepts he used—is practically
forgotten.1*Yet the republic did see the outpouring of a large body of historical
and political prints or cartoons, some 2000 surviving for the period 1588-1632.'
Beyond the meticulous cataloguing of the early collectors these ‘diagrammatic’
prints have not been studied systematically in the way that, for instance, ‘textual’
pamphlets have been analysed.' Intimately connected, as they are, with the
political life of a novel and potent polity, they compose a significant source—a
pivotal source if we agree with Moxey’s argument that ‘the social values
expressed by the visual sign systems of these media cannot be distinguished
from some underlying reality that they are assumed to “reflect” ’, but that ‘to all
intents and purposes, these systems constituted the society whose values they
articulated’.4
We reach a similar conclusion when we turn our gaze to late nineteenth-
century Brazil—what many, indeed, would refer to as another great early
modern Republic. What we may—-cautiously—call ‘A total history of Brazil'
has not yet been written because (again) historians have restricted themselves to
1 Johannes Althusius, Politica methodice digesta (1603, enlarged version 1610 and
1614). A Westphalian, Althusius (1557-1638) was Syndic of Emden, a town not
technically part of the Republic, but garrisoned by the Dutch, to whom his work was
dedicated. The work is now most accessible to the English reader in abridged form in F.
S. Carney, The Politics of Johannes Althusius (London: Eyre and Spottiswoodc, 1965). In
his introduction to Carney's work, J. Friedrich described the Politica as ‘a culmination of
medieval social thought and a watershed of modern political ideas’ (p. xix). However,
many of Althusius' terms (such as ‘symbiosis’ and ‘consociato’) are not now part of the
political lexicon.
The period covered by our research into Dutch prints is set within the famous Eighty
Years' War of 1568-1648 in which the Dutch won their independence.
The catalogues of F. Muller, De Nederlandsche Geschiedenis in platen. Bereden-
eerde beschrijving van Nederlandsche historieplaten. zinneprenten en historische
haarten, 4 vols (first published Amsterdam, 1863-82: reprint Amsterdam: N. Israel,
1970); and G. van Rijn, Katalogus der historie- spot- en zinneprinten hetrekkelijk de
geschiedenis van Nederlanden. verzameld door Atlas van Stolk, 10 vols (Amsterdam,
1895-1933) include most of the prints of the period. Recently, Horst and Tanis in Images
of Discord (Grand Rapids: Eerdmans. 1993) have raised the profile of the prints.
4 K. Moxey, Peasants. Warriors and Wives (Chicago: University of Chicago Press,
1989), p. 131.
Culling o ff lhe king 's head 189
textual sources. Not merely in the sense that (as in many parts of early modem
Europe) to be literate was to be a tiny segment of society—adult, male, rich and
white, but rather that though prolific, laws, reportage and formal debate (i.e.
textual representations), simply codified one particular version of politics and
culture. The privileged—in Brazil at least—the public rather than the private
realm; the public events of legislative debate, the public war between stated
interest, and they set aside the intense and all-pervasive private violence of
privilege and deference—a significant strand in the configuration of power.
What is clear is that within both of these stories of ‘early modern’ polities, our
enduring proclivity to text has both limited our interpretation and shaped our
presentation of the past. Yet what is just as true is that those historiographical
communities that are bringing images in from the margins (literally and
metaphorically) are producing edifying, relevant, and credible new takes on the
past. Within, for example, the studies of early seventeenth-century England, the
formulaic portraits of Jacobean and Caroline culture are now emerging as
embodiments and acts—rather than just ornaments and depictions—of the
Jacobean and Caroline elites/ Similarly, the lavish scenery designs of the
opulent royal festivals of King James I of England are, today, approached as
components of—rather than simply backdrops to—the ideology of absolute
monarchy/ Likewise, due to the widening of our, as it were, evidential franchise,
the architectural façades and building styles of the English Renaissance stand,
today, less as mere stages for, and more as active engagements with, the political
fracture of the time.7 In fact, within this new multi-format—dare one say ‘multi
media’—historiography, to consider the temporary images of (for example) civic
pageantry more as technologies of power, and less as insubstantial ephemera,
seems not only possible but necessary/*6
For an accomplished study of one such portrait, see Ellen Chirelstein, ‘Lady
Elizabeth Pope: The Heraldic Body', in Renaissance Bodies: The Human Figure in
English Culture c. 1540-1660, ed. by Lucy Gent and Nigel Llewellyn (London:
Reaktion, 1990), pp. 36-59. The most sensitive and informed analysis of early Stuart
portraiture is John Peacock, ‘The Politics of Portraiture’, in Culture and Politics in Early
Smart England, ed. by Kevin Sharpe and Peter Lake (Basingstoke: Macmillan. 1994), pp.
199-229. For a more panoptic take on the subject, see David Howarth’s very accessible
Images of Rule: Art and Politics in the English Renaissance. 1485-1649 (Basingstoke:
Macmillan. 1997).
6In this respect the work of Stephen Orgel has been instructive. See his ‘The Royal
Theatre and the Role of King’, in Patronage in the Renaissance Court, ed. by Guy Fitch
Lyte! and Stephen Orgel (Princeton: Princeton University Press, 1981), pp. 261-73; as
well as his important earlier work. The Illusion of Power: Political Theatre in the English
Renaissance (Berkeley: University of California Press, 1975).
See Tim Mowl and Brian Earnshaw, Architecture Without Kings: The Rise of Puritan
Classicism Under Cromwell (Manchester: Manchester University Press, 1995).
g
For a close examination of the cultural influences within one such arch used in
190 C olson. Colson. Parry , and Sawyer
But what the recent historiography of early Stuart England also demonstrates
is that more than just increasing the quantity of meaningful evidence available to
us, images, in fact, provide levers and conduits, glimpses and echoes of
discourses that are simply not as accessible in our other documents. This
argument, about theory but developed from an engagement with the remaining
artefacts, or sources, of three different milieu, can be provided with a nexus by
considering one particular (though, admittedly, rather complex) discourse—the
discourse of ‘power’. With particular reference to those three ‘early modem’
polities already mentioned, we argue for the existence of a potent bond between
the nature of power and the nature of images. And that, perhaps, the story of
power is—occasionally—one best found within and told through images.
Our definition of ‘power’, here, is, of course, crucial. The ‘power’ (that we
are suggesting synchronizes so adroitly to the nature of images) is that as
described by Michel Foucault—specifically as laid out in his 1976 essay on the
subject. Echoing (if not a confessed subscriber to) the parameters of post
structuralism, Foucauldian power is net-like and dynamic, not hierarchical and
stable. Rather than centralized it is diffuse, shifting rather than static. Polysemie,
polyvalent, and ubiquitous, Foucauldian power is a constant interplay of inter
relationships, a continuing struggle, an-ongoing negotiation within which every
definition (by speech, by thought, by gesture, by text, and, yes, by image) is a
moment of—and cause for—the reconfiguration of power itself. In this concep
tualization, individuals do not ‘exercise’ power, but are themselves the points of
its manifestation. Foucault, however, appends a caveat: we are not conditioned
to think of power in this way. Modem political theory, he explains, has inherited
and bound itself to the ‘Leviathan’ of sovereignty—and a notion of an ordered,
concentric, and descending power, even if the actual figure or body of the
sovereign itself is no longer there. The Foucauldian challenge, therefore, is for
us to ‘Cut off the King’s Head’ within our conceptions of power—and,
consequently, unfix and de-gravitate that presumed ‘centre’."
James’ triumphal entry into London, see Gervase Hood. ‘A Netherlandic Triumphal Arch
for James i’, in Across the Narrow Seas: Studies in the History and Bibliography of
Britain and the Low Countries Presented to Anna E. C. Simoni, ed. by Susan Roach
(London: The British Library, 1991), pp. 67-82. For the text of the procession in which
the arches were used, see Thomas Dekker and Ben Jonson, The Magnificent
Entertainment Given to King James, Oueen Anne his wife, and Hemy Frederick the
prince, upon the day of His Majesty 's triumphant passage (from the Tower) through his
Honourable City (and Chamber) of London, being the 15. of March. 1603, as reproduced
in Jacobean Civic Pageants, ed. by Richard Dutton (Keele: Ryburn Publishing, 1995),
pp. 19-115.
’ ‘Truth and Power: Interview with Alessandro Fontano and Pasquale Pasquino’, trans,
by P. Patton and M. Morris, in M. Foucault, Power, Truth, Strategy (Sydney: Feral
Publications, 1979), pp. 29-48 (p. 38). For Foucault’s thinking on power in the context
Culling o ff lhe king 's heati 191
And this is our point: the structure of meaning (the discourse) created by this
act of theoretical regicide creates a system of meaning web-like, topographic,
where (freed from the chain of hierarchy) anything can connect to a plethora of
others. In other words, Foucault’s notions of power—notions that resist
hierarchical, centralized systems of authority—would seem to be more readily
and more substantively visualized, and more intuitively understood in and
through images. Or (to state the corollary), text is a system whereby symbols are
arranged linearly and sequentially within a (consequent) hierarchical structure;
i.e., the very antithesis of Foucauldian power. Consequently, the frame of
reference/referents which represents and demonstrates the complex
interconnecticity—the topography—of modem theories of power, might be, we
suppose, the image. (Text is but a sub-set of Image, the latter explicitly
promoting and encouraging more than one view.) Or put another way: though
we understand text is in itself ‘power’, we suspect ‘power’ may be diagrammed
more readily and more effectively than it can be textualized.
D iagram m ing the D utch R epublic
Our own observations tell us that practitioners of power politics are aware of the
significance of gesture, of ‘power dressing’, of all that is encompassed by the
word image. Similarly, in the past, power was often manifested through media
other than the word. Moreover, some forms of power-—physical forms—simply
cannot be described with words (one cannot describe the electrical system that
transfers power around a car or a building: one diagrams it). And there are
certainly some historical cultures that seem to expound this. For instance, John
Montias has demonstrated that the Dutch Republic was saturated with images.1"
That such an image-conscious society included a vigorous brand of political
imagery (closely connected with its rejection of Habsburg rule and its Republic’s
burgeoning power) may not be unconnected. The Republic was wired for a
discourse of diagrammed power in the way that today’s information societies are
wired for Information Technology.*10
of a wider debate, see S. Lukes, Power (Oxford: Basil Blackwell, 1986). and for a context
admirably free of postmodern jargon, see S. R. Clegg. Frameworks of Power (London:
Newbury Park; New Delhi: Sage, 1989). Anybody who doubts that our conceptions of
power are shaped by this ‘sovereign’ construct might take up Microsoft’s PowerPoint
software, set to diagram an organization; they will find it practically impossible to create
anything but ‘top down' structures, presumably because those who specified the software
could conceive of nothing else.
10J. M. Montias' major work was Artists and Artisans in Delft (Princeton: Princeton
University Press, 1982), which demonstrated the quantities of paintings and prints, which
could be found in almost all homes in Delft in the early seventeenth century.
192 C olson. C olson. Parry , and Sawyer
Fig. 1. 'Penning' dated 1588: from H. E. Greve. De Tijd van den Tachtig-
jarigen Oorlog in Beeid (Amsterdam: Uitgevers-Maatschappy "Elsevier". 1908),
p. 67
The Dutch, who had thrown out the man who would be king in the Abjuration
of 1581 (their own, as it were, Foucauldian moment in the redefinition of
power), established a polity defined by constant negotiation: a definition far
closer to Foucault’s concept than that defined by a trajectory that includes
Bodin, Hobbes, and sovereign power." And it is perhaps significant that, in the
work of Johannes Althusius, we find analyses of the polity carried out in an
ascending manner—as Foucault argued we should. This form-giving power of
this dynamic, imbalanced state could be comprehended more clearly if we could
graphically contrast a ‘sovereign’ view of the Dutch situation with a domestic
counterpart. Fortunately, due to the involvement of the English in the developing
maelstrom in the Low Countries, we have English observations from the period.
A pictorial example may be seen in a coin produced by the partisans of the
English Earl of Leicester, sent to aid the Dutch in 1585 (fig. 1); we see their
situation depicted sovereignly, with Elizabeth of England, Leicester, an
apocalyptic beast, and the provinces kneeling at the monarch’s feet in
supplication. The Latin text translates: ‘To the best and greatest God be praise
and honour forever, because . . .’ and then the text ends, the rest (i.e., the reason
for the praise) only being expressible in the image within.12 In short, here it is
" The classic work is F. H. Hinsley, Sovereignty, 2nd edn (Cambridge: Cambridge
University Press, 1986). He traces the trajectory Bodin-Hcnry VIII-Hobbes. and regards
sovereignty as he defined it as the pinnacle of political theory.
: The reverse shows clerical figures, seven in all, who appear to be falling. They can
be distinguished by their dress as. at the lowest point, the pope (losing his grip on a
Culling o ff the king 's head 193
Fig. 2. C. J. Visscher, ‘AJbeelding van 7 Nederlandts Bestandt’ [Image o f the

Nethedands Truce], c. 1609. FMI267. Reproduced by permission o f the
only an image that can express the coin’s message. And that message can be
simply diagrammed or described in hierarchical terms.
In contrast to this hierarchic image, far more complex ‘diagrams’ flourished
among the political prints of the Republic. An example can be seen in a
depiction of the truce negotiations of the early 1600s, Muller’s print number
SI267c (fig. 2).13 ‘Reading’ it primarily as a political diagram, we see, from the
left, military forces in confusion at the end of the fighting, a great carriage
representing the truce, aided by the monarchs of England and France, led by
allegorical figures and the Archdukes, and carrying Peace and other virtuous
characters. It is steered by Johan Ney, a Jesuit and the Habsburg negotiator of
the Truce. It heads towards a sunlit region—indicated stage right, as it were,
rather than illustrated—between two booths, representing the Independent
Netherlands and the Habsburg Netherlands. Clearly much could be said about
such an image. The point here, however, is that dynamic, shifting patterns and
monstrance and a chalice), followed by a group consisting of a cardinal, two monks, two
bishops, and possibly a Jesuit. Above, the Tetragrammaton is shown in a cloud from
which lightning strikes the clerics. The text reads QVEM DF.VS CONFICIET SPIR1TV
ORIS SVI. ‘Whom the Lord shall consume with the spirit of his mouth’, n Thess. 2. 8
The coin is attributed to the Earl of Leicester, or rather his partisans in the Netherlands, in
major coin catalogues (e.g., E. Hawkins, A. W. Franks, H. A. Grueber, Medaille
Illustrations of the History of Great Britain and Ireland to the Death of George It, voi. 1
(London: British Museum, 1885), p. 139, their no. 99).
' AJbeelding van 7 Nederlandts BestandI [Image of the Netherlands Truce], c. 1609,
C. J. Visscher. FM1267. An allegorical print about the Twelve Years of Truce 1609-21.
194 C olson. C olson. Parry , and Sawyer
Fig. 3. Detail, ‘Afheelding van 't Nederlandts Bestand/’ [Image o f the

Netherlands Truce], c. 1609. FMI267. Reproduced by permission o f the
networks of power are diagrammed. For example, in one detail (fig. 3) a cameo
represents the Independent Netherlands. The intense debate within the Republic
occasioned by the negotiations is graphically projected: the Netherlands maid
gestures to Maurits, the Dutch commander on her left; to her right, the gestures
and glances of the figures representing the States General are, likewise, there to
be noted. This scene stands as a microcosm of the entire image, alive with
debate, negotiation and movement. Here—in interplay, interconnecity, diffu
sion—is Foucault’s concept of power diagrammed.
Moreover, if we take each ‘element’ of Leicester’s image—crown, throne,
monarch, and kneeling figures, naked in supplication—and analyse a sample of
Culling off the king 's head 195
Dutch imagery for the same elements,14 it can be demonstrated that such
elements were mostly portrayed in a negative fashion, rarely neutrally, and only
on some occasions with any positive connotations. Thus the discourse projected
by Leicester was, it seems, quite inappropriate to his circumstances as Governor
General in the Netherlands from 1585 to 1588. It is in imagery, more than in
major canonical texts, that the constant negotiation implicit in the ‘un-central-
ized" and disaggregated manifestation of power that structured the nascent Dutch
State is most strikingly obvious.
P ictu rin g B ra zil ’s R epública
However it is not just the non-sovereign inter-relations of power of the Dutch

Republic that seem to sharpen when viewed pictorially. For when we turn to our
other Republican example, we likewise find a discourse of power that was—and
needs to be—recalled through images. For instance: it is hardly surprising that
scholars attempting to examine the extraordinary events of the period 1868—
1910 (generally regarded as the founding years of contemporary Brazil) should
have wrestled with the fact that though slavery ended and the Republic emerged,
the structure of social relationships inherited from the colonial era was retained.
Brazil has continued as a society characterized by social apartheid. In 1883 a
political slogan proclaimed ‘Brazil is coffee, coffee is the “Negro”, the “Negro”
is Brazil’. But by 1900 the ‘Negro’ was forgotten, coffee was hegemonic: the
‘Negro’ was emphatically not Brazil? In fact, South America’s premier power,
centre of one of the world’s most lucrative trades, a continental polity, entered
into a crippled existence in the twentieth century—one from which it has not
recovered. Hitherto the reasons for this have remained obscure. Until, we think
about the nature of power, because scholars have failed to take on board the
dynamic, net-like, diffuse nature of power in this continental polity. Indeed,
however sophisticated such statements as ‘the structure collapsed in May 1888’
(when the Abolition Law was approved by the Legislature)IS and ‘revolution of
Abolition’ might appear at first glance, they have become simple tributes to the
14 Kleio IAS enables each element of each image to be registered for searching and
data retrieval, down to a very detailed level if required, by the use of textual image
descriptions linked to digital representations of the prints themselves. Thus, for example,
if during research the gesture of kneeling is found to be of significance, all the scenes
showing such a pose or gesture can be called, viewed, and referenced.
I! Seymour Drescher, ‘Brazilian Abolition in Comparative Perspective’, Hispanic
American Historical Review (August, 1988), 429-60 (p. 459). Drescher develops the
useful notion of an hierarchical regime of notables (p. 458), ‘demoralized’ by the mass
mobilization of the crisis of 1886-88. He notes that the transition from slavery has barely
been discussed.
tactical ability of oligarchs."' They do not examine the threat to private absolute
authority explicit during the final years of the struggle to end slavery. Because—
and this is the crucial point—they fail to confront the fact that power in Brazil
remained neither hierarchical nor stable. Such approaches, therefore, do not
enable us easily to understand either the fragility of the state in twentieth-century
Brazil nor the continuation of its colonial heritage as social apartheid. In short,
the intrinsic poverty of textual sources that deal with the public rather than the
private limits their explanatory value in explaining change in a dynamic
structure, because they do not enable scholars to cross the door from public to
private. Scholars of Brazil have been aware of this, and the last decade has seen
increasing recourse to the work of literary critics’ expressions of ‘misplaced’
[fora de lugar]'*1and to the anthropologist’s sense of the public place as incessant
war.1* But, exciting as these are as idealizations, they have only the most limited
power as explanation, because they operate at the wrong remove. Though
contextualizations, they do not enable us to describe in the most approximate
fashion the extraordinary impact of 1887-89 on the future course of Brazilian
history. Such models do not, for example, explain the extraordinary sense of
alienation felt by army officers against the rule of merchants and lawyers—
‘frock coats’—nor the extraordinary influence of foreign ideas—casting a long
shadow over the twentieth century. They allude to but do not elucidate the
“ Jacob Gorender’s A escravidão reabilitada (São Paulo: Ática, 1990) discusses the
recent historiography of 1887-88 as that of an ‘abolitionist revolution’, pp. 133-38.
Gorender correctly identifies the role of slaveocrat leaders in a transition which would be
‘slow, gradual and secure', p. 146, and notes scholarly attention given to the perception
of the ‘faithful slave’ during the crisis of 1887-88; Lilia Schwarcz, Retrato em branco e
negro. Jornais, escravos e cidadãos em São Paulo no final do século XIX (São Paulo:
Companhia das Letras, 1987).
1 Roberto Schwarz, Misplaced Ideas. Essays on Brazilian: Culture (London: Verso,
1992), especially ‘Misplaced Ideas. Literature and Society in Late Nineteenth Century
Brazil’, pp. 19-32 (original title: ‘As idéias fora de lugar’ [misplaced] Estudos Cebra
(1973), 3).
'* Roberto A. da Matta, A Casa <£ a Rua. Espaço, cidadania, mulher e morte no Brasil
(São Paolo: Editora Basiliense, 1985), especially pp. 71-102. Da Matta’s House [Casa]
and street [Rua] are basic dynamic opposites within the web of relationships described by
Foucault. The Brazilian anthropologist’s essay, ‘For an Anthropology of the Brazilian
Tradition’ or ‘A Virtude esta no Meio’, in The Brazilian Puzzle. Culture on the
Borderlands of the Western World, ed. by David J. Hess and Roberto A. da Matta (New
York: Columbia University Press, 1995), pp. 270-93, argues (p. 276) that ‘we will not
advance towards significant understanding of Brazilian and Latin American reality if we
do not discover the deep relations between the impersonal commands of law (conceived
as a function of ‘individuals’) and 'friends' (a universe governed by the implicit and
personalized rules of parentela (the extended family structure on which social influence is
based)).
extraordinary gulf between the tenacious belief in modernity that is the public
doctrine of state organizations and the brute power of private privilege that
remains untouched by law. They take little account of the way in which men
perceived that in 1887-88 the war between private privilege and general security
that had surfaced in the 1830s had suddenly re-surfaced and the savage nature of
private power again lay exposed. Most of all, they do not address the central
debate of the time: the ownership of progress and science—the pervasive
heritage of the French Revolution: and the acute and specific threat of
‘degeneration’ that haunted the political elite of the age. Using text we might as
historians describe the words of the proclamation of abolition as a ‘tersely
worded death warrant for a collapsing structure', but we cannot easily
understand its aftermath—text gives rise to the sense of political relationships
which remain ill defined, hence ‘misplaced’.
Therefore, in order to understand this sense of the ‘misplaced’, other forms of
expression have to be brought to the argument,19 for example images—in this
case, specifically mass-produced cartoon images. For though it was only the poet
and the novelist who could venture beyond the public to the livid intimacies of
private power, it was only the cartoonist who had the medium through which to
represent these complex discourses.20 Acknowledged fictions have their uses, but
even literary giants such as Machado de Assis, writer, newspaper columnist, and
in 1888 a well-placed clerk in the Ministry of Agriculture could only obliquely
refer to the ‘absolute oligarchy’ in a minor publication—their novels amply
describing the effect of its existence. Nevertheless his column in the leading
newspaper, the Gaieta provides the vital bridge, a sense of the ‘vie quotidienne’,
which allows the historian to move towards a fuller explanation, one which
includes the argument of images and their role in explaining the political
Pereira Barreto: ‘On the one side are the abolitionists, riding on a sentimental
rhetoric and armed with a revolutionary metaphysics, on the other side are the
landowners, silent and humiliated, in the attitude of those who meditate upon their guilt
or meditate an impossible revenge’ (quoted from Paula Beiguelman, Teoria Aedo no
Pensamento Abolicionista, vol. i of Formação Politica do Brasil [São Paulo: Livraria
Pioneira, 1967], p. 159).
;o Not simply because cartoons were so diffused, but because images play such a large
role in the discussion of the complex grammar of encompassment and passages between
‘this world’ and the ‘other world’ [da Matta, p. 276]. The open hypermedia environment,
Microcosm, allows individual links and associations to be catalogued and classified
according to the various perceptions of contemporary viewers and latter-day scholars
alike. The flexibility of the environment enables the identification of critical components
of the various images and their referents, limiting allusion and inference while clarifying
arguments. Frank Colson. ‘Case Study C. Hypermedia Database Management Systems:
Microcosm as a Research Tool’, in Databases in Historical Research, ed. by Charles
Harvey and Jon Press (London: Macmillan, 1996), pp. 69-71.
198 Colson. C olson. Parry , and Savvylr
l.,.< !
Fig. 4. 'Livre’, Revista Il·lustrada, Rio de Janeiro, 14 October 1884

topography of everyday life.21
The day-to-day chronicle of official political life in late nineteenth-century
Brazil was, by early modem standards, extraordinarily amply documented. Some
thirty newspapers and periodicals followed and created daily life in Rio de
Janeiro, capital of the Empire. Sheer plethora is daunting enough, and the study
of the page is rewarding. It is however crucial to note that, thanks to French
influence and the advent of a market among the urban elites, the period
witnessed the remarkable flowering of the political cartoon. Systematic
cataloguing of the cartoons for Rio at least is complete and some editions of the
long run of the Brasil I¡lustrada, edited by Angelo Agostini, have been
reproduced in facsimile. A prolific caricaturist, Agostini’s understanding of the
intricate topography of power has meant that his work has long been used
by scholars to enliven their accounts. Evoking currents of political life, the
caricatures are brilliant, strident, yet building a subtle vocabulary of
power-—typified by the cartoon ‘free’ [/;'v/-e] published in 1883
■' Arguably the greatest of all Portuguese-language novelists, author of nine novels
and more than two hundred short stories (1839-1908). For his political allusion, see
Machado de Assis, Bons Dias! Crónicas (1888-89), edição, introdução e notas de John
Gledson (São Paulo: Editora Hucitec, 1990). pp. 13-14.
Culling off' lhe king 's head 199
Fig. 5. ’Não vos aproximeis de mim! Fossas mãos ainda tintas de sangue dos
escravos manchariam as minhas vistas! Retirae-vos, tu não vos quero. ’ Revista
Illustrada 15 June 1888 [original orthography]
(fig. 4).22 Yet they have not been ‘read’—perhaps because they allude to that
constant negotiation that remained a hallmark of political life under Empire and
(after 1889) the Republic or, more prosaically, because historians have found it
difficult to collate the visual clues projected by these images. The sense of
concealed brutalities that the limits of such negotiation between had been
reached in 1888 is nowhere more powerfully expressed than in another cartoon
‘The Republic courted by the ex-proprietors of slaves’ published only a few days
after the abolition act had been signed.23 The fact that the cartoons would be
~ ‘Livre’—depicting a jangada [raft] commonly used as a lighter to ship goods over

the reefs of Ceará to ocean-going steamships. In 1884 the jangadeiros [lightermen]
refused to transport slaves for sale in neighbouring Maranhão or the Southeast. Agostini
depicted their leader Nascimento steering a raft carrying a slave family and the figure of
‘liberty’. The depiction of the family, and the figure of ‘liberty’ contrast richly with
stereotypical views. From R. Maghalães Jr„ Deodoro. A Espada contra o Império (São
Paolo: Companhia Editora nacional, 1957), vol. t, facing p. 192.
■' The radical press, led by the Cidade do Rio on 25 May wrote that ‘we have notices
from the interior that various ex-slave owners have refused to obey the law and have
informed their captives that they can only leave their lands or receive a salary once they
have received a letter of manumission from the Regent'. The cartoon was criticized in the
press on 18 June. It has been reproduced in Maghalães Jr (as in note 22), p. 348.
posted up in country Venda and city cafes, corner kiosks, or coffee-houses

(available to the millions of illiterates) determined that they would inhabit a
world of invocations and votives. Indeed, the appeal to ‘modernity’ in its French
allusions, by 1888 well known to viewers, makes this image a vital evocation of
the crisis which broke as slavery collapsed.
This image is part of an historical moment when sovereign power was
disaggregated—when the link between the crown (supreme mediator) and its
natural supporters was fractured. It depicts a procession of ex-slave-owners
dressed in traditional country garb (arguably similar to slave-catchers) following
a tattered banner reading ‘Down with the Abolitionist Monarchy—Long Live the
Republic with Compensation’ (fig. 5). Here República unlcaves the ruptured
discourse of power (or, as one radical put it, the bestialization of the people)
which characterized 1888-89. Published within weeks of the proclamation of
abolition on 13 May, and much discussed in the Press (for Agostini had long
campaigned in the abolitionist cause), the cartoon works at many levels.
República epitomizes the civic virtues of a Robespierre, her robes, the cap and
the tricolour rosette, and the stance, denoting purity, right, and reason, liberty
and progress. With her foot forward on the march of Progress we are asked to
wonder whether she will allow her purity to be stained by those who pursue her
to other more bestial ends. Her protests are vehement and proud: ‘Do not come
near to me! Your hands are still red with the blood of slaves. You will stain my
robes. Leave me! I do not want you!’ But the ex-slaveowners remain insistent.
They will choose ‘revolution’ (even the French Revolution’s tricolour) rather
than face the hierarchical polity. For the planters seek to expropriate Progress
and divorce it from liberty. Their campaign flag demands 'compensation' for the
slaves they have lately owned and dressed. Their eyes are avid, their intention
evil. The days when the cartoon was being sold in Rio were ones in which the
Press was dragging example after example of planters refusing to release their
slaves, imprisoning and whipping them, to the attention of the authorities, who
often looked the other way. The depiction of ex-slaveowners as harbingers of
unreason and bent on the prostitution if not subversion of the Republic was a
powerful one. Fear and avidity characterize the image of the planters, as they
face a world in which authority does not appear to be theirs alone. For months,
abolitionist agitation had begun to isolate planters from traditional support
among the free poor and local propertied, as well as administrative organiza
tions, which they had long controlled. By late 1887 slaves were not simply
resisting planters, but also were running off the plantations and receiving support
from clandestine and increasingly vociferous abolitionists. As early as Novem
ber 1886 the Imperial government had refused to arm local militia (the National
Guard) to prevent the militarization of the plantations. By October 1887 units of
the army refused to return slaves to their owners. The Church, long a supporter
of the status quo, finally abandoned the planters in late 1887, leaving their
authority to rest on the degree to which they could informally organize local
Cutting ojf the king 's head 201
violence. Their private ‘rights’ to govern their Ashantis with complete impunity
seemed in 1887-88 to have been usurped, by the government they expected to
own.
Their desperation and pursuit of unreason so evident in the piercing and
distraught impression of the ‘archetypical’ planters is quite remarkably docu
mented. Since the abolition of the lash in late 1886 they had petitioned the
Crown for guarantees of their property, for protection against a population of
vagrants.24 In May and June 1888 the abolitionist Press reported numerous
threats made by ex-proprietors to the freedmen on their plantations. Yet that
unreason is evidenced not merely by the visual contrast to the secular sanctity of
República but also by the picture of workers (blacks and whites alike) gathering
in a coffee crop widely documented to be one of the largest on record. The
bemused expression on the face of the overseer indicated that the procession of
planters is not really concerned with the crop, still less with slaves and their
families, but is concerned with the business of recovering the web of
connections and patronage which were, as far as the planters were concerned, the
essence of the state. The demand for immediate compensation is not merely a
literal assault on the Republic of reason and purity but also a reassertion that
Progress shall be impelled by savage dominion rather than law.
In effect, what these images capture and diagram is the extent of the social
rupture that emerged in 1887-88, when a Crown, ruling by hereditary right and
through established laws, clashed with provinces and their historic liberties, as
well as proprietors who regarded themselves as having absolute dominion over
their lands. Scholars have noted that the ‘absolute oligarchy’ immediately
emerged as ally of the Republic, and many ex-slaveowners profited from the
years of high prices commanded by coffee. But most have failed to appreciate
the extent to which many felt that the mobilization of the free population in
support of abolition that occurred in 1887 presaged a loss of control over their
private local societies. The motifs of freedom characterized by the stance of the
liberated slaves in Ceará in 1884 were widely diffused. By mid-1888 few would
disagree that abolition had been won by the slaves themselves, with explicit
support from several different segments of society and the ‘golden law’ of 13
May 1888 was (like so many previous reforms) a simple act o f ‘public security’.
After the trauma of March-May 1888, planters were desperate to reconquer
local hegemonies—hence the apparently irrational flight to the Federal Republic.
The irony, therefore, that a regime of secular purity and order would protect the
privacy of the lash was not lost to a contemporary reader of this and other
images. Similarly, the depiction of the planters in the procession in traditional
24 'Fuga de escravos em Campinas discursos pronunciados no senado pelo Exmo Sr.

Barão de Cotegipe (Presidente do Conselho e Ministro de Estrangeiros) (Rio de Janeiro:
Imprensa Nacional, 1887), pp. 16-19. The speeches of the Council President addressed
the issues raised by these petitions, sent to Parliament from May 1887, to abolition.
202 Colson, Colson. Parry , and Sawyer
Fig. 6. Inigo Jones. 'Pallas o f Perfection Devonshire Collection. Chatsworth.

Reproduced by permission o f the Duke o f Devonshire and the Chatsworth
Settlement Trustees. © Photographic Survey, Courtauld Institute o f Art. London
rough country garb (compared, at least, to the overseer), is no casual act—by
early 1888 some of the planters who controlled the largest and most
industrialized establishments already had freed their slaves. Angelo Agostini’s
layered image of June 1888 has, therefore, captured the moment of disruption
which brought about the invocation of Order and the Republic by those whose
legitimacy was that of custom and brute force. It points to forms of power which
most regimes—including the Dutch Republic—have been unable to do little

more than mitigate.
M apping E nglish Sovereignty
It was a similar struggle over (and appropriation of) the agents of ‘progress’ that
characterizes our third and final example—the illusionary images of the English
court masque. The court masques were exclusive royal entertainment performed
before the king, his family, and his court as well as any (invited) visiting
dignitaries. Though Elizabeth I (in the sixteenth century) and Lord Protector
Oliver Cromwell (in the middle of the seventeenth century) enjoyed similar sorts
of festivals, the English masque was almost entirely confined to a short history
from 1605 to 1640. In content they were a combination of music, poetry, visual
effect, dance, and song; ‘pictures with light and motion’ claimed the polymath
Inigo Jones, designer of most of the shows.:s Usually taking place on the evening
of Twelfth Night or at Shrovetide (or, occasionally, at times of court wedding),
these were lavish, conspicuously expensive events in the royal calendar, taking
several hours to perform (sometimes, evidently, to the tedium of king and
courtiers alike)*3" and, more times than not, staged in the royal Banqueting House
at Whitehall.37 The masques were a complex blend of indulgent frivolity and
' Jones offered this description in the text of the 1632 masque Tempe Restored (lines
49 50); reproduced in Stephen Orgel and Roy Strong, Inigo Jones: The Theatre of the
Stuart Court, 2 vols (London: Sotheby Parke Bernet; and Berkeley: University of
California Press, 1973), vol. il, pp. 479-504.
" 'Why don't they dance?’ the king was reported to have called out at one eventful
masque in 1618. ‘What did they bring me here for?’ See Orazio Busino’s full account in
Calendar of State Papers: Venetian, 1617-19, 24 January 1618,'Anglipotrida’, pp. MO-
14. The original MS is in the archive of St Mark’s, Venice (CLV11, cod. MCXX11. fols
72-75).
3 In contrast to that of its successor (still standing on Whitehall today), there is little
extant evidence relating to the first Jacobean Banqueting House. A sketch by Elizabethan
architect Robert Smythson from c. 1609 (British Architectural Library, RIBA, London)
provides an impression of the groundplot of the site. Similarly, Andrew Kerwyn’s
accounts relating to royal works at the time of Banqueting House’s construction (a
protracted process from 1604 to 1610) proffers valuable clues as to the building’s internal
and external design; see ‘Declared Accounts of the Pipe and Audit Office, Works and
Buildings’, Public Record Office, London, E 351 / 3240-43. Consequently, studies of the
space have been few and far between. See: Per Palme, The Triumph of Peace: A Study of
the Whitehall Banqueting House (London: Thames and Hudson. 1957). pp. 1-5 and 115-
18; Dramatic Records in the Declared Accounts of the Office of Works 1560-1640, The
Malone Society Collections, voi. 10, cd. by R. F. Hill and F. P. Wilson (London: Malone
Society. 1977), pp. xvi-xvii; and most recently—and most usefully— The Histoty of the
204 Colson. Colson. Parry , and Sawyer
state solemnity, sober didactics, and cutting satire. And, invariably, they turned
upon an epiphanic moment where discord and antic movement were transformed
into political order, serene spiritual and intellectual harmony. Throughout, the
masquing space served (in a metonymic rather than metaphoric way) as a
microcosm of the kingdom—with the sovereign central, pivotal, and privileged.
An image that shows well the stylized illusionary vista that was presented
before the king is Jones’s preparatory sketch (fig. 6) for a scenery design from c.
1618-19. The drawing, made in pen and brown ink washed with warm grey, was
used, most likely, for the performance of George Chapman’s Masque o f the
Twelve Months, the last festival, in fact, to be performed in the Banqueting
House before an accidental fire destroyed it in January 1619.“ Reporting on the
entertainment, a contemporaneous diplomatic despatch gives us an indication of
how this two-dimensional image was realized—in performance—as three-
dimensional scenery:
. . . at the foot of the room opposite his Majesty a curtain which hid all the wall at
that end was let drop, revealing a perspective with very lovely ornaments which
stood in the air between the ceiling of the room and the solarium. In it were seated
all the lords of the masque, in the most beautiful order. By means of a hidden
device it descended very, very gently down to the ground. Behind this perspec
tive, in proportion as the lords’ seat descended, and at their backs, there were seen
in another perspective castles, towers, palaces, rooms and pictures in fore
shortening. In truth it seemed to me that 1 had never seen anything that gave more
cause for wonder.. . . ”
The account makes clear that this image—like so many of Jones’s masque
designs before and after it—exploited the visual and spatial implications of
King's Works , ed. by H. M. Colvin (London: Her Majesty’s Stationery Office, 1982)
iv:ii, pp. 322-24.
:x A persuasive case was made for the connection between this drawing and
Chapman's masque in a paper by Martin Butler entitled 'A re-identified Chapman
masque and the Jacobean response to European crisis’, delivered at Texts and Cultural
Change: History. Politics and Interpretation. 1520-1660. The Reading Literature and
History Conference, 16-19 July 1995. However, David Lindley, in his ‘Select
Chronology of Stuart Masques’, makes no mention of the masque. Instead, the masque
for 1619 is cited as Thomas Middleton, The Masque of Heroes: see his Court Masques
(Oxford and New York: Oxford University Press, 1995), pp. xxvii-xxviii. Similarly,
Stephen Orgel and Roy Strong's ‘Calendar' in their Inigo Jones (as in note 25), voi. 1, pp.
85-87. makes no reference to Middleton or to Chapman in respect to a 1618/9 masque.
Their only relevant entry is for a ‘Lost Masque for Prince Charles' for 1619.
G. B. Gabalence’s diplomatic dispatch of 12 February 1619 reproduced in J. Orrell.
'The London Court Stage in the Savoy Correspondence', Theatre Research International.
4(1978-79). 84-85.
linear perspective to fix the king as primal spectator, as ‘Britain’s glorious eye’."
The geometrical construct of linear perspective, of course, privileges one eye in
space—especially when that perspectival effect is the receding flats of a
theatrical scene. The prospect of a street scene in perspective, for example, such
as that which opened Ben Jonson’s masque The Vision o f Delight in 1617, was
only aligned to the king’s point of view. In contrast, a spectator positioned up in
the balcony (see fig. 7, plate 26), askew to the stage, would have been presented
with a very different (awry) arrangement of scenic objects—intentionally,
explicitly, didactically.
But as well as underscoring the notion of the English king’s unique acuity,"
Jones’s image also connected to discourses of stoic learning and civic instruc
tion, of'royal education and right’.” For this image of the ‘Pallas of Perfection’
(Jones usefully labelling his drawing on the far left of the paper) can also be read
as a visual mnemonic of state-sanctioned discourse. Indeed, by 1618 it was
something of a convention to present before the king and his court abstract
themes and values rendered into (seemingly) substantive architectural form.
From the ‘Temple of Peace’ (in 1604) to ‘The Throne of Beauty’ (in 1605) and
from the ‘House of Fame’ (in 1609) to the 'House of Chivalry’ (in 1610) and the
‘House of Honour’ (in 1614), the scenery of the English court masque
repeatedly used images of buildings as analogues of complex thought. Drawing
upon the medieval and early modern traditions of artificial memory, as well as
the practice of contemplating and presenting conceptual space as physical
space,'*5 the ‘Pallas of Perfection’ brought (to use the words of one Elizabethan
511Thomas Campion, Lord Hay's Masque ( 1607), line 374. as reproduced in Orgel and
Strong, Inigo Jones (as in note 25), voi. I, p. 119.
" See Ross Parry, ‘The Careful Watchman: James I, didacticism and the perspectival
organization of space’, in Disziplinierung im Alltag des Mittelalters und der frühen
Neuzeit, Veröffentlichungen des Instituts für Realienkunde des Mittelalters und der
frühen Neuzeit 17, ed. by Gerhard Jaritz (Vienna: Der Österreichischen Akademie der
Wissenschaften, 1999), pp. 275-97.
' The line is from Ben Jonson’s masque Pleasure Reconciled to Virtue, performed in
the Banqueting House on Twelfth Night 1618. As the young Prince Charles descends
‘The hill of knowledge', a song calls for the figure of Atlas to allow men to 'read' the
lines and signs of royal education, in both the spectacle itself (performed to, and centred
upon King James) and the presence of Charles as one of the masques. Orgel and Strong,
Inigo Jones (as in note 25), voi. 1. pp. 285-88.
5 See, for instance, the four treatises written as dialogue between a master and scholar
on the importance of knowledge and science (and especially astronomy and the ‘spheres")
that make up Robert Record's, Castle of Knowledge ( 1556). Similarly, for a later text that
describes—at length—the elements of the world using the metaphors of ‘a magnificent
house' (‘heaven is the roofe; . . . the Moone and Starres, night Lamps; severall regions,
severall roomes; . . . the earth, the floor, the sea a mote; the surface of the earth, an
206 C olson. C olson. Parry , ano Sawyer
mnemotechnician) ‘a multitude into one’.*5'1It provided an immediate, accessible

and memorable diagram of a unified and ordered realm. It mapped, in a sense,
the shape of Jacobean power.
Therefore, though just a preliminary sketch, implicit within this image is
Jacobean monarchy in all its complexities. The early seventeenth century was,
after all, a moment of massive epistemic change, at which a medieval court
stared steadfast into the face of modernity. In this respect the ‘Pallas of
Perfection’ represents an appropriation by the monarchical culture of the resur
gent visual technologies of linear perspective that were empowering the eye of
the Renaissance individual. Likewise, it denotes a moment when the tech
nologies of visual mnemonics that were empowering the mind of early-modern
learners was brought, judiciously, into close association with monarchical
culture, particularly James Ts cultural role as royal pedagogue, God’s proxy as
the ‘sorest and sharpest schoolmaster’.55 In this regard, Jones’s image is a com
plex, layered diagram of power, within which the agents that could have pro
moted the sovereignty of the individual have instead been spun to promote the
individuality of the sovereign. ‘The Pallas of Perfection’ (and the other scenes
like it) relates a system of power at a time of epistemic rupture—a system of
power reduced to an accessible, immediate, and intuitive representation. A
lasting image born from ephemera, Jones’s design can be read as a visual
construction of the political reality of early Stuart England. And yet, signifi
cantly, it is also a construction (on privileged sight and mnemonic shape) that is
difficult to replicate in text.
embroydered carpet’), see Samuel Purchas, The Kings Tower and Triumphant Arch of
London. A Sermon preached at Paul Crosse. 5 August 1622 (London: W. Stansby, Henry
Fetherstone, 1623), pp. 25-27. Likewise for an author conceiving the act of faith as an
architectural project, see Robert Barrell, The Spiritual Architecture, or The balance of
Gods Sanctuary to discente the weight and solidity of a true and sincere, from the Leuiti.
and varitie of a fase and counterfeit prefession of Christianity ... a Sermon preached at
Pauls Cross 16. November 1623 (London: Augustine Matthews and John Norton, 1624).
5 See Thomas Fulwood’s The Castel of Memorie: wherein is conteyned the restoring,
augmenting and conserving of the Memotye and Remembraunce, with the safest remedies
and the best preceptes thereunto in any wise appetieyning . . . (London: Rouland Hall,
dwelling in Gutter lane, at the sign of the half eagle and the key, 1562), Chapter 6.
55 James 1, The Trew Law of Free Monarchies: or The Reciprock and MutualI duetie
betwixt a free King and his naturali Sujects, as reproduced in King James vt and t:
Political Writings, ed. by Johann P. Sommervillc (Cambridge: Cambridge University
Press. 1994), pp. 62-84 (p. 83).
Culling off the king 's head 207
C onclusion
What perhaps examples such as the cartoon of Brazil’s República, or the print of
the Dutch truce negotiations, or even the design for a Jacobean masquing scene
show, is that ‘ruptured’ discourses of power can sometimes be discerned more
effectively through non-textual sources. When we are considering power (to use
Bryson’s phrase) ‘not as a monolith, but as a swarm of points traversing social
stratifications and individual persons',w’ it appears to be (in these three examples
at least) the image that thrives as a representational framework. For Jacobean
England it served as a cogent mnemonic; for revolutionary Brazil it served as the
space of private discourse; and for the early republican Low Countries it
manifested and projected a radical but workable alternative to the world of
absolutist, sovereign powers envisaged by Bodin and Hobbes.
Furthermore, when we are confronted with moments when this diffuse loca
tion of power is itself dislocated—whether we call that moment epistemic
change, or ‘out of placeness’—it is the structure (or, indeed, the lack of
structure—the ‘potentiality’) within the format of the image that seems to lend
itself so ideally to the discourse. It is there in the polyvalency of the Dutch
prints. It is there in the self-conscious fissures of the Jacobean festival designs.
And it is captured in the layered networks of meaning within the cartoons carried
by the Brazilian Press. In short Images would seem to convey a structure o f
meaning more applicable, and more responsive, to structures o f power, even (or
maybe especially) at moments of dislocation—whether that dislocation was seen
as perilous, or indicative, or functionally apt in the circumstances where it was
recognized.
Now that History has promoted its images from secondary illustration to
primary evidence (from scenery to protagonist), it is perhaps time to explore
how in this new leading role it can allow us to see very different stories of the
past. And. particularly with regards to these ‘Diagrams of Power’—as Net rather
than Hierarchy, Texture rather than Text—it does seem that there is some
evidence to suggest that there are stories that cannot be told with words alone.
Through the lens of ‘text’—in our sources and in our publications—the past has
all too easily become an ordered, fixed, hierarchical place. With the
kaleidoscope of ‘image’ (above all the digital image) we now look to a past of
movement and interplay.
’ N. Bryson. ‘Semiology and Visual Interpretation", in Visual Theory. Painting and

Interpretation, ed. by N. Bryson, M. A. Holly, and K. Moxey (Cambridge and Oxford:
Polity Press, 1991), pp. 61-73 (p. 71).
‘Serra ex ferro’—‘Serra ex vitro’:1
medieval history—com puters— image messages
reconsidered
G e r h a r d J a r it z
he discourses initiated, determined, and influenced by the contents and
T messages of images are about depicted actions, persons, objects etc. In

this respect, there is certainly no basic difference between images and the
representations of spoken or written text. In our, the historians’, constructions, we
clearly always have to be mindful of these factors and contexts. Of particular
interest are those fields and aspects that are discussed regularly and intensively
in our sources. There, we are confronted with connections of the qualitative and
the quantitative, opening the possibility to search for patterns, their role, rules
1 Thomas Aquinas, Summa Theologiae, I, q. 39, a. 8: ‘Quilibet autem artifex intendit

suo opere dispositionem optimam inducere non simpliciter, sed per comparationem ad
finem. Et si talis dispositio habet secum adiunctum aliquem defectum, artifex non curat,
sicut artifex, qui facit serram ad secandum, facit eam ex ferro, ut sit idonea ad
secandum; nec curat eam facere ex vitro, quae est pulchrior materia, quia talis
pulchritudo est impedimentum finis' [Every artist strives to put the best arrangement
into his work - not the best in absolute terms, but the best in relation to his aim. If such
a composition involves some adverse effect, the artist docs not worry about this. If, for
example, the artist is making a saw for cutting, he makes it of iron, because iron is
suitable for the purpose, and he docs not dream to make it of glass, which is a more
beautiful material, for such beauty is in conflict with the aim]; see Wladyslaw
Tatarkiewicz, History o f Aesthetics It: Medieval Aesthetics (The Hague and Paris, and
Warszawa: Mouton and PWN, 1970). p. 262.
210 G erhard Jaritz
Fig. 1. 'Visitation o f the Virgin’, Master o f the Schottenaltar, pane! painting,

Austrian, 1469/1480. Vienna, Schottenstift. The idealized paved and dean street
in Vienna is appropriate to the ‘Visitation o f the Virgin’. © Institut fü r
Realienkunde des Mittelalters und derfrühen Neuzeit, Krems.
and raising connotations.
When using pictures for such types of analysis, we certainly have to be
conscious of the fact that we are dealing with various levels of the ‘reality of
images’,2 just as we have to be alert when using written or any other source
2 Concerning late medieval images, see Keith Moxey. ‘Reading the ‘Reality Effect’, in
Pictura quasi fictura. Die Rolle des Bildes in der Sachkultur des Mittelalters und der
frühen Neuzeit, ed. by Gerhard Jaritz, Forschungen des Instituts für Realienkunde des
'Serra ex ferro - 'Serra ex vitro ’ 211
material. When, for instance, we analyse medieval visual sources, we have to be

aware that the ‘reality of images’ and this ‘reality of contexts’ influenced the
medieval perception of information by the pictures’ beholders, just as they, albeit
in another way, influence our own perception.
In order for the message to be apprehended, the contents and contexts have to
be proper and purposeful ones. The qualities have to be appropriate to the
objects, to the depicted persons, to the actions. The saw has to be made out of
iron, as Thomas Aquinas stressed in the thirteenth century; the painter cannot
make it out of glass, the more beautiful material, such beauty being in conflict
with the aim, for only an iron saw can convey the act of sawing.
There is always a need for the beholder to have a suitable means to recognize
the message—something that we already find in the well-known example from
the Libri Carolini (end of the eighth century). This concerns two identical
pictures of a beautiful woman: one represents the Virgin Mary, the other
represents Venus. The first is to be venerated, the latter discarded.' The painter
provided the beholders with the solution, with a clearly recognizable sign for
those who were able to read and to understand: one picture received the caption
‘Virgin Mary’; the other received the caption ‘Venus’. In this example it was the*3
Mittelalters und der frühen Neuzeit. Diskussionen und Materialien, 1 (Vienna: Verlag der
Österreichischen Akademie der Wissenschaften, 1996), pp. 15-22; Gerhard Jaritz, “ Et est
ymago ficta non veritas'. Sackultur und Bilder des späten Mittelalters’, in ibidem, pp. 9
13; idem. Zwischen Augenblick und Ewigkeit. Einführung in die Alltagsgeschichte des
Mittelalters (Vienna: Böhlau, 1989). pp. 71-92.
3 Libri Carolini IV, 16: ‘Offerentur cuilibet eorum qui imagines adorant . . . duarum
feminarum pulchrarum imagines, superscriptione carentes, quas ille parvipendens abjicit,
abjectasque quolibet in loco jacere permittit, dicit ille quis: Una illarum sanctae Mariae
imago est, abjici non debet; altera Veneris, quae omnino abjicienda est, vertit se ad
pictorem quaerens ab eo, quia in omnibus simillimae sunt, quae illarum sanctae Mariae
imago sit, vel quae Veneris? Ille huic dat superscriptionem sanctae Mariae, illi vero
superscriptionem Veneris: ista quia superscriptionem Dei genitricis habet, erigitur,
honoratur, osculatur; illa quia inscriptionem Veneris, dejicitur, exprobatur, exsecratur;
pari utraeque sunt figura, paribus coloribus, paribusque factae materiis, superscriptione
tantum distant’ [A man who venerates pictures was shown two pictures of beautiful
women without any captions, which someone had thrown away, caring little for them.
Someone said to him: one of these is a picture of the Virgin Mary and should not be
thrown away, and the other is of Venus, and should at all costs be thrown away. The man
turned to the artist and asked him which one was the picture of Mary, and which one was
of Venus, for they were completely alike. The painter supplied one picture with the
caption: The Virgin Mary, and the other with the caption: Venus. The picture with the
caption: Mother of God, was elevated, venerated, and kissed, and the other, because it had
the caption: Venus, was maligned, scorned and cursed, although both were equal in shape
and colour, and were made of identical material, and differed only in caption]; see
Tatarkiewicz (as in note 1), p. 100.
212 G erhard Jaritz
Fig. 2. 'Christ Visiting St George in Prison panel painting. 1516. Spisská

Sobota (Slovakia), parish church. The idealized paved and clean urban streets is
appropriate to ‘Christ Visiting Saint George in Prison’. © Institut für
Realienkunde des Mittelalters und der frühen Neuzeit. Krems
'Serra ex ferro — 'Serra ex vitro ' 213
Fig. J. 'The Dream o f St Joseph’, Friedrich Pacher, (detail: the hard road to
Egypt), pane! painting, Tirol, before 1500. Innsbruck, Landesmuseum
Ferdinandeum. The negative model o f a bad and dirty road is appropriate to the
Dream o f Saint Joseph ' suggesting that they start the difficult ‘Flight to Egypt ’.
© Institut fur Realienkunde des Mittelalters und der frühen Neuzeit, Krems
214 G erhard Jaritz
Fig. 4. 'The Temptation o f St Martin by the Disguised Devil’, panel painting,

before 1500. Göflan (Südtirol), parish church. The ground is strewn with
stones. © Institut für Realienkunde des Mittelalters und der frühen Neuzeit,
Krems
'Serra ex ferro - Serra ex vitro ' 215
Fig. 5. 'Si Florian and St Sebastian ’ (detail: Emperor Frederick HI as St

Sebastian), panel painting. Styrian. c. 1480. Obdach (Steiermark), church o f the
hospital. The long and curled hairstyle o f Emperor Frederick III. © Institut für
Realienkunde des Mittelalters und der frühen Neuzeit. Krems. Austria
216 G erhard Jaritz
caption that counted; the caption ‘Venus’ created the image of Venus for those
who could read or who were told; the caption ‘Virgin Mary’ turned the second
image into the Mother of God. In other cases, however, it is the presented form
and quality of the actual contents of the pictures that creates the recognition. But
always some kind or type of context of the signs in different levels plays the
important role—as also for our own analyses.
Let us come back to the discourse initiated and influenced by images, and let
us concentrate on intensity, regularity, and the particular connotation offered by
the contents: this combination of the qualitative and the quantitative. For our
studies and our search for relevant patterns, computer-supported image analysis
has proved particularly useful. Some Central European groups of images and
their interpretation may serve as examples. The basis for analysis is about 20,000
late-medieval images,4 described in a very detailed, structured, and standardized
way that are available in a KA.ei(0 -database,5 thus making text-image links
possible in different variants.6 The issues of visual representation that we are
interested in are those that often play a decisive role in late-medieval written
evidence dealing with aspects of daily life and material culture. As examples we
will deal with streets and roads and their quality and cleanliness;7 male
hairdressing;'* other material signs of the outward appearance of social groups,
4 These images are integrated in the database REAL of the ‘Institut für Realienkunde
des Mittelalters und der frühen Neuzeit’ of the Austrian Academy of Arts and Sciences.
See Gerhard Jaritz, Everyday Life in the Middle Ages and Digital Image Analysis ( 1995):
http://rubens.anu.cdu.au/chart/jaritz.html; idem, ‘Computergestützte Bildanalysen in der
Geschichte mittelalterlichen Alltags', Österreich in Geschichte und Literatur, 39 (1995),
156-61. Parts of the contents of REAL arc available on:
http://www.imareal.oeaw.ac.at.
5 Concerning KXeuo, see Manfred Thaller, KXeuo. A Database System (St Katharinen:
Scripta Mercatura Verlag, 1993); Matthew Woollard and Peter Dcnlcy. Source-Oriented
Data Processing for Historians: A Tutorial for kleio (St Katharinen: Scripta Mercatura
Verlag, 1993); see also http://www.gwdg.de/kleio/ and http://gilgamesch.hki.uni-
kocln.de/develop/site/. With regard to Kteico-imagc analysis, see Gerhard Jaritz, Images.
A Primer of Computer-Supported Image Analysis with kleio IAS (St Katharinen: Scripta
Mercatura Verlag, 1993); see also idem, ‘Medieval History and Digital Image
Processing’, Schede Umanistiche. Rivista semestrale dell'Archivo Umanistico Rinasci-
mantela Bolognese, nuova serie ( 1998), 201-09.
6 See Gerhard Jaritz. Bound Images: Encoding and Analysis (1995):
http://www.uky.edu/~kiernan/DL/jaritz.html.
7 On medieval streets and roads, see, e.g., Jean-Pierre Leguay, La rue au moyen âge
(Rennes: Ouest France. 1984): Lynn Thorndike, ‘Sanitation, Baths, and Street-Cleaning in
the Middle Ages and Renaissance’, Speculum 3 (1928), 192-203.
s See Jean Loubier, ‘Das Ideal der männlichen Schönheit bei den altfranzösischen
Dichtem des XU und Xin Jahrhunderts’ [Phil. Diss. Friedrichs-Universität Halle Witten
berg] (Halle a. S., 1890), pp. 44-56; Oskar Voigt, ‘Das Ideal der Schönheit und
'Serra ex ferro Serra ex vitro ' 217
such as pointed shoes,* 9 and the image of peasantry.101They here will be used as
examples to show the general possibilities and uses of visual sign language from
the fourteenth to the beginning of the sixteenth century." For this, they also can
be seen as important individual sources of information for a history of
mentalities of the late Middle Ages.12
The quality of streets and roads regularly played an important role in a variety
of late medieval written sources, particularly those of urban areas: there are
distinct references in laws, craftsman contracts, donations, account books,
chronicles, travel accounts, etc. The poor quality of medieval roads is still a
stereotype in today’s historical presentation of the Middle Ages, particularly the
one taught in schools. But in the many late-medieval discussions about roads,
their cleanliness, the contrast between paved and unpaved, and between easy and
difficult travel over them are all of critical importance. And it seems to be quite
obvious that images in their contexts also deal with such questions and problems
and use them to mediate a variety of different messages.13
On the one hand, we are confronted with the image of the highly idealized
urban street conforming to every requirement as to cleanliness, of top quality,
Hässlichkeit in den altfranzosischcn chansons de geste’ (Phil. Diss., Universität Marburg.

1891). pp. 34-36: Gerhard Jaritz, ‘Young, Rich, and Beautiful. The Visualization of Male
Beauty in the Late Middle Ages’, in The Mart o f Many Devices, Who Wandered Full
Many Ways. Festschrift in Honor o f Jemos M. Bak, cd. by Balázs Nagy and Marceli Scbök
(Budapest: Central European University Press, 1999), p. 63.
9 See Gerhard Jaritz, ‘Schnabelschuh und Hömerhaube oder: Bild, Sachkultur und
Kontextualisierung’, in (9. Österreichischer Kunsthistorikertag. Vergangenheit in der
Gegenwart—Gegenwart in der Kunstgeschichte?, Kunsthistoriker 11/12, 1994/1995
(Vienna: Österreichischer Kunsthistorikerverband, 1996), pp. 8-12.
111 See, generally. Gerhard Jaritz. ‘The Material Culture of the Peasantry in the Late
Middle Ages: “Image” and ‘‘Reality’” , in Agriculture in the Middle Ages. Technology,
Practice, and Representation, cd. by Del Sweeney (Philadelphia: University of Pennsyl
vania Press. 1995). pp. 163-88.
11 Concerning the literary sign language of dress, see Gabriele Raudszus, Die
Zeichensprache der Kleidung. Untersuchungen zur Symbolik des Gewandes in der
deutschen Epik des Mittelalters, Ordo. Studien zur Literatur und Gesellschaft des Mittel
alters und der frühen Neuzeit, i (Hildesheim: Olms, 1985).
12 See Axel Bolvig, ‘Images of Late Medieval ‘Daily Life’. A History of Mentalities’,
Medium Aevum Quotidianum, 39(1998), 94-111.
13 See Gerhard Jaritz, ‘Umweltbewältigung. Der Beitrag der Geschichtswissen
schaften’, in Umweltbewältigung. Die historische Perspektive, ed. by Gerhard Jaritz and
Verena Winiwarter (Bielefeld: Verlag für Regionalgeschichte, 1994). pp. 14—16 and
errata: idem. ‘Alltag in der Stadt des 15. und 16. Jahrhunderts’, in Alltagserfahrungen in
der Geschichte Österreichs, ed. by Emst Bruckmüller (Vienna: ÖBV Pädagogischer
Verlag, 1998), pp. 52-55.
218 G erhard Jaritz
Fig. 6. ‘Parable o f Dives and Lazarus ’ (detail: the sei-vant o f Dives), panel
painting, Upper Rhine, end o f the fifteenth century. St. Paul im Lavanttal
(Kärnten), collections o f the Benedictine monastery. The long and curled
hairstyle o f the servant o f Dives. © Institut fur Realienkunde des Mittelalters und
derfrühen Neuzeit, Krems
'Serra ex ferro - 'Serra ex vitro ' 219
Fig. 7. ‘Labours o f the Months', month o f August (detail: hairstyle o f an

‘improper’ peasant), ink drawing. South German. 1475. Vienna, Österreich.
Nationalbibliothek, MS. 3085, fol. 5r. The chaotic curly hair o f an ‘improper’
peasant. © Institut fü r Realienkunde des Mittelalters und der frühen Neuzeit,
Krems
and in excellent condition with a good standard of pavement. For example, the
particular religious event of the ‘Visitation of the Virgin’ (fig. 1) is integrated
into a depiction of the community for which the image was produced: in this
case Vienna.14 The representation of the street follows two patterns of ideali-
14 See Heinrich Ferenczy, Das Schottenstift und seine Kunstwerke (Vienna: Orac,
1980), p. 194; Walther Brauneis, ‘Beitrag zur mittelalterlichen Topographie der Stadt
Wien". Österreichische Zeitschriftfür Kunst und Denkmalpflege, 27 (1973), 121-31.
220 G erhard Jaritz
zation, that of one’s own town and one of a highly important religious action.
This ideal of one’s own community need not necessarily be a ‘real’ image, but
may obviously just be a type. Such an example, for instance, may be seen in a
representation of ‘Christ Visiting Saint George in Prison’, out of a winged
altarpiece in today’s Slovakia (fig. 2). Again, the paved and clean streets play a
relevant role in the idealization of action and place.15
On the other hand, a contrasting image of roads may be recognized in cases
where a generally negative connotation should be mediated. This is, for instance,
particularly true of the flight of the Holy Family into Egypt. One typical example
is a representation of the ‘Dream of St Joseph’, in which the angel appears and
tells him that he and his family should flee (fig. 3). It shows the very bad, diity,
and hard road to Egypt, a road on which animals forage for food. Bad roads,
paths, and squares may also turn up in connection with other negative
connotations that images should mediate. One of these examples is the
‘Temptation of St Martin by the Disguised Devil’ showing the ground strewn
with stones (fig. 4).16
Religious messages, therefore, obviously use a language of signs that is based
soundly on the practical experience and knowledge of the beholders. In this case
that language is based on regular discussion about an important material aspect
of everyday life: streets and roads. The contrasts in their quality are the means of
providing the audience with auxiliary information to reinforce religious and
spiritual messages. In order to apprehend such results, it is useful to employ the
above-mentioned text-image databases and analyse them qualitatively and
quantitatively with regard to contents and contexts. This means, for instance,
searching the databases for the topics and contents of late-medieval works of art,
where streets and roads are described verbally as well-paved or clean, or where
there are descriptions of dirt, stones, puddles, or foraging animals in the context
of streets. A comparison of the results of these two queries shows ‘message-
bound contrasts’, of which a few examples have been selected above.
Another example may be taken with regard to aspects of the outward appear
ance of people. For the whole of the Middle Ages, the image of beautiful males
followed a certain pattern.17*One of the key indicators was their long and curled
hair of fair or light brown colour.ls Such hair was also used to represent the
actual inner qualities of these beautiful, rich, and young males, or other men
closely connected to them. In images, it was typically used for rulers (fig. 5), rich
and socially important men, their servants (fig. 6), angels, or the apostle John the
15 See Libuse Cidlinská, Gotiche kridlové oliare na Slovensku [Gothic winged

altarpieces in Slovakia] (Bratislava: Tartan, 1989), p. 82.
16 As in note 13.
1' See Gerhard Jaritz, ‘Young, Rich, and Beautiful’ (as in note 8), pp. 61-77.
ISAs in note 8.
'Serra ex ferm - ‘Serra ex vitro ' 221
Evangelist.19 Other groups of lower status in the social hierarchy of medieval

society do not have this type of hair. If such men wear long and curled fair hair
in images, they may obviously be connected with patterns of the ‘world upside-
down’.2" This becomes very clear, for instance, in the obviously satirical and also
didactic illuminations of a South German manuscript of 1475, representing the
Labours of the Months. There, the peasants have a coiffure that by no means fits
their role and social position (fig. 7), but the depiction also shows that they were
not quite successful in their attempt to overturn the order of the world; their curls
look rather chaotic.21 These Labours of the Months, therefore, show proper
peasants’ work done in the proper season. But parts of the outward appearance of
the persons doing this work obviously do not correspond to their social status
and function. In a humorous and satirical way, the pictures seem to represent the
attempt of peasants to rise in the social hierarchy by changing aspects of their
outward appearance. This is a topos that we find regularly, particularly in
medieval literary sources.22 The peasants try to adopt the hairstyle of the nobility
but are not really able to do so. Thus, such images or texts mediate not only
humour and satire to their audience but also a moral: Look what happens to
peasants who act like this; one laughs at them, so be careful that something
similar does not happen to you, if you try to leave your proper place in the social
order.
Late-medieval pictures also operate in a similar way with regard to specific
items of dress. Amongst the most discussed pieces of clothing in the late Middle
19 See Gotik in der Steiermark. Katalog der Landesausstellung Stift St Lambrecht

(Graz: Styria, 1978). p. 135; Jaritz, ‘Young. Rich and Beautiful" (as in note 8), pp. 64-65.
20 Concerning the motive of the ‘world upside-down’ generally, see, e.g., Hedwig
Kenner, Das Phänomen der verkehrten Weh in der griechisch-römischen Antike
(Klagenfurt: Geschichtsvercin für Kärnten, 1970); Roger Chartier and Dominique Julia,
‘Die verkehrte Welt’, in Die unvollendete Vergangenheit und die Macht der Welt
auslegung, cd. by Roger Chartier (Berlin: Wagenbach. 1989), pp. 73-82: Helmut
Hundsbichlcr, Tm Zeichen der ‘verkehrten Welt’, in Symbole des Alltags—Alltag der
Symbole, cd. by. Gertrud Blaschitz, et al. (Graz: Akademische Druck- und Verlagsanstalt,
1992), pp. 555-70: Christa Grössinger, The World Upside-Down. English Misericords
(London: Harvey Miller, 1997).
21 Jaritz, ‘Young, Rich, and Beautiful’ (as in note 8), p. 65.
22 See. e.g., Helga Schüppert, ‘Der Bauer in der deutschen Literatur des Spät-
mittclalters’. in Bäuerliche Sachkultur des Spätmittelalters, Sitzungsberichte der
Österreichischen Akademie der Wissenschaften, phil.-hist. Klasse. 439 (Vienna: Verlag
der Österreichischen Akademie der Wissenschaften, 1984), pp. 125-76; Herman Bract.
“ A thing most brutish’: The Image of the Rustic in Old French Literature’, in Agriculture
in the Middle Ages. Technology, Practice, and Representation, ed. by Del Sweeney
(Philadelphia: University of Pennsylvania Press, 1995), pp. 191-204.
222 G erhard Jaritz
Fig. 8. ‘Duke Albrecht o f Austria’, ink drawing, Tirol, last quarter o f the
fifteenth century. Vienna, Österreich. Nationalbibi., MS. s. n. 12820, fol. 40'.
The pointed shoes o f Duke Albrecht o f Austria. © Institut fü r Realienkunde des
'Serra ex ferro - 'Serra ex vitro '
Fig. 9. ‘Sermon o f St John Capistran in Bamberg’ (detail: Bonfire o f the

Vanities), panel painting, c. 1470. Bamberg (Germany), Staatsgalerie. Pointed
shoes are burnt as an object o f the Vanities. © Institut fü r Realienkunde des
Mittelalters und derfrühen Neuzeit, Krems
224 G erhard Ja ritz
Ages were pointed shoes.23 Throughout Europe, particularly around the mid-
fourteenth and mid-fifteenth century, sumptuary laws and sermons deal with
them, especially prohibiting or criticizing town-dwellers’ wearing of them;
chronicles regularly report problems that occur for men who use them. They are
to be seen as a sign of superbia for large parts of late-medieval society. If one
analyses pictorial sources from the period, one comes across a pattem similar to
what we have seen when examining male hairstyles. Pointed shoes seem to have
been very fashionable and legitimate for people of the upper classes, such as
rulers (fig. 8) or noblemen, and again their servants (fig. 6).24 All of them seem
to have been allowed and privileged to wear them as a sign to show their
exceptionally high status or their close connection to such persons. This clearly
emerges from written as well as pictorial evidence. Conversely, pointed shoes
are regularly used in images to represent negative connotations for specific lower
class people, particularly persons such as tormentors of saints or other enemies
of order and faith (see fig. 4). An example closer to practical life is represented
in an image of the Bonfire of the Vanities initiated by a sermon of St John
Capistran in Bamberg (fig. 9). Here, pointed shoes are shown as one of the
objects of superbia to be destroyed.
Thus, on the one hand, a concentration on individual objects and details of the
outer appearance of persons is enhanced by these new computer-supported
approaches.25 On the other hand, the complete and general appearance (i.e., the
sum total of signs) of the individuals depicted may also be analyzed. An
interesting example is again the use of the ‘image of peasants’ as a positive or as
a negative model.26 The typical legitimate working dress of peasants consists of
the undycd, knee-long, coarse woollen tunic; they wear simple shoes or boots, or
are barefoot. In that manner, they may be depicted publicly and didactically to
23 As in note 9.
24 Concerning obvious problems, also occurring for members of the nobility, cf. the
well-known 'Berner Twingherrenstreit’ in the 1470s. See Leo Zehndcr. Volkskundliches
aus der älteren schweizerischen Chronistik [Schriften der Schweizerischen Gesellschaft
für Volkskunde. 60] (Basel: Krebs, and Bonn: Habelt, 1976), pp. 81-83; Gerhard Jaritz.
‘Kleidung und Prestige-Konkurrenz. Unterschiedliche Identitäten in der städtischen
Gesellschaft unter Normierungszwängen’, Saeculum, 44 (1993), 16-17.
25 See, generally, Gerhard Jaritz, ‘ “Seiden Päntel an den Knien“ oder: Die Hoffart
liegt im Detail’, in Ut populus ad historiam trahatur. Festgabefür Herwig Ebner zum 60.
Geburtstag, cd. by Gerhard Dienes, et al. (Graz: Leykam. 1988), pp. 63-74.
26This use of peasants as a positive and negative model can also occur at the same time
in the same source; see Jaritz, ‘The Material Culture of the Peasantry’ (as in note 10), pp.
165-66; Quellen zur Geschichte des Bauernstandes im Mittelaller, ed. by Günther Franz
[Freiherr vom Stcin-Gcdächtnisausgabc. 31] (Darmstadt: Wissenschaftliche Buchgesell
schaft, 1974). pp. 548-52.
Serra ex ferro 'Serra ex vitro ’ 225
Fig. IO. ‘Labours o f the Months ', month o f July (detail: peasants at mowing and
hay-making), wall-painting, c. 1400. Trento, Castello Buonconsiglio, Torre
d'Ac/uila. The ideal outer appearance o f a field-working peasant.
© Institut für Realienkunde des Mittelalters und der frühen Neuzeit, Krems
show their positive function in medieval society (fig. IO).27 The peasants are
doing the right work at the right time in the right way, and they are correctly
dressed. They are a positive model for the whole of society. But peasants can
certainly also function as a negative model; as we have seen, they may be used
as an object to demonstrate satirically, and also didactically, the wrong way of
behaving. This might be the case in a representation of the Last Judgement
(fig. 11). There, the sinful peasant is clearly clad in dress that certainly does not
fit his work and status: doublet, hose, codpiece, and duckbill-shoes. His negative
inner state is visualized by his outer appearance. The peasants, whom we
mentioned before, with their chaotic curled hair, make the same point (see
fig. 7). They wear doublets and hose that are pointed at the feet, again a type of
dress that certainly does not belong to rural fieldwork (fig. 12). The image’s
27 See Nicolò Rasino, The Frescoes at the Torre Aquila in Trento (Rovereto: Manfrini,
1962); Gerhard Jaritz, ‘Lebensbilder? Die mittelalterlichen Fresken aus dem Adlcrturm
von Trento (Trient)’, in Das andere Mittelalter. Emotionen, Rituale und Kontraste
(Krems: Kunst-Halle Krems, 1992), pp. 127-33.
226 G erhard Jaritz
Fig. 11. ‘Last Judgement' (detail: the devil grips a peasant), panel painting,
Schnatterpeck-workshop, first quarter o f the sixteenth century. Kortsch (Süd-
Tirol), trustee church. The ‘improper’ peasant gripped by the devil on
Judgement Day. © Institut für Realienkunde des Mittelalters und der frühen
Neuzeit, Krems
'Sena ex ferro 'Serra ex vitro ' 227
Fig. 12. 'Labours o f the Months ’, month o f August (detail: one o f the ‘improper’
field-working peasants), ink drawing, South German, 1475. Vienna, Österreich.
Nationalbibliothek, MS. 3085, fo i 5r. ‘Improper’ outer appearance o f a field-
working peasant. © Institut für Realienkunde des Mittelalters und der frühen
Neuzeit, Krems
message can only have been meant comically and satirically and, thus, also
didactically and with a moralizing intention.28 They are not proper peasants any
more.2
2S For the satirical-didactic treatment of peasants in late medieval and early modern art
generally, sec. c.g., Hans-Joachim Raupp. Bauernsatiren. Entstehung und Entwicklung
des bäuerlichen Genres in der deutschen und niederländischen Kunst, ca. 1470-1570
(Niederzier: Lukassen, 1986); Keith Moxey, Peasants. Warriors and Wives. Popular
Imagery in the Reformation (Chicago: University of Chicago Press, 1989), pp. 35-66.
228 G erhard Jaritz
The ‘contrasting’ image messages used painted material objects and pheno
mena that belonged to everyday knowledge and experience in the late Middle
Ages. They have been analyzed through the use of a large database of digitalized
images, linked with detailed, standardized, and highly structured image
descriptions. It is not the ‘realities’ of medieval roads, of hairstyle, of pointed
shoes, or of peasants’ dress that are important, but the messages intended to be
mediated by these painted examples. The application of computer-supported
methods to apprehend these patterns has been mentioned at the beginning of this
contribution but not dealt with extensively. With the help of selected examples,
this article intended rather to discuss some trends in the development of the aims
of (art) historical research in dealing with the source evidence of late medieval
images. An awareness of the relation among—and the context of—the images’
contents, their patterns, and intended messages has started to play an important
role in medieval studies. The computer-supported methods are very suitable for
this type of analysis, and by using them, some of the purposes of analysis have
themselves been modified and developed; comparative research into the ‘texts’
of image representations in a qualitative and quantitative way has increased,
becoming more sophisticated and refined.
It has definitely been one of the major successes of computer-supported
(image) analysis in the historical disciplines during the recent years that many
scholars are now so familiar with it that it no longer seems necessary to place a
strong emphasis on explaining or describing it in detail. Most of the currently
available methods using computer-supported image analysis have been explored
extensively; their theoretical background has often been clarified. As we have
reached this point, we have become much more able to concentrate on our
specific historical research interests.
Quantitative image analysis:
the Painter o f w ooden shoes
AXEL BOLVIG
Introduction
he preceding two decades have witnessed a fundamental change in
T historians’ access to new source material.1 The digital revolution has

created facilities for storing huge amounts of images, the results of which
we have now begun to see. Images constitute a growing part of the material for
research—and presentation too—deployed by medievalists/ The digital revolution
may result in a paradigmatic change in our understanding of the Middle Ages and
in the presentation of our research. Working with images—qualitatively and
quantitatively—we need to establish new methods of analysis and interpretation.
Art history is a discipline that concentrates on images, but art historical
research traditionally has excluded many different kinds of imagery from its field
of operation. Art historians are facing an expansion of their research material as
well as the necessity for a new methodology to take over from that of the
Warburg School. Late in his life Erwin Panofsky gave up the term iconology and
Irvin Lavin confessed that sometimes he thought iconography was an invention
of the devil.1These two statements form a good starting point for art historians to
1 François Gamier, L'âne a la lyre. Sottisier d'iconographie médiévale (Paris: Le

Léopard d'Or. 1988). p. 27.
: Gerhard Jaritz. 'Images. A Primer of Computer-Supported Analysis with Kleio IAS’.
Halbgraue Reihe zur historischen Fachinformatik. 22 (1993). pp. 11-14.
Irving Lavin. 'Iconography as a Humanistic Discipline', in Iconography at the
Crossroads, ed. by Brendan Cassidy (Princeton: Princeton University Press. 1993),
230 A xel B olvig
elaborate on a new methodology, and within the preceding decades we have

witnessed an intense and fruitful discussion concerning new approaches to
images and new definitions of what is art?
Historical methods are firmly based on working with texts and the whole
apparatus for research has been based on and around the written word. At the
time when Nieburg, Ranke, Seignobos, and their followers elaborated on a
scientific historical methodology, images could be reproduced only in black and
white copies as woodcuts or engravings.*5 In a way, pictorial source material did
not exist for the fathers of the historical sciences, or, if it did, then only to a
limited extent. 'Following the recovery in the fifteenth century of so much
ancient literature, written records came to monopolize the sources on which they
drew’.6 Consequently they turned towards philology. In a sense, they represent
the same attitude towards images expressed by Gregory the Great and the many
erudite medieval scholars who thought—or at least wrote—that images are the
Bible of the illiterate. So why use visual source material when you are able to
read a text?7*Until the present, historians have hesitated to fashion new methods
that deal with pictures.6 Of course we shall not completely give up the terms:
Tradition/Überreste or Document/Monument, but the binding of history to philo
logy was invented by the devil. In the years to come he will be defeated.9
Easy access to thousands of images changes the way research can take place
and, consequently, also the contents of that research. One hundred years ago,
access to images required travel. Art could be experienced almost exclusively in
pp. 33,38.
J The list of contributions from Baschet and Baxandall to Warlike and Wirth is
impressive. As a matter of fact there is a need for a bibliography on this subject matter.
5 William M. Ivins, Jr., Prints and Visual Communication (Cambridge, MA: MIT
Press, 1978).
6 Francis Haskell. History and its Images. An and the Interpretation of the Past (New
Haven: Yale University Press, 1993), p. 2.
The Danish founder of a scientific historical methodology. Kristian Erslev. published
in 1911 a book Historisk Teknik, which is still in use at the universities. In the book he
only once refers to an image, namely the depiction of green horses on the Bayeux
tapestry. He states that we do not believe that horses were green in the eleventh century!
Concerning the lack of a methodology among Danish historians towards visual material
see Axel Bolvig, ‘Med passende ændringer’. in Spiv og salte. ed. by Tove Hansen
(Copenhagen: Rhodos, 1990). pp. 225-37.
6 Rainer Wohlfeil, ‘Methodische Reflexionen zur Historischen Bildkunde’, in
Zeitschrift für Historische Forschung. Beiheft 12 ( 1991 ). pp. 17-35.
9 Concerning the relationship between history and art history see Jérôme Baschet, ‘Les
images: des objets pour l’historien?’, in Le Moyen Age aujourd'hui. Actes de la rencontre
de Cerisy-la-Salle, ed. by J. Le Goff and G. Lobrichon. Cahiers du Léopard d’Or. 7
(Paris: Le Léopard d'Or. 1997), pp. 101-35.
The Painter of wooden shoes 231
museums and churches, particularly cathedrals. The museums and a number of

the cathedrals were situated in the most important cultural centres of Europe. The
subjects of research were therefore centralized and reflected the purchasing
policy of the great museums. Very often, art-historical research was concentrated
on rather small numbers of subjects, which were defined as Fine Art. Historians
very seldom visited museums in order to research.
The experiences of the visitors to museums and cathedrals were only fixed on
their retinas. Many art historians of the nineteenth century developed an out
standing visual mnemonics, but still it was an unstable means of storing visual
information. From that personal image-store, art historians might retrieve many
examples with forms and contents from artworks experienced years ago pertinent
to the subjects of their current investigations. Checking and modifying hypo
thetical connections required another trip to museums and cathedrals perhaps far
away from the subject of investigation. The result was often arbitrary and
naturally dependent on the museums and cathedrals actually visited by the
investigator."1
Later, the situation was improved by a growing number of illustrations in
printed editions. The available visual material began to grow and is still growing.
For a long time, it was still a matter of access to relevant printed material and of
what material has or has not been actually printed. A subject that was made the
object of massive printed reproduction tended to become a centripetal force for
explanation of artistic imitation and influence. Theories of artistic influence and
migration of style grew dependent on what was printed and accessible. During
recent decades, the visual history of the past is most often the history of slides
and reproductions."
Today, the accessibility of visual material is growing rapidly. The establish
ment of huge image databases is an expanding undertaking in numerous insti
tutions. The computer revolution has embraced the visual side of the past. In the
future, art history will have to expand its field of research to new kinds of
images, and historians will be forced to include images as a vital source material.
The accessibility of images will turn history and art history upside down. Today,
you can write about medieval art in one volume, but you still can print only a
very limited selection of the works of art. In future, you will—in principle—be
able to gather pictures of all the medieval artefacts you like on huge hard disks
and browse through thousands of images. But nobody will have the energy and10
10 Regarding connoisseurship. see Jaynie Anderson. Collecting Connoisseurship and

the Art Market In Risorgimento Italy. Giovanni Morelli’s Letters to Giovanni Melli and
Pietro Zavaritt (1866-1872) (Venice: Instituto veneto di scienze, lettere ed arti, 1999)
[Dr. Phillip Lindley has kindly drawn my attention to this book].
" Charles S. Rhyne, 'Computer Images for Research, Teaching and Publication in Art
History and Related Disciplines,’ Visual Resources 12 (1996), pp. 19-51 (p. 25).
232 A xel B olvig
time to read about this enormous body of images.12*15

The rapid diffusion of the Internet is facilitating easy access to widespread
visual source material." Among other things, the Internet will erode traditional
nation-bound research and presentation. The construction of the Internet rests on
a kind of equality among the thousands of individual websites and, consequently,
also among the growing number of image databases." There is no centre on the
Net and therefore no periphery. Looking for similarities between pictures or
eventual models of your object of investigation will take you in many directions
and not to a specific centre for artistic creativity and innovation. Centre-
periphery theories are homeless in cyberspace. They are replaced by intercultural
and interactive structures that cannot avoid influencing our medieval studies. The
question of individual authorship will become less important." In conjunction
with a growth of research in visual history we shall witness a corresponding
growth of interdisciplinary research that will not work within and accept today’s
national frontiers.
The computer revolution involves new ways of defining an image. An image
manifests itself in or is presented by means of physical material. This implies
that it has a physical demarcation in proportion to something that is not a picture.
The cropping or the frame sets the limit. Of course a painting can hold several
individual subordinate motifs but the frame sets a limit to the contents and the
extent of the visual message. With the invention of photography, cropping has
had a growing importance to the definition of an image."’ This development has
reached its peak with the digitalization process. Cut-and-paste functions are
indispensable tools when working with computer-supported research in
pictures.17 By these means we, as users, create our own images, and each of us
produces an individual verbal definition and description. The process of
transcribing visual information to a linguistic system is twofold. It is based on
conventions—that is, iconography in a broad definition—and it is based on
12 Axel Bolvig, ‘Ars longa—vita brevis'. Medium aevum quotidianum 39 (1998). pp.
9-20(15).
" Rhyne (as in note 11 ).
IJ A collection of more than five hundred links to web-sites with medieval images is to
be found on www.kalkmalerier.dk/links.
15Cf. Ivan Gaskell. ‘History of Images', in New Perspectives on Historical Writing, ed.
by Peter Burke (University Park: Pennsylvania State University Press. 1993), pp. 168—
92.
14 Harold Evans. Pictures on a Page (London: William Heinemann. 1978). pp. 203-
36.
17 Charles S. Rhyne, ‘Images as Evidence in Art Histoiy and Related Disciplines', in
Museums and the Web. ed. by David Bearman and Jennifer Trant (Pittsburgh: Archives &
Museum Informatics. 1997), pp. 347-61.
analogy, which also itself is based on coded systems that by analogy refer to an
outside world. Having no grammar or linearity, an image is an open system that
cannot by itself present a precise amount of information.IS Consequently, each
scanning and, even more, each cutting-out of details requires new definitions that
can vary ad infinitum. Defining or indexing these new digital images is a huge
problem in the establishment of image bases but contrariwise also gives the
researcher new opportunities to create relevant source material and to bring
together visual information from a variety of pictures.
The definition of an image, then, is bound to its framing and/or cropping and
cutting.'1’ With wall paintings, the problem is that there is no material frame in a
traditional sense. An oil painting has a frame, which does not belong to the
image. A wall painting needs no frame, as the material of which it consists is the
same as the surrounding walls.1 81920 Consequently the building itself, the archi
tecture, constitutes the frame, within which many individual iconographie or
analogue motifs can be traced. When dealing with wall paintings, you always
make a selection based not on framed subjects but on iconographie definitions or
the recurrence of motifs, be it the Crucifixion, Saint Laurence, the heavenly
Castle, or a wooden shoe. The selected subject may receive a new ‘identity’, but
it still belongs to the totality of the architectural interior.21
The above-mentioned considerations will, in the following section, be
exemplified by the inclusion of the image base with Danish wall paintings,
which is accessible on the Internet on www.kalkmalerier.dk/.22
D anish W all P aintings: A ltern a tive Im ages
In some twenty churches situated in the district of Southeast Funen and the
island of Langeland, we find vault decorations that do not fit traditional iconog
raphy or our understanding of artistic expression. It is a matter of a very strange.
18 Susanne K. Langer, Philosophy in a New Key. A Study in the Symbolism of Reason.

Rite and Art (Cambridge, MA: Harvard University Press, 1942); Roland Barthes,
‘Rhétorique de l’image’, in Communications. 4 (1964). pp. 40-51.
19 Gerhard Jaritz, ‘Comparative Analysis of Historical Image Information', in
Electronic Filing. Registration and Communication of Visual Historical Data. ed. by
Axel Bolvig (Göttingen: Max-PIanck-Institut für Geschichte, 1995), pp. 29-33.
20 Sometimes we find painted frames. This goes especially for pictures of St
Christopher. Still the painted frame belongs to the total decoration of the church.
21 Axel Bolvig, ‘Hæc pictura completa fuit—Om kalkmalerier og rammer’, in Hikuin,
ed. by Annedorte Vad (forthcoming).
22 Jesper Jerne Borrild. ‘Medieval Danish Wall Paintings—an Internet Database’, in
Medium aevum quotidianum. 39 (as in note 12), pp. 21-36.
234 A xel B olvig
Fig. I. Gudme church. Wall painting, ‘wooden shoe’. 1488. ©Author

perhaps ‘primitive’, and certainly alternative imagery.’1 Traditional indexing of
these paintings based on conventional iconography or visual analogy is difficult.
The churches were decorated during the last three decades of the fifteenth
century by an artist or a workshop that has been named the Painter of Wooden
Shoes (‘træskomaleren’). The name has been given because the workshop has
painted one or more wooden shoes, among the peculiar imagery and ornamenta
tion, in most of the churches (fig. I ).*2425
The painted wooden shoes are not framed in any way, so even if they possess
their own connotations they cannot be treated as if they were isolated from the
rest of the decorations in the churches. They belong to an imagery consisting of
strange and grotesque figures and birds, almost abstract configurations, with very
few traditional religious subjects. All the configurations appear as isolated
motifs, but none of them have painted frames to give them an explicitly indi
vidual existence. They belong to one ‘picture’, that is to say the whole decoration
within the church. All the picture’s information must be interpreted on the level
of the ‘reality’ of the image and not of the ‘reality’ of actual life.15 The depicted
wooden shoes must be evaluated in their context.
21www.kalkmalerier.dk—search: HMJ in arkivnr.

24www.kalkmalerier.dk—search: HMJ-152 (Gudme).
25Jaritz (as in note 2), p. 16.
Fig. 2. 0rbœk church. Wall painting, 'hand with a knife’. 1490. ©Author
In 0rbæk church the Painter of Wooden Shoes has depicted an arm stretching
out of one of the built-in holes in a vault. The hand is holding a pointed knife
(fig. 2).26What does that mean? What are the traditional connotations of a knife?
In the long history of mankind, the knife has always been the murder weapon of
the ordinary man. Today we talk about ‘die Dolchstoss Legende', referring to the
German army being stabbed in the back by the civilian population during the
First World War. The Night of the Long Knives refers to Hitler’s execution of
many of the S. A. leadership without trial. When forced to retire in 1908, the
Danish Prime Minister J. C. Christensen said that he was ‘stabbed in the back’.
Since 1808 the expression ‘they are at daggers drawn' has meant unbridled
hostility. 'Mack the Knife" is a famous scoundrel in one of the plays by Bertold
Brecht. The Serbian ultra-nationalist Vuk Draskovic published in 1982 a book
entitled 'The Knife", in which the knife symbolizes the Muslims, the ugly and
dangerous enemy. Looking farther back in history we see that the knife always
has been connected to negative and deadly conceptions. Lady Macbeth says
‘Give me the daggers: the sleeping and the dead are but as pictures. . ,’.27 The
Danish king Erik Klipping was murdered in 1286: the annals write that the
www.kalkmalerier.dk—search: HMJ-125 (0rbæk).

27 William Shakespeare, Macbeth, act. II. scene 1 (murder of Duncan, king of
Scotland).
236 A xel Bo i.vig
king’s body had fifty-six wounds.28 In another source we read: ‘In the year 1286
king Erik, son of Christopher of pious memory, died on St Cecilia’s night,
caused by the knives of the impious, terribly murdered by his own men in his
own bed’.” According to the written material, murderous knives belong to the
impious.
In the wall paintings, too, the knife occurs as a negatively connoted tool, a
tool for murdering, wounding, and cutting. The Circumcision of the child Jesus,
which is the first step on his bloody story of salvation, is performed with a
knife.*
50The executioners skin St Bartholomew with knives,51and the breasts of St
Catharine are cut off with knives.52*Of course the knife is the most practical tool
to be used for these purposes—what else should they use?—but it hurts the
victims. It is being used for torture. The fatal function of the knife can be seen in
depictions of the deadly sin Ira—Anger. In Jprlunde church a visualization of the
seven deadly sins was painted around 1510. Anger is depicted as a murderer with
a knife.55 In Fanefjord church there is a depiction of Joab stabbing Abner from
behind with a knife: it is a betrayal, just as is the kiss of Judas.54 In Vigersted
church a juggler is playing with and throwing two pointed knives. The dress
accentuating his body and the proximity of a woman performing a lascivious
dance classify the juggler as the opposite of established social groupings.55
In contrast, the killing performed by professional soldiers by order of the
28 ‘Annales Ryenses", Danmarks middelalderlige annaler (Selskabet for udgivelse af

kilder til dansk historic, 1980), p. 175: ‘Rex E[ricus] occisus est in lecto in nocte sancte
Cecilie a suis, quos maxime dilexerat. inflictis sibi ulneribus DVI'.
29 Libri datici Lundenses, Land domkapitels gavebpger (Selskabet for udgivelse af
Kilder til dansk Historie. 1973), p. 296: ‘Anno incarnacionis M.CC.LXXX.V1. rex Ericus,
pie recordacionis Chrisofori regis quondam filius, pugionibus impiorum occubit in nocte
beate Cecilie a suis in proprio lecto crudelissime trucidatus'.
50www.kalkmalerier.dk—search: sh/46. 14/139 (Tingsted).
51 www.kalkmalerier.dk—search: 20/72 (Tyvelse).
52www.kalkmalerier.dk—search: 12/57 (Árhus Cathedral).
55 www.kalkmalerier.dk—search: 16/19 (Jorlunde); Danmarks Kirker, Frederiksborg
amt, 1975. pp. 2259-60: in some woodcuts in Pèlerinage de la vie humaine the vice
Treachery is depicted with a pointed knife in her hands: see Michael Camille. ‘Reading
the Printed Image', in Printing the Written Word. The Social History of Books circa
1450-1520. ed. by Sandra Hindman (Ithaca and London: Cornell University Press, 1991 ),
pp. 259-91.
54 www.kalkmalerier.dk—search: 12/19 (Fanefjord): Ulla Haastrup, ‘Danske kalk-
malerier 1500-1536’, in Danske kalkmalerier 1500-1536. ed. by Ulla Haastrup (Copen
hagen: Nationalmuseet. 1992). pp. 37-38.
55 www.kalkmalerier.dk—search: 22/198 (Vigersted).
The Painter o f wooden shoes 237
authorities in the Massacre of the Innocents normally is performed with swords.1''

And Abraham uses not a knife, of course, but a sword to sacrifice his son Isaac.”
Real heroes fight each other with lances or swords, never with knives.1* I need
not mention that St Martin cuts his coat with a sword, even if a knife would be
more useful.''1And in the Agony, St Peter is armed with a sword, which he uses
to cut off the ear of Malchus.40Of course a holy man cannot use a knife to attack
a soldier, even if he might have had one in his belt.
The knife, therefore, generally has murderous connotations, and such conno
tations must be attached to the knife in the hand stretching out of the hole in 0rum
church. This knife is not the only one depicted in the church. On a painting—now
whitewashed over—could be seen a very strange person putting a pointed knife
into the anus of a devil or a monster in order to tap fluids into a jug.4142*In this motif
the knife is used to exploit the devil or Evil itself. In Allesp church the Painter of
Wooden Shoes has depicted another hand holding a knife.4' In Ryslinge church the
same workshop presented a grossly caricatured male person equipped with knives
at both his hips.41 The knife depicted by the Painter of Wooden Shoes in these
parish churches is loaded with dangerous, murderous and rebellious connotations.
How does the negative knife correspond to the rest of the decorations?
Historians and art historians have had difficulty relating to the imagery of the
Painter of Wooden Shoes. ‘For a modern viewer they are not easily under
stood1.41 It makes you wonder how they ever came to be accepted as a suitable
decoration in the church.45*
Grotesque masks with a fool's cap and hands that grasp a rib or a knife might be
considered as a reflection of an extremely baroque caprice which combined with
images marked by dark superstition and horror gives you an insight into emotional
life of the late Middle Ages seen from its most glaring sides.*
J<>www.kalkmalerier.dk—search: barnemord*.
” www.kalkmalerier.dk—search: 22/27 (Glesborg).
18www.kalkmalerier.dk—search: 15/33 (Dronninglund).
" www.kalkmalerier.dk—search: 30-3/79 (Tuse).
411 www.kalkmalerier.dk—search: Tilfangetagelse: John 18.10 writes that Peter drew
his sword.
41www.kalkmalerier.dk—search: HMJ-123 (0rbtek).
42www.kalkmalerier.dk—search: HMJ-219 (Allesp).
4' www.kalkmalerier.dk—search: HMJ-32 (Ryslinge).
uDanske kalkmalerier ¡475-1500. ed. by Ulla Haastrup (Copenhagen: National-
museet. 1991 ). p. 46.
45 Hans J. Frederiksen. 'Troens kunst'. in Ny demsk kunsihistorie. vol. 1 ( 1999). p. 217.
* Egmont Lind. Report to the National Museum, quotation from Lise Gotfredsen,
‘Nogle sære billeder'. in ICO, nos 3-4 (1977), p. 20.
238 A x e l B o l v ig
Fig. 3. Stege church. Wall painting, ‘two wooden shoes'. 1494. ©Author
Another scholar, Lise Godtfredsen, concludes that the painter was psychotic—
maybe schizophrenic, ‘which means that we must be content with not being able
in a satisfactory way to interpret his images. It also means that we no longer need
to spend time researching them to find a deeper theological meaning. Their logic
eludes daylight and our senses’.47
Nevertheless, researchers have thrown themselves into interpreting the
decorations of the Painter of Wooden Shoes. Ulla Haastrup writes that these
primitive decorations performed by bricklayers become meaningful representa
tions of symbols of the course of the year with its ecclesiastical festive days and
secular calendar entries. She suggests that the many depicted wooden shoes,
rather than being emblems, must be regarded as symbols. She asks if a wooden
shoe symbolizes an artisan because workers used this special footwear.4849Henrik
M. Jansen has written a book about the Painter of Wooden Shoes, which offers a
broad survey of decoration campaigns of this workshop.4'’ Leif Sondergaard takes
a standpoint opposite to Lise Godtfredsen’s. While she thinks that the images in
0rbæk church were painted by a psychotic person, Leif Spndergaard is
convinced that they stem from the thoughts and beliefs of ordinary artisans in the
countryside in the late fifteenth century. This means that they possess logic and
are meaningful, but in a different way from images made by professional
painters. He concludes that most such wall paintings in churches decorated by
47 Gotfredsen (as in note 46), p. 28.

* Danske kalkmalerier 1475-1500 (as in note 44). pp. 30-3 I.
49 Henrik M. Jansen, Trceskomaleren—Et fynsk kalkmalerivcerksted i 1400-tallet
(Svendborg: Svendborg & Omegns Museum. 1991).
the Painter of Wooden Shoes are intended to protect the vaulting against
demonic forces from outside: they are apotropaic.50The database of Danish wall
paintings offers an opportunity for quantitative comparison, and querying the
database about wooden shoes provides you with another answer. In the relatively
few images that depict artisans, we never find a wooden shoe worn by them.51
One might believe that peasants wore wooden shoes, but this is not the case.
Peasants are depicted in the First Labour and the Legend of the Fast Growing
Seed. A search of these subjects reveals no wooden shoes at all.52In the Dance of
Death the peasant does not wear wooden shoes.53 And neither do the two
hunchbacked peasants in the Dominican monastery at Arhus.54 We get the same
negative result when making a query on shepherds in the database.55 In the
depiction of the Prayer of the Rich and the Poor Man, neither man is wearing
wooden shoes.56 Apparently all the above-mentioned social groups wore a kind
of leather shoe; certainly not wooden shoes.5758
The wooden shoes depicted by the Painter of Wooden Shoes and the few
other equivalent depictions are not designed as an ordinary foot-shaped piece of
footwear. Normally they have high wedges and long pointed toes (fig. 3).5SThey
50 Leif S0ndergaard. "En hândværkers tro og tanker—om kalkmalerieme i 0rbæk

kirke", Ârbog for Svendborg & Omegns Museum 1995 (1996), pp. 55-65: Leif
Spndergaard. "Magiske tegn. figurer og formier i senmiddelaldcrlige kalkmalerier’. in
Billeder i middelalderen. Kalkmalerier og alterlavier, ed. by Bisgaard, Nyberg. and
Sondergaard (Odense: Odense Universitetsforlag, 1999). pp. 165-216.
51 www.kalkmalerier.dk—search: 16/147 (Âstrup): 22/199 (Vigersted): sh/598 (Kirkerup):
11/90 (Tved): 15/6. 15/8. 15/9 (Sulsted); Gislinge. reproduced in Niels M. Saxtorph, Danmarks
kalkmalerier (Copenhagen: Polilikens Forlag. 1986). p. 92: Tirsted. reproduced in Axel Bolvig,
Dagligliv i middelalderen (Copenhagen: Gyldendal, 1979), p. 30.
52 www.kalkmalcrier.dk—search: 20/150 (Hyllested); 12/180 (Lyngby); 14/141
(Tingsted): 14/112 (Tâgerup); 30-3/62 (Tuse): 17/57 (Keldby); 14/44 (Reerslev): sh/316
(Gjerrild): 32-3/73 (Elmelunde); 17/130 (Lem).
55www.kalkmalerier.dk—search: ko 136 (N0rre Alslev).
54www.kalkmalerier.dk—search: 11/125 (Arhus Dominican monastery).
55www.kalkmalerier.dk—search: sh/358 (Reerslev): 30-3/53 (Tuse): 9/34 (Morkov).
56 www.kalkmalerier.dk—search: fattig*.
57 The publication of the comprehensive excavations in Svendborg, close to the
territory of the Painter of Wooden Shoes, mentions a large number of leather shoes but no
wooden shoes: see Willy Groenman-van Waateringe. Leather from Medieval Svendborg
(Odense: Odense University Press, 1988). The same can be said of the huge collections of
shoes at the museums of Coventry; see Susan Thomas, Medieval Footwearfrom Coventry
(Coventry: Coventry Museum. 1980). Encyclopaedia Britannica informs us under the
reference "sabot" that wooden shoes are used by peasants in France and the Netherlands.
www.eb.com: 180/cgi-bin/g?DocF=micro/515/84.html.
58 www.kalkmalerier.dk—search: HMJ-104 (Vindinge); HMJ-107 (Frprup); 17/67
240 A xel Bolvig
Fig. 4. Nibe church. Wall painting, 'pig wearing pattens’. Early sixteenth
century. © Author
The Pointer o f wooden shoes 241
Fig. 5. Gislev church. Wall painting, 'Crowning of Thorns’, detail, c. 1520. ©

Author
connote the dangerousness of pointed objects exactly as does the knife.
A search in the database on wooden shoe (‘træsko’) reveals quite another
picture of the figures using this footwear.51' In the chapel of The Three Magi at
Roskilde Cathedral, the youngest of the kings is depicted as an extremely dressy
young man.6061*He has long bright hair under his hat, which looks like a crown. His
dress shows off his body. The sleeves are slit up to his upper arms. In the belt a
pointed dagger is placed like a phallic symbol. On his feet are long elegant
wooden shoes— ‘pattens'—with high wedges. This type of shoe functioned as an
overshoe, which raised the person further above the dirty and muddy ground.
The detailing of these wooden shoes followed the fashion of leather shoes.'’1It is
not unusual that the youngest of the kings is sumptuously and provocatively
dressed (which did not prevent guardians of morality all over Europe making
sarcastic remarks about clothing that reveals the body).
Balthazar is not the only one to wear elegant pattens. Among the paintings in
Nibe church we find a pig with a crossbow walking on its hind legs with an
equivalent kind of wooden shoe (fig. 4).“ In medieval imagery pigs are nega
w www.kalkmalerier.dk—search: træsko in ting (the Danish characters can be replaced

by * in the search field).
“ www.kalkmalerier.dk—search: sh/188 (Roskilde Cathedral).
61 Francis Grew and Margrethe de Neergaard, Shoes and Patterns. Medieval Finds
from Excavations in London. 2 (London: HMSO 1988). pp. 91-101.
K Reproduced in Kurt Nielsen and Spren Skovfo, Nibe kirke. Dens bygning,
242 A xkl B olvig
tively connoted.6' The same type of shoe is to be found in Udbyneder.61Here it is

worn by a devil, which at The Day of Judgement is chasing souls down to the
Mouth of Hell. The devil’s patten is worked out with a very long pointed toe. His
other leg also is connected to wood: under his knee a wooden leg is fixed. In the
Day of Judgement in Hprve church, a devil wears wooden shoes of a more
ordinary kind on both feet.*65
In Gislev church, one of the tormentors at the Crowning of Thorns wears
wooden shoes (fig. 5), as does the tormentor at the whipping of Christ in the
Post-Reformation paintings of Sulsted (fig. ó).66 In Hyllested church, in a
marginal image, a man fighting a woman is wearing wooden shoes. On an arcade
pillar in Ârhus Cathedral we find a caricature of a lansquenet.67 His head with a
long pointed beard is placed on a long bird’s neck. In his hands he is holding a
sword and a halberd. He has lost one of his legs, which is replaced by a wooden
leg. On the other foot he wears an elegant pointed wooden shoe. This caricature
of a lansquenet, like the above-mentioned figures, represents the evil and the
dangerous. All examples, therefore, of the use of wooden shoes, are negatively
connoted, regardless of who is wearing them. In the imagery of medieval Danish
wall paintings, the wooden shoe does not exist as the footwear of the ordinary
man, either for work or for leisure. There seems to be no difference between
pattens and ordinary wooden shoes within the ‘reality’ of the images.68
udsmykning og mennesker (Nibe: Nibe menighedsrâd, 1995), p. 33.

65 Frithiof Dahlby, Symboler og tegn i den kristne kunst (Copenhagen: J. Fr. Clausen.
1965). pp. 68 and 145: Dorthe Falcon Mpller. Music Aloft (Copenhagen: Falcon. 1996). p.
99; see also Michael Camille, ‘At the Sign of the Spinning Sow: The "Other” Chartres
and Images of Everyday Life of the Medieval Street’, in this volume.
64 www.kalkmalerier.dk—search: 11/22 (Udbynedcr).
65 www.kalkmalerier.dk—search: 16/83 (Hprve).
66 www.kalkmalerier.dk—search: sh/426 (Sulsted).
67 www.kalkmalerier.dk—search: 66/68 (Ârhus Cathedral).
68 The wooden shoe with high wedges (patten) was used as an overshoe for outdoor
use. It is now and then depicted in Flemish painting where apparently it is impossible to
find expressed any specific symbolic value. On a small altarpiece ‘Christ carrying the
Cross’ Kunsthistorisches Museum, Vienna. The Paintings (London: Scala Books. 1997),
p. 60, Hieronymus Bosch has a man near the cross wear pattens. It is difficult to say
whether he represents Simon of Cyrene or an executioner; reproduced in Wolfgang
Prohaska, Kunsthistorisches Museum, Wien. The Paintings (London: C. H. Beck/Scala
Books. 1997). p. 60. In ‘The Temptation of Saint Anthony’ at Lisbon's Museo Nacional.
Bosch depicts a strange figure who probably represents an evil messenger wearing pattens
(Luis Afonso has kindly drawn my attention to this picture). On an altarpiece by Joos van
Cleve (Kunsthistorisches Museum, Vienna), the Virgin Mary and the Jesus Child are seen
together with Joseph, who has taken off one of his pattens, probably in respect for the
holy child. In Jan Gossaert’s ‘St. Luke depicting The Virgin Mary with the Child'. Luke
Jean-Claude Schmitt has written about the material and the symbolic realities
in the Middle Ages with a starting-point in shoes. He refers to a story, a variant
of the tale of the Jongleur de Notre-Dame, about a performer who for lack of
money played a piece of music in honour of the representation of Jesus Christ at
Volto Santo. The figure of Jesus gave his right shoe as a reward to the musician.
Schmitt points to the fact that taking off a shoe indicates that you relinquish your
possession of a woman or a field or that you convey your right to the one who
is kneeling at his drawing desk; he has taken off his pattens, apparently to show deference
to Mary and her child: reproduced in Prohaska (op. cit.), p. 63. Taking off the pattens as a
sign of deference is shown in several images. In Carlo Crivelli’s 'The vision of the
blessed Gabriel' (The National Gallery. London: NG 668), the saint is kneeling by the
side of the Virgin Mary with the Child and has taken off his pattens. In Robert Campin’s
'Merode altarpiece’. the right wing is devoted to Joseph as a supplement to the central
motif of the Annunciation. He is silting in his workshop making a mousetrap. In all its
realism Campin's painting is saturated with symbolism, and the mousetrap refers to St
Augustine’s statement that God had to make his appearance on earth in the shape of a
human being in order to cheat the devil: ‘The Lord's cross was the devil's mousetrap'.
Under Joseph’s long gown we see his left foot wearing a patten. Joseph, not showing
deference but creating the symbolic devil's trap, keeps his wooden shoe on; reproduced in
Jutta Held and Norbert Schneider. Socialgeschichte der Malerei vom Spätmittelalter bis
ins 20. Jahrhundert (Cologne: DuMont. 1993). p. 25. In the "Presentation in the Temple’
(1450-1500) (The National Gallery. London: NG 706). Joseph is wearing beautiful
pattens. In this scene he is not showing deference to the child and consequently he has not
taken off his pattens. The most famous pair of pattens undoubtedly are those depicted in
Jan van Eyck’s ‘The Betrothal of the Arnolfini' (The National Gallery, London). On the
floor in front of Arnolfini lie his pattens with high wedges (see plate 1). They have long
pointed toes like the pattens of Balthazar and the devil of Udbyneder. It seems obvious
that he has taken off his overshoes, but their central position in the painting indicates that
like St Luke he is showing deference to his fiancée. Arnolfini belonged to a wealthy
milieu: drawing from the finds of excavations in London. Francis Grew and Margrethe de
Neergaard state that especially during the late fourteenth century, pattens were worn in
the City chiefly as a useful fashion accessory to protect the feet and the shoes of the well-
to-do: see Grew and Neergaard (as in note 61 ). p. 91. In a wall painting by Domenico Di
Bartolo in Santa Maria delle Scale. Siena (1443), a man is seen carrying a bier (?) wearing
the same kind of wooden shoes: reproduced in Georges Duby. An 1000 an 2000. Sur les
traces de nos peurs (Paris: Textuel. 1995). p. 35. All the above-mentioned representations
of wooden shoes (pattens) differ from the pointed wooden shoes of the painter of Wooden
Shoes. I dare say that the latter’s wooden shoes (sabots) might contain a locally and
socially delimited iconography of their own. but an iconography that anyhow connotes a
widespread stratum of opposition and rebellion.
I have collected the above mentioned examples of depicted wooden shoes by travelling
or looking in illustrated books and not by systematic research and certainly not—yet—by
a search in image databases, which in the future without doubt will give me much more
information to compare with the Danish material.
244 A xel Boi.vici
Fig. 6. Sulsted church. Wall painting, ‘Whipping of Christ’. 1548. © Author

receives the shoe.“ This notion connected to the shoe goes back to the book of
Ruth in the Old Testament.’" Whether there is a symbolic connection between
this notion of shoes and the many depicted wooden shoes in the Danish churches
I cannot say. It seems most likely, I think, that the connotations in the Danish
paintings point in another direction.*70
Jean-Claude Schmitt, ‘Réalité materielle et réalité symbolique. A propos du soulier

de Christ', in Pictura quasi fictura. Die Rolle des Bildes in der Erforshung von Alltag und
Sachkultur des Mittelalters und der frühen Neuzeit, ed. by Gerhard Jaritz (Vienna: Der
österreichischen Akademie der Wissenschaften. 1996). pp. 73-85. The fact that Amolfini
has taken off his shoes at Van Eyck's Betrothal of the Arnolfini might be evaluated in this
perspective, but apparently it does not make logical sense.
70 Ruth 4.7-8.
The many wooden shoes associated with the Painter of Wooden Shoes are not
at any rate depictions of the footwear of peasants and artisans. They cannot
function as symbols of bricklayers. Their design does not fit an analogue
depiction of a piece of footwear. As symbolic representations, the wooden shoes
connote provocation (Balthazar, the lansquenet, and the pig), violence (the
tormentors), and alternative forces in opposition to the mightiest ruler (the devil).
As a symbol, the long pointed toe connotes the dangerous. The murderous knife
in the hand sticking out of a hole in 0rbæk church is grouped with the pointed
wooden shoes in connoting danger, provocation, and rebellion. In this connection
it is important to note that the French word for wooden shoe is sabot and that
sabotage means both the making of wooden shoes and vandalism. About 1300,
saboter meant to thrust or to wound, and during the sixteenth century the word
meant to shake, shock, or to torment.71
We do not know if there is a connection in meaning between French
etymology and the iconography of Danish wall paintings, but the identical nature
of the contents is remarkable. The rebellious significance hidden in these
'accidental’ wooden shoes is underlined by the time of their execution, the years
around 1500. Exactly during these decades Germany experienced violent
peasants' rebellions that culminated in the 1520s. The revolutionary movement
was named after its symbol of unity—a ‘Bundschuh’, which was the typical
footwear of peasants in late medieval Germany. On their standards the peasants
painted a shoe: the Bundschuh they wore themselves. The Bundschuh thus
became the sign of rebellion.
At the same time as the Bundschuhbewegung, wooden shoes were painted in
some twenty Danish churches gathered on southeast Funen and the surrounding
islands, shoes that are depicted with a long pointed toe, which itself indicates
rebellion. Shoes that in the imagery of the wall paintings are used by tormentors
and devils. Shoes that can be found in decorations that deviate very much from
traditional wall painting iconography. Faced with possible criticism from
ecclesiastical authorities, the local peasants could answer: this is just an ordinary
wooden shoe. But with René Magritte they would think: this is not a wooden
shoe. With van Gogh they would feel that an image of a wooden shoe contains
an infinity of meanings. Mutually they would know that the configurations on
the vaults represent a sign of gathering, solidarity, resistance, and rebellion.
Martin Luther pinned his theses on church-doors as a natural place to disclose
written communication. Inside the churches it was just as natural to communi
cate ideas and messages. The Painter of Wooden Shoes was not merely one
71 O. Bloch and W. von Wartburg, Dictionnaire étymologique de la langue française,

8"' edn (Paris: Presses universitaires de France, 1989). p. 567: A. Greimas. Dictionnaire
de l ’ancien français, 2nJ revised edn (Paris: Larousse. 1980). p. 576. Jérôme Baschet has
kindly given me this information. Paul Robert. Dictionnaire alphabétique et analogique
de la langue française, 2'"1edn. 9 vols (Paris: Le Robert. 1985). VIII, pp. 520-21.
246 A xel B olvic.
psychotic person; as a matter of fact, the decoration of more than twenty

churches during a couple of decades involves several painters. They all have in
common the use of alternative iconography in opposition to traditional religious
imagery. They use the depiction of a wooden shoe with high wedges and long
pointed toes as a common sign.
The present use of the database results in a theory that wooden shoes only
belong to negatively connoted figures and situations, never to artisans’ or
peasants’ daily life. Do the depicted wooden shoes in this specific part of
Denmark at this specific period of European history indicate the parishioners’
wish to take over their local churches? Do we see a visualized rebellion against
the established church? Or is the wooden shoe a sign of resistance directed
towards the land-owning lords?
At the sign o f the ‘Spinning Sow’:
the ‘other’ Chartres and images
o f everyday life o f the medieval street
t MICHAEL CAMILLE
he vast majority of people who visit Chartres, a short day-trip from Paris
T today, do so to see the world-famous Gothic cathedral with its superb

twelfth- and thirteenth-century sculpture and stained glass. Modern
interpreters of this building, beginning with Alphonse Didron in his ‘Icono
graphie Chrétienne’ of 1841, through Emile Mâle and down to the thousands of
bibliographical entries listed in Van der Meulen's recent bibliographical survey,
have made it into a great iconographie machine for the production of meanings.1
Like all machines in the common science-fiction trope, this too becomes
autonomous, not only outliving its creators but also ‘taking over’ the world,
generating new interpretations in its increasing control of the Gothic imaginary.
It has become a ‘summa’ in stone, educating the illiterate masses (today as well
as in the past) in the basic truths of the Christian faith. Chartres Cathedral often
is described as striving to contain everything in one vast encyclopaedia and
microcosm: Old and New Testaments, past and future, body and spirit, virtue
and vice, even Mow' as well as ‘high’ culture. The western complex includes not
only the systematic order of the famous royal portal but also the statue of the ass
playing the lyre carved at the base of the south tower. Michel Foucault argues
that this insatiable yearning to encompass ‘everything’ results in the monument’s
1 Jan Van Der Meulen. Chartres. Sources and Literary Interpretation. A Critical
Biography (Boston: Hall, 1989).
250 M ic h a e l C a m i l l e
t i
Fig. I. Chartres. Distant view of the cathedral from the south east with ‘Basse-
Ville’ in foreground. ©Author
ultimate loss of meaning because of excess, a literal over-production of
signification.2 I would put this slightly differently: I would suggest that the
cathedral manifests the desire to encompass and to structure official exegesis in
opposition to the chaotic instability of the countless unofficial and inde
cipherable meanings that are projected onto it and that proliferate in the profane
: Michel Foucault, Madness and Civilization, trans, by Richard Howard (New York:
Vintage Books, 1973), p. 18.
At the sign of the ‘Spinning S a w ' 251
world it seeks, but fails, to exclude. This is its failure, because the proliferation
of images pressing in from the outside ultimately prevent Chartres Cathedral
from sustaining its isolated and separate place. However, it is also its triumph,
since the boundaries of the sacred extend from its walls out into all spheres of
the socius beyond.
In questioning why we still isolate the sacred from the profane as separate
spheres in medieval culture. I shall be turning things upside down. I shall not
discuss the contamination of the cathedral at the centre by ‘popular’ images from
the outside, which is where one might want to begin with an image like that of
the musical ass. but rather shall place the cathedral itself on the margins. I am
going to talk instead about what I call the ‘other’ Chartres, hoping, in the
process, to blur the image we might have of the cathedral as the spotless
reflecting mirror of medieval society. This ‘other’ Chartres is not where clerics
perform the liturgy and people confess, pray, and contemplate the next world. It
is in this one, where they are born, puke, suck, play, work, eat, drink, defecate,
breathe, sleep, and eventually die. It is the town itself, the Chartres of the
marketplace, of the body, the mouth, the gut, and the anus, the ‘material lower
bodily stratum’, as the Russian scholar Bakhtin termed it.3 What was called the
‘basse ville’ or lower town contained both policed gates open to certain influxes
of goods as well as sluices running out of it, waterways that flushed away the
detritus and pollutants from the town's textile and tanning industries (fig. 1).
These were the sites of entry and exit, the mouth and anus of this social body.
The town in this sense embodied the ‘high’ and ‘low’ sacred/profane metaphor
stretching steeply down below the walled enclosure of the holy city. During the
Middle Ages, Chartres flourished as one of the largest and most active urban
centres in northern France, especially its production and market in wine and
cloth. In addition to the cathedral, the extant medieval architecture of Chartres
includes a number of canonical residences with sculptural decoration that were
inside the cloister, a wall that demarcated the ecclesiastical domain in the town
and that was controlled by the bishop and chapter.4 There are also important
architectural remains from the town itself, most notably a group of wonderfully
preserved timber-framed houses or ‘maison de pans de bois’ built during the
fifteenth century. Timber-framed houses were crucial sites of sculptural
decoration during the Middle Ages, a genre ignored by art historians today who
’ Mikhail Bakhtin, Rabelais and His World. trans, by Hélène Iswolsky (Cambridge,
MA: MIT Press, 1968). pp. 368-A36.
4 For the house opposite the cathedral with carvings of a pair of wrestlers and dice-
players in the lunettes, see Yves Esquieu, Quartier Cathédrale: une cité dans la ville
(Paris: Desclée de Brouwer. 1994). p. 96. For another interesting twelfth-century house,
also probably a canonical residence, with monstrous human and animal figures carved in
the lunettes, see A. Mayeux. ‘Maison du xif siècle à Chartres’. Bulletin Monumental. 79
(1920), 217-22.
252 M ichael C amille
*v n
F/". 2. Chartres, ‘Maison du Saumon © Author

Al lhe sign o f lhe 'Spinning Sow' 253
relegate these sculptures to the status of 'patrimoine folklorique' rather than

admitting them to the status of "art' proper."’ One of these is the ‘escalier de la
Reine Berthe’ containing numerous figurai sculptures, but the one I want to
focus on, just opposite the Hotel de Ville only a stone’s throw from the south
portal of the cathedral, is the building known as the ‘Maison du Saumon’
(fig- 2).
Named after the beautifully carved fish on the central console of the first
storey, this three-storey, five-bay façade is divided by a series of sculpted
figures, many now damaged beyond recognition. At the centre, flanking the
great fish, are the angel and Mary of an Annunciation group much damaged
during the Revolution/’ As is usual in the other fifteenth-century houses I have
examined with numerous sculpted elements, those which survive at Angers,
Bourges, and various towns in Brittany, the sacred personages are placed on the
lower storeys while the higher storeys are reserved for animal and satirical
figures. As well as standing saints there are a number of animals carved on the
façade in addition to the salmon: a butterfly and snail on a vine-leaf, for
example. Most striking of all these wooden carved signs, however, is that on the
far right of the second storey, diagonally adjacent to the watery sign that has
given this house its name, the image of the ‘truie qui file’ or the spinning sow
(fig. 3). The porcine creature is represented as a strange upside-down ‘fileuse’ or
spinner. Lying on her back, she is busy working the spindle, which flies up in
the corner, rather than dangling as it should from the long distaff, held in place
by the belt that encircles the animal’s plump body. This spinning equipment and
even the belt to hold it in place can be paralleled in contemporary images of
human female spinners, but before 1 come to analyse this aspect of the image 1
first want to establish some of the perimeters of pig-perception in fifteenth-
century France.56
5 For French half-timbered houses in general, see: Camille Enlart. Manuel

d’archéologie française, //. Architecture civile et militaire (Paris: Picard. 1904), pp. 185—
88: R. Quenedey, L’Habitation rouennaise (Rouen: Lestringant. 1926: new ed. by
Brionne. 1977); Pierre Garrigou Grandchamp, Demeures médiévales: coeur de la Cité
(Paris: Desclée de Brouwer. 1994). pp. 118-203: Yves Esquieu. ‘La maison médiévale
urbaine en France: état de la recherche'. Bulletin Monumental. 153 (1995), 109-41; and
the most in-depth study yet undertaken of carvings located on houses on one site, by
Daniel Leloup. La Maison urbaine en Trégor aux xve et xvf siècles (Rennes: Éditions
Ouest-France. 1996).
6 Roger Joly, Histoire de Chartres (Roanne: Horvath. 1982). p. 189: see also Archives
de la Commission des Monuments Historiques, ed. by A. M. Baudot and A. Perrault. lit
(Paris. 1920). p. 10.
Fig. 3. Chartres, the ‘Truie qui file ’ on the ‘Maison du Saumon’. ©Author
Fig. 4. The so-called ‘Truie qui file' on the south tower of the cathedral, from
L ’Abbé Bulteau, Monographie de la Cathédrale de Chartres, Il (Chartres, 1888)
At the sign o f the 'Spinning Sow ' 255
Pig in the City
From medieval folktales and modern children’s stories to the recent film ‘Pig in
the City’, the myth of the urbane porker, the swine become socialite, the pig that
leaves the filth of the farm for the swagger of city life, is a compelling myth
articulating the deep-seated nature/culture contradiction. The very use of painted
and sculpted animal images as both house signs and shop signs throughout the
medieval urban space is an important aspect of this trajectory from countryside
to city, making the question of what this pig is doing here especially significant.7
The spinning sow is not a unique image, though this particular configuration
with the sow upside down is, as we shall see, unusual. The image of the ‘truie
qui file’ existed in many towns in medieval France. It is documented as a house
sign at Dijon, Orléans, Rouen, Rheims, Sens, Troyes, and Paris.8 Apart from a
spinning sow carved on a house in Malestroit in Brittany, which will be
discussed later in this paper, this one at Chartres is the only other medieval
example I know—of what was a popular theme in house sculpture—to have
survived to the present day. Compared to a theme such as the ‘ass playing the
lyre’—which, partly because it appears on the cathedral, has been the subject of
numerous articles since the middle of the nineteenth century—there are very few
studies of the spinning sow as an iconographical subject, an example of how we
tend to prioritize ecclesiastical over non-ecclesiastical themes.9
7 S. Roux, Images des bêles urbaines, milieux naturelles, espaces sociaux: Etudes
offertes à Robert Delort (Paris: Presses de l'Université de la Sorbonne, 1997), pp. 229-
40.
s Albert Babeau. Les Vieilles Enseignes de Troyes (Troyes: P. Nouel. 1898). p. 21:
Marie-Camille Palay. Les Enseignes, emblèmes et inscriptions du vieil Orléans (Orléans:
H. Herluison, 1878), pp. 10-48: C. Porée, Histoire des rues et des maisons de Sens (Sens:
Duchemin, 1920); and for Paris see Edouard Fournier, Histoire des Enseignes de Paris
revue et publiée par la Bibliophile Jacob (Paris: E. Denlu, 1884), who discusses the 'truie
qui file’ on p. 115. For house signs in general in the Middle Ages, see E. de la Quérière,
Recherches historiques sur les enseignes de maisons particulières (Paris: E. Didron,
1852) and Jean-Pierre Leguay, La Rue au moyen âge (Rennes: Editions Ouest-France,
1984), pp. 104-10. The best illustrated discussion is Yves Papin, ‘Les Enseignes
médiévales: un langage pittoresque et universel'. Archeologia, 69 ( 1979), 37-42.
9 For the meaning of the ass playing the lyre carved at Chartres, see the articles by Guy
Villette listed in Van der Meulen (as in note 1), nos 710, 2068. 2069. as well as: Helen
Adolf. "The Ass and the Lyre'. Speculum. 2 (1950), 49-57: Martin Vogel. Onos lyras.
Der Esel mit der Leier, Schriftenreihe zu Grundfragen der Musik. 13-14 (Düsseldorf:
Gesellschaft zur Forderung der systematischen Musikwissenschaft. 1973), voi. I. pp. 361 —
64; Alessandro Vitale-Brovarone, 'The Asinus Citharaaedus in the Literary and
Iconographie Tradition of the Middle Ages’. Marche Romane. 28 (1978), 125-29: M. de
Meyer, De Volks-en Kinderprent in de Nederlanden (Antwerp and Amsterdam. 1962), p.
417: Wilfried Schouwink, Der wilde Eber in Cottes Weinberg: Zur Darstellung des
256 M ichael C am ille
On the Chartres house the small sculpted figure of the pig fits perfectly into
the triangular form of the spandrel-shaped bracket, and is quite visible from the
street below, in a way that many marginal sculptures carved in stone on churches
are not. Is there anything like this image on Chartres cathedral? On the base of
the south tower is a group of incongruous animal performers. One of these is the
already-mentioned theme of the ass playing the lyre carved on the south side of
the southwest tower of the cathedral. This musical donkey is based upon
Classical sources, and one of the few scholars to mention the spinning pig at
Chartres has argued that, like the harping ass here, our spinning sow 'is based on
a fable by Phaedrus’.10 The Roman playwright’s fable ’Asinus ad lyram' was
known to medieval authors indirectly through a phrase in Boethius’s
Consolation o f Philosophy but had entered vernacular proverbial discourse
already by the twelfth century. The foolishness theme takes on a more specific
Christian significance at this particular site. As an image of exclusion, it depicts
the incongruity of lay people’s vain attempts to perform the sacred themselves,
for example, to read the Psalter. Paired with the image of the ‘Ane qui vielle’ is a
much-damaged figure of an animal also up on its hind legs performing in human
fashion, which, fascinatingly, many nineteenth-century scholars labelled the
‘truie qui file’ (fig. 4). In 1852 E. de la Quérière in his book Recherches
historiques sur les Enseignes îles maisons particulières, described this as the
very same ‘truie qui file’ known in so many towns in France as a house sign.
Even more influentially, in his 1888 monograph on the cathedral. Marcel J.
Bulteau also described the vestiges of this statue as ‘une quenouille, sa patte
droite tenait le fil', probably projecting back onto the hardly legible statue on the
cathedral an image he knew from the well-known local image in the town." He
saw it as a case of the contamination of the cathedral’s symbolic structure by an
extraneous and, in this case, inappropriate sign. However, here I disagree with
scholars such as Van der Meulen who believe that meaning exists only where it
has been inscribed by its medieval makers: the fact that many people in the
nineteenth century saw and read this cathedral sculpture as the spinning sow,
however erroneous this reading was, makes it part of the cumulative meanings
adhering to the building and is just as relevant as those lingering on from the
Middle Ages. In 1916 the Comte d’Armancourt presented a paper arguing that
Schweins in Literatur und Kunst des Mittelalters (Sigmaringen: Thorbeke, 1985): Niis-
Arvid Bringeus. Volkstümliche Bilderkunde (Munich: Callewy. 1992). pp. 69-72: M.
Pastoureau, 'L'homme et le porc: une histoire symbolique’, in his Couleurs, Images,
Symboles (Paris: Le Léopard d'Or. 1989), pp. 237-82; and Mythologies du porc: Actes
du colloque de Saint-Antoine-!'Abbaye (Isère), ed. by P. Walter (Grenoble: Millon.
1999). p. 263.
10Joly (as in note 6). p. 189.
" L'Abbé Bulteau. Monographie de la Cathédrale de Chartres, il (Chartres: Librairie
R. Sellerei, 1888), pp. 96-97; E. de la Quérière (as in note 8), p. 17.
Al lhe sign o f lhe 'Spinning Sow ' 257
Fig. 5. Chartres Cathedral, North transept, right porch, right outer archivolts,
'December pig killing © James Austin
the so-called ‘sow’ indubitably displayed male genitals and that what was
represented here was in fact another popular animal type: ‘the wise dog which
dances’.*12*The spinning sow then is one of the few things that did not appear
'' Armancourt, Société Archéologique de l'Eure-et-Loire, 13 (1919), 413-16; see also

Van Der Meulen (as in note 1), no. 2052; and Rüdiger Hoyer. Notre Dame de Chartres,
Der Westcomplex: Systematische Grundlagen der Bauarchäoligischen Analyse
(Frankfurt: Lang, 1991). Abb. 66. p. 91.
originally in the stone encyclopaedia of the cathedral. Its lack on the cathedral is
made up for by its appearance in the marketplace.
Writing about medieval marginal sculpture. Nurith Kenaan-Kedar states that
the images found carved on fifteenth-century French houses are taken from the
traditional repertoire of themes already used for centuries in church corbels, like
these figures at Chartres, as well as gargoyles.13 Images such as the spinning
sow, which occasionally appear in church contexts, are not the same, however,
when they are placed on a civic building or house. For they are here linked to the
life of the street rather than to the liturgy. The marketplace, which in Bakhtin’s
words was ‘the centre of all that is unofficial enjoyed a certain extraterritoriality
in a world of official order and official ideology, it always remained ‘with the
peo-ple” \ 14 When pigs appear in church sculpture, which they do quite rarely
compared to other animals, it is usually as part of the calendrical cycle for
December, as they do on the north transept, right porch sculptural archivolts of
Chartres Cathedral (fig. 5). They refer to the sins of greed and lust as first
defined by Clement of Alexandria, when they are shown being ridden by an ape
or playing bagpipes. The pig had a long negative history in Christian art as ‘an
abomination’, its uncleanness defined by Jewish dietary laws and by Christ
casting out demons from a man into swine.15 Most well known is the ‘Judensau’,
carved at a number of cathedrals (but not at Chartres) and later popular in
woodcuts, which shows the Jews suckling on the very creature they view as
unclean.16
At the same time it would be also wrong to argue that the pig has a totally
secular meaning on the house because it appears in a secular place, thereby
separating the spheres of sacred and profane culture that in reality constantly
overlapped. Writing about contemporary southern French communities from an
anthropological perspective, Claudine Fabre-Vassas has written what is the most
anthropologically complex account of the pig as sign in her book The Singular
Beast. She makes the important point that ‘the slaughter of the animal coincides
with the rhythms of the destiny of Christ’, making it a complex and volatile sign
15 Nurith Kenaan-Kedar, Marginal Sculpture in Medieval France: Towards the

Deciphering of an Enigmatic Pictorial Language (Aldershot: Scolar Press. 1995). pp.
134, 139. and 143.
u Bakhtin (as in note 3), pp. 153 ss.
15 See: F. C. Sillar and R. M. Meyler. The Symbolic Pig. An Anthology of Pigs in
Literature and Art (Edinburgh: Oliver and Boyd, 1961): Claude Gaignebet and J.
Dominique Lajoux, L'Art profane et religion populaire au moyen âge (Paris: Presses
universitaires de France. 1985), pp. 98-100: and Milo Kearney, The Role of Swine
Symbolism in Medieva! Culture (Lewiston: Mellen. 1996).
16 See Isaiah Shachar. The Judensau. A Medieval Anti-Jewish Motif and its History
(London: Warburg Institute, 1974).
At the sign o f the 'Spinning Sow ' 259
of seasonal and pseudo-sacred authority in various communities.17*

As signs, animals served a totally different function in the social imaginary
than do their modern counterparts. Also, whereas in the countryside the pig was
‘an important member of the family’, once herded into cities their symbolic
ambivalence turned more towards hate, not only because the pig’s dung was now
a nuisance on the street but also because the pig ‘now stood as a sign of the gross
immiseration of its fellow slum-dwellers’.IS Precisely because they lived among
poorer people, they might occasionally eat unattended babies, for which they
were accused, tried, sometimes tortured, and then executed like human
criminals. The monks of Saint-Geneviève ordered a pig to be publicly burned
alive, and in 1403 another was imprisoned and had to pay room and board like
other prisoners before its execution.19
The anthropologist Edmund Leach, in his study Anthropological Aspects of
Language: Animal Categories and Verbal Abuse, views the pig as a particularly
striking example of what he calls ‘the intermediate category’—both man and
animal, friendly and hostile.20 The pig was associated with the outcast, the Jew,
the prostitute, the leper. So why does it appear here? Peter Stallybrass and Alton
White, in their ambitious attempt to ‘sketch a theory of its ambivalence',
describe the pig as an animal of the threshold, between the country and the city,
the outside and the inside, the human and the animal.
Not only did the pink pigmentation and apparent nakedness of the pig
disturbingly resemble the flesh of European babies (thereby transgressing the
man-animal opposition), but pigs were usually kept in peculiarly close proximity
to the house and fed from the household’s leftovers [. . .]. It is precisely the
‘creature of the threshold’ which becomes the object of fascination. Mixing faeces
and food, human and animal, human skin and animal hide, the household and the
farmyard and the field, the pig also lived to die—unlike sheep, cows and
chickens, the pig was useful only to eat and proverbially, only became valuable
when dead.21
17 Claudine Fabre-Vassas, The Singular Beast: Jews, Christians and the Pig, trans, by
Carol Volk (New York: Columbia University Press, 1997). p. 185.
Is See Robert Delort, ‘Les Animaux dans la ville occidentale à la fin du moyen âge’, in
Villes, bonnes villes, cités et capitales: études d'histoire urbaine (xne-xvn siècle).
Mélanges offerts à Bernard Chevalier, ed. by Monique Bourin (Tours: Presses de
l'Université. 1989), p. 345; F. Harrison. Strange Land: The Countryside: Myth and
Reality (London: Sidgwick & Jackson. 1982), p. 61.
n E. P. Evans, The Criminal Prosecution and Punishment of Animals (London:
Heinemann, 1906), pp. 140-44.
20 Edmund Leach. ‘Anthropological Aspects of Language: Animal Categories and
Verbal Abuse’, in New Directions in the Study of Language, ed. by E. H. Lenneberg
(Cambridge, MA: MIT Press. 1964), pp. 23-63.
21 Peter Stallybrass and Allon White. The Politics and Poetics of Transgression
260 M ichael Cam ille
They describe how 'the pig was duly celebrated as well as reviled’, becoming
'the symbolic analogy of scapegoated groups and demonized "others" especially
the Jews who were forced in some cities to perform races with pigs and be
stoned’. The pig becomes ‘a focus of what we call displaced abjection, the
process whereby the "low” social groups turn their figurative and actual power,
not against those in authority, but against those who are even "lower” (women.
Jews, animals; particularly cats and pigs)’.22 By eliding the Jew with the pig, the
carnival crowd was producing a grotesque hybridization of terms expressly
antithetical to each other according to the dietary rules of their victims, who. at
carnival-time, would be excluded from the great pig-feast. For between Decem
ber and March the killing of the family pig was a time of enormous rejoicing and
festivity, incorporated into the liturgy of the ‘cochonailles’ when the community
came together for the sacrificial death of the creature. The most important cluster
of associations for the purposes—visible in calendar pages of many Flemish
manuscripts showing the violent death of the pig in December, its hot blood
caught by an old woman, and which kept folk alive through the winter—is the
association between pigs and women.
Spinning P igs a n d W omen
The gender of this pig is made visible neither through her body nor through any
teats, which in some examples show her suckling her young, but through what
she does. Spinning was exclusively the work of women, practised by duchess
and peasant alike and part of the traditional iconography of gender
differentiation throughout the Middle Ages. While Adam delved out in the
fields, 'Eve span' in the home, even while she tended her children.23 Spinning
their yarn simultaneously, they spun tales, and like the fates, time itself became
associated with their unwinding actions. This feminine cultural space is
described in the fifteenth-century French text Les Evangiles des Quenouilles or
'Distaff Gospels’, a text that sets up exactly the same pseudo-sacred resonance
that is at work in our image. The distaff was the woman’s sceptre but also the
weapon used to beat henpecked (another animal word used to articulate gender
(Ithaca: Cornell University Press. 1986). p. 47.

- Ibid. p. 53.
23 See the excellent discussion by Francesca Canade Sautman, 'La maturation du fil’,
in her La Religion du quotidien: rites et croyances populaires de la fin du moyen âge
(Florence: Olschki. 1995), pp. 37-43; also Fouquart de Cambrai, Antoine du Val. and
Jean d'Arras (attrib. to). Les Évangiles des Quenouilles, ed. by M. Jeay (Paris: Vrin-
Presses de l'Université de Montréal. 1985): and Frances M. Biscoglo. ‘Unspun Heroes:
Iconography of the Spinning Woman in the Middle Ages’, Journal of Medieval and
Renaissance Studies. 25.2 (1995). 163-76.
At the sign o f the Spinning Sow 261
Fig. 6. Chartres Cathedral, North transept, ¡eft porch, left outer archivons,
'Women carding and spinning’. © James Austin
Fig. 7. ‘Woman with distaff defecating1. Wooden ‘aisselier’ from an apothe

cary’s shop. Late fifteenth century, Nantes, Musée Dobrée. ©Author
relations) husbands. Therefore, when an image shows a man bearing the distaff it
was of course to turn the world upside-down and give women rule. In practical
rather than symbolic terms, the distaff (in French the ‘quenouille’) held the mass
of raw wool fibres on top which had to be twisted and manipulated by the
dextrous fingers of the female worker in order to produce yarn, wrapped around
the dangling spindle (or ‘fuseau- in French). The pig’s cloven hoof of course is
incapable of any such movement, a point emphasized in its delineation in the
Chartres woodcarving. On the north portal of Chartres cathedral, women are
carved in stone, spinning as exemplars of the active, as opposed to the contemp
lative, life; the latter are shown as women holding open books (fig. 6). This
association was suggested by the liturgy.24 In the realm of the street, the distaff
24 For the spinning figures on the Chartres Cathedral north portal, see Adolf
Katzenellenbogen, The Sculptural Programs o f Chartres Cathedral (Baltimore: Johns
Al the sign of the Spinning Sow' 263
Fig. 8. ‘Spinning sow’. Wood. Malestroit, Brittany, fifteenth century. ©Author

appears as the woman’s emblem in the lowest as well as the highest of realms:
thus as she is carved squatting to defecate in a carving once placed on an
apothecary’s shop now in the Musée Dobré at Nantes (fig. l ) f
The other extant example of the spinning pig theme that I know of appears on
a house at Malestroit in Brittany (fig. 8). Here the sow wears a similar belt but
stands in an upright position like a human being. The spinning instruments being
held in proper gravitational force allow the weight of the spindle to pull down*
Hopkins University Press, 1959), p. 74.

" Dominique Costa. Catalogue: Fascicule /, du XII'' au x v f siècle. Musée Dobrée
(Nantes: Musée Dobrée, 1959), no. 192. p. 95.
the long twisted thread twisted by the fingers from the flax. She seems to be
chewing on the flax held on top of her distaff, as the proverbial 'greedy pig’. In
her excellent study of popular culture in late medieval France, which she calls 'la
religion du quotidien', Francesca Sautman describes the image of the spinning
sow in late medieval house signs as 'a sort of distorted spinner [. . .] in its place
the porcine female spinner was impure, a carrier of leprosy like the menstruating
woman’. She is the antithesis to the Virgin Mary carved below. The same
contrast can be found on the page of a Rouen Book of Hours in which a prayer
to the Virgin has a spinning sow painted in the lower margin.26 In a fifteenth-
century French herbal, the mother sow spins while a group of monkeys snatches
her infant from its cradle (fig. 9). Having a world-turned-upside-down motif on
the outer corner of the house façade is paralleled in a beautiful fifteenth-century
house at Bourges, where the Annunciation is juxtaposed with a tiny inverted
image of Marcoul the trickster showing his anus in a sign of male sexual and
social power.2728At Chartres the contrast is even more explicit between Mary and
the porcine Eve, spinning her sickness and sexual sins in a morality that is social
rather than ecclesiastical in nature. She is an antithesis to the Virgin Mary,
whose robe-relic was the centre of the pilgrimage trade of the cathedral, a kind
of gross anti-Virgin.
As Fabre-Vassas makes evident, the image of the sow had sexual overtones as
well:
The pig, like the child, thus lives in the maternal aura [. . .] the sow is the most
irregular of females. The time is so short between its rhythmic cycles that [. . .) it
forms the basis for the peasant's wariness: 'The sow is in heat all the time’.2s
In Attic comedy, the pig is associated with the prostitute who was called a 'pig
merchant'. The animal of the goddess Venus throughout the Middle Ages and
the prostitutes of Southwark in London operated with a sign of a boar’s head
hung in front of their establishment. It is another chapter of my book, but this
French stone sculpture now preserved at Wellesley College in the United States,
and called ‘Pigwoman with dog’ in the catalogue of Gothic sculpture in
American collections, is one of a small number of fourteenth-century brothel
signs.29 According to Varro the porcus was a ‘nursery word’ used by women to
26 Paris, Bibliothèque nationale de France, ms. lat. 1178 fol. 35'. 'Sancta Maria
intercede pro nobis’, cited in Sautman (as in note 23). p. 8.
2' Michael Camille, 'Adam's House at Angers: Sculpture. Signs and Contrasts on the
Medieval Street’, in Konstraste im AlItag des Mittelalters, cd. by Gerhard Jaritz (Vienna:
Verlag der österreichischen Akademie der Wissenschaften. 2000). pp. 143-78.
28Vassas (as in note 17). p. 108.
Dorothy Gillerman. Gothic Sculpture in America, l: The New England Museums
(New York: Garland, 1989), no. 195. pp. 250-51.
Al lhe sign o f lhe 'Spinning Sow' 265
Fig. 9. ‘Spinning sow and monkeys', Livre des simples médecines. Paris
Bibliothèque nationale de France, ms. fr. 9136. fol. 1. Late fifteenth century. ©
Bibliothèque nationale de France, Paris
describe the pudenda of little girls. This woman, a pig from the waist down,
offers the newcomer a drink and is being sucked in front, not by a dog but by
another pig. Within the delicate Gothic gestures, this is a deeply misogynistic
Gothic image.
Spinning was likewise loaded with erotic possibilities. Sautman describes
how:
although in the field of labour, spinning is an exclusively female task, its sexual
symbolism was divided between the function of the ‘quenouille’ or ‘distaff
which was ‘female’ and the 'fuseau' or spindle, which is masculine both as a noun
and symbolically as that which the woman fondles with her hand. The act and its
apparatus were implicated in a hot-bed of sexual double-entendres.
A contemporary poem among the ’Adevineux Amoureaux’ describes a sug
gestive interchange between a man and a woman in which the lady appropriates
the male ‘fuseau’ and he the female ‘distaff, so that each offers the other the
sexual organ they both desire. She offers her silver ‘spindle’: ‘Je vous vens le
fuseau d’argent? Vous avez le corps bel et gant’. He replies ‘Lady, I will offer
you the distaff/1 like to manipulate women, although especially those who have
hairy nuts’ [‘Dame je vous vens la fusée/ Bien me semblez femme rusée/
Ailleurs avez escaillié noix’].30 To spin also signified to create or give birth to a
human being, adding a register of positive prosperity to the image.
“ Sautman (as in note 23). pp. 41-42.

266 M ichael Ca m ille
Like women, pigs were both celebrated and reviled in medieval culture.
Although at Chartres Cathedral women were represented in images as the
daughters of Eve, forever consigned to be secondary and sinful and always in
danger of corrupting their own bodies and those of men with their sexual
appetites, women in the city of Chartres played a considerable economic role.
While they could not, of course, become canons in the cathedral, they could
become leaders of important mercantile confraternities and groups such as the
textile workers. Although marginal to the Church, women were fundamental to
the growing market economy, as producers and consumers. In a recent
discussion of the house and home as the domestic sphere of women,
architectural historian Mark Wigley makes much of the metaphor of the body as
a building. He cites the fourteenth-century Parisian physician Henri de
Mondevile, who describes the body as a house, ‘the house of the soul, which like
any house can only be maintained as such by constant surveillance of its
openings. The woman’s body is seen as an inadequate enclosure because its
boundaries are convoluted [.. .] turned inside-out’.31 Peter Stallybrass has argued
that ‘the surveillance of women concentrated upon three specific areas: the
mouth, chastity and the threshold of the house. These three areas were frequently
collapsed into one another’.32 Entrances and interiors are ‘coded’ female,
especially churches, and even whole cities.33 A wooden carving in the Stadhuis
in Damme in Flanders shows a man looking up the anus of a vast sow, which is
in fact a positive pun on the name of the waterway, the ‘Zwyn’, which connected
Bruges to the sea.34 If watery connections, as well as sacred and enclosed space,
were coded feminine, as a number of studies have shown, the unexplored area of
the street, I would argue, was masculine. However, as old prints depicting the
Maison du Saumon in the early nineteenth century show, the space before this
particular house often was filled with women, since it was the site of the
flourishing fish market. In the modem imagination, fish, for a whole set of
31 Mark Wigley, 'The Housing of Gender', in Sexuality and Space, ed. by Beatriz
Colomina (New York: Princeton Architectural Press. 1992), pp. 337-55. For cities and
gender, see Elizabeth Grosz, Space. Time and Perversion: Essays on the Politics of
Bodies (London and New York: Routledge, 1995). pp. 103-11.
’2 Peter Stallybrass, ‘Patriarchal Territories: The Body Enclosed’, in Rewriting the
Renaissance. The Discourses of Sexual Difference in Early Modern Europe, ed. by M. W.
Ferguson (Chicago: University of Chicago Press, 1986), p. 126.
33 A fundamental study is F. Choay, ‘La ville et le domaine bâti comme corps-.
Nouvelle revue de psychanalyse, 8 (1974), 239-52; but see also the fascinating
archaeological study by Roberta Gilchrist, 'Medieval Bodies in the Material World:
Gender, Stigma and the Body’, in Framing Medieval Bodies, ed. by Sara Kay and Miri
Rubin (Manchester: Manchester University Press, 1994), pp. 43-61.
L. Maeterlinck. Le Genre satirique fantastique et licencieux dans la sculpture
flamande et wallonne (Paris : J. Schemit. 1910). p. 115 [fig. 68].
Al the sign o f lhe 'Spinning Sow ' 267
complex reasons, has become associated with women’s bodies, but the gender
mapping was different during the Middle Ages.
The Pig, the Fish a n d the Virgin
Medieval signs are linked to social functions. The reason for having the fish
carved on the house (fig. 10) is not symbolic in the same way as the fish carved
on the calendar of the north portal of Chartres Cathedral, a sign of the zodiac
(fig. 11). This also was a sign linked to temporality, though not to the cycles of
the year but to the actual selling of goods. The house was on the "rue du Poisson-
Doux’, which was a centre for the selling of fresh-water fish throughout the
Middle Ages. It was where merchants who brought mackerel and salmon from
far afield sold their splendid specimens.35 Compared to house signs, which were,
for the most part, arbitrary in their signification, shop signs had to announce
things for sale or services offered. Originally, merchants suspended the actual
objects produced as signs: a loaf of bread for a baker, a ‘plat à barbe’ for the
barber, or a pot on the end of a long pole for a potter. Throughout Europe,
branches of vegetation had long been wound in a circle and suspended on a pole
to indicate places where wine was sold, such as that visible in a fifteenth-century
Parisian miniature.36 These natural objects of decomposable material gradually
came to be replaced by more permanent images. Shops were structured as an
opening, the shutters coming down to form the counter or ‘etal’ while the top
shutter or ‘auvent’ protected goods from sun and rain. The lack of literacy
among the urban population is not, in my view, the main reason for names not
being used. Instead, the way in which signs were organized articulated not
personal identity but occupation and social position. Signs of métiers could be
quasi-religious, such as roast-meat sellers using the sign of St Lawrence’s grill
and St Martin for inns, but most were based upon the similitude of service or
goods sold, such as a shoe for a cobbler or a fish for a fishmonger as here.37 The
sign of the fish at Chartres would have been directly above barrels or panniers of
real fish, rotting, smelling, and shining—perhaps some still squirming.
,s Geneviève Aclocque. Les Corporations, l ’industrie et le commerce à Chartres du

XI' siècle a la Révolution (Paris: Picard. 1917), p. 181.
36 For house signs, see Michael Camille, ‘Signs in the City: Place. Power and Public
Fantasy in Medieval Paris', in Medieval Practices of Space, ed. by M. Barbara Hanawalt
and Michael Kobialka (Minneapolis: University of Minnesota Press, 2000), pp. 1-37; and
the documents listed in Sautman (as in note 23), pp. 4-5: and Pierre Turpin. “‘L'Escoive’"
ou "escouve” dans le language juridique et de la toponymie du Nord de la France’. Revue
du Nord, 22(1936). 204-16.
37 For St Lawrence, see Sautman (as in note 23), p. 3: and for a stone sign of a cobbler
in Cordes, see Georges Duby. Histoire de ta France urbaine, il: la ville médiévale (Paris:
Seuil. 1980). p. 255?
Fig. 10. Chartres. ‘Salmon' from the ‘Maison du Saumon'. © Author

Something unusual about Chartres, which is significant for our image, is that
in the fourteenth and fifteenth centuries, the selling of fish had been entirely
concentrated in the hands of butchers. In 1147, the count of Chartres, Thibaut,
had granted fish-sellers in Chartres a monopoly, but there was, as in all towns,
great competition between the purveyors of Lenten and of non-Lenten fare.3S
Social conflicts between groups like meat-sellers and fish-sellers sometimes
erupted in actual violence, and not only during Lenten and carnival celebrations
when mock-battles between the two groups took place. The sign of the
fishmongers is also multivalent and, like the sow, is linked to calendrical cycles
and rituals in popular culture. The month of April (of course marked temporally
by the sign of the fish, Pisces) also marked the resurrection of Christ, and the
Lenten period leading up to it, being a time when fish as prescribed, was also the
period of the ‘Poisson d’avril’. This festive merrymaking in which young girls
could choose new boyfriends has been linked etymologically to the word
‘passion’, since this is the period of spring rebirth and sexual growth. Many
sixteenth-century texts play on the male ‘mackerel’, and one describes a young
girl losing her virginity to a ‘franc poison d'apvril’.**39 Perhaps this explains why
this splendid scaly fish does not hang head down but has its head erect and
upright in contrast to the pig that lies on its back. Once we realize that the fish is
just as much an erotic sign as a sacred one, and that, like the adjacent sow, it
articulates ritual and not just symbolic associations, we are in a position to
understand just how rich a system of meanings was woven in the urban
imaginary of the medieval street.
Although known in later sources as the ‘maison du Saumon’, this house in
certain times or for particular communities took on the name of the ‘Truie qui
file’. We know of a number of houses that went under this name. The names of
houses can change. One either suspects that the same visual sign, as it was worn
and battered, perhaps less legible, was given a different designation. In this way
a Parisian house called ‘des trois chandeliers’ in 1257 received a new sign till
1423 when it became the ‘Cheval a louer’; but when it was referred to as ‘des
Sagittaires’ in 1509 it may have been the same equestrian image which evoked a
different name. The same is the case with the ‘chasteT (1369) getting a new coat
of paint and becoming the ‘Château d’or’ in 1613.40 The proliferation of signs
also meant that there was more than one sign of one subject. Paris had at last two
houses of the spinning sow in the fifteenth century. A ‘maison de lTmaige Saint
Jullien’ existed on the Rue de la Juiverie, as well as on the rue Galande on the
,s See Aclocque, Les corporations, p. 128; C. Billot, Chartres ii la fin du Moyen Age
(Paris: EHESS. 1987). p. 134.
39Sautman (as in note 23). p. 57.
40These examples are taken from the study of Parisian house signs by Adolphe Berty.
‘Trois Ilots de la Cité’. Etudes archéologiques. 12 (1855), 5-6.
At the sign o f the 'Spinning Sow' 271
left bank. His statue also graced the portal of the chapel of the hôpital Saint-
Julien-des Ménétriers, where he served to embody the boundaries of a sanctified
space. By contrast, the saint on the house stood in a different relation both to his
divine prototype and to the viewer, who came towards it, not in pious contem
plation. but crossed its path in everyday banal experience.41 Such signs became
the property of those they protected. Cities were divided according to different
trades and their popular saint protectors. Signs like those of St Catherine or St
Julian functioned in a totally different way in the space of the street than they
would in a church. The most common sign in all medieval cities was that of the
Virgin Mary. Not only was her image standing at every street comer but her
carved figure also appeared as in this example integrated into scenes like the
Annunciation. We should not regard such house signs as any less sacred than
those images carved in wood on altars in churches. In 1418 a drunken Swiss
soldier was not only publicly executed for attacking a corner statue of the Virgin
in Paris but also burnt in effigy on the anniversary for years after.42*
Images, art historians have realized in recent years, are not static representa
tions of a pre-existing reality. They partake in that reality and often help
construct it.4-' In terms of religious images, the liturgy is important, but for our
image too we can discern in the scanty records of pig-human relations a series of
rituals associated with this type of image. An example, once at no. 24 of the
Marché aux Poirées in the Les Halles area of Paris, was an image of the spinning
sow which was described as ‘remarkable and more famous through the follies
that the boys from the nearby shops performed there during Lent’. The source
for this anecdote is Henri Sauval’s Histoire et Recherches des Antiquités de la
ville de Paris (1724), in which he goes on to describe it as ‘un reste du
paganisme’ or relic of paganism.44 This image of the sow was like that at
Chartres, located on the rue des Poirées aux Halles, that is, on the building
associated not with meat but with root vegetables. The fact that this ritual took
place during Lent, the period of strict abstinence from any meat, is highly
significant and chimes with those Lenten festivities surrounding the pig
41 See Camille (as in note 36).

Edouard Fournier, Chroniques et Légendes des rues de Paris (Paris: Dentu, 1893),
pp. 364-65; illustrated in a 1633 engraving in Yvan Christ, Eglises parisiennes actuelles
et disparues (Paris: Éditions Tel. 1947). p. 96: for this theme see also Edward Muir. ‘The
Virgin on the Street Corner: The Place of the Sacred in Italian Cities’, in Religion and
Culture in the Renaissance and Reformation, ed. by Steven E. Ozment, Sixteenth-
Century Essays and Studies 11 (Kirksville, MO: Sixteenth-Century Journal Publishers,
1990). pp. 24-40.
" Michael Camille, Mirror in Parchment: The Luttrell Psalter and the Making of
Medieval England (Chicago: University of Chicago Press, 1998).
4 Henry Sauvai. Histoire et Recherches des Antiquités de la ville de Paris (Paris,
1724). p. 57.
272 M ichael C a m ille
described in Fabre-Vassas. These included ritual battles between Fishmongers

and butchers on the morning of Holy Saturday.4’
However, this image is more than a relic of paganism. It is known in other
sources from the medieval period as one that took different forms in different
communities. All involved the young men or ‘bacheliers’ of the community
acting out a special semi-erotic relationship with the pig, either before or after its
slaughter, and its being handed over to the women for the making of sausages, as
a liminal sign of both danger and fecundity in the yearly cycle. These signs are
folded into complex rituals of the sacred but they are primarily signs of
commerce. An essential question of the Chartres image is why is the sow lying
on its back? Is the pig dead? As soon as it is killed and the women collect the
blood to make the blood sausages, it is scraped and hung upside down, and thus
begins a complex of rituals dividing men and women, in which in some
communities the youths steal the pig and play ‘nocturnal games’ with it.
pretending to hide it from the rest of the population. A creature divided, the pig
incarnates the sins of lechery and gluttony; demons take up residence within it.
At the same time it is Christian flesh, endowed with a soul of blood, called upon
to appear at the meals of Christmas and Easter.4 546
What all this suggests is that just like sacred images, these signs could also
change over time as well as being linked to certain temporal rhythms and rituals.
There may have been a play on words here as occurs in many medieval house
signs such as ‘Au Lion d'or’—‘Au lit on Dort’. The verb ‘filer’ could indicate a
pig doing something far more pig-like than spinning. A second example of the
spinning pig known in Paris was also in a significantly charged mercantile site at
no. 134 Rue Saint Antoine. This one is very well preserved in the Musée
Carnavalet and probably dates from the early seventeenth century (fig. 12).47 A
low relief in terra cotta and wood, it was placed on the lintel of the first storey of
a charcutier’s shop. It shows the sow standing on her hind legs and with a belt
around her waist to hold the distaff, just as in the Chartres example, although she
manages to spin while at the same time suckle three piglets. This too has its
legends, one of which relates it to a series of events involving an entertainer who
is supposed to have dressed up an animal in women’s clothing, an act that
45Vassas (as in note 17), p. 243.

4,1 For the role of youth groups in extending the rites of the countryside into an urban
setting see Robert Muchembled. Popular Cullare and Elite Cullare in France 1400-
1750, trans, by Lydia Cochrane (Baton Rouge: Louisiana State University Press. 1986),
pp. 165-69: and Nicole Pellegrin. "Les Bachelleries: organisations et fêtes de la jeunesse
dans le Centre-Ouest XVe—xviif siècles’. Mémoires de la Société des Antiquaires de
l'Ouest 15 (1982), 73-86.
47 Jean-Pierre Willesme. Enseignes du Musée Carnavalet—Histoire de Paris (Paris:
Musée Carnavalet. 1996). no. 123. p. 79. which also describes the previous example from
the Marché aux Poirées.
Ar the sign o f the 'Spinning S o u ' 273
Fig. 12. 'À la Truie qui file ’ sign o f a charcutier. Early seventeenth century.
Paris. Musée Carnavalet. © Musée Carnavalet
eventually got him accused of witchcraft and burnt, along with his pig, at the
stake.48 This evidence also points to another register of negative associations of
the pig with witchcraft: the capacities of women both to bewitch healthy
valuable pigs with a glance of envy and to use them as demonic servants.
While the cathedral contains in its encyclopaedic ‘summa’ the various trades
in the famous stained glass windows, where the fishmongers and indeed butchers
both have their place, these are presented in narrative forms controlled by
ecclesiastical prerogatives. As the late Jane Welch-Williams showed so
brilliantly, they were not offerings made by the guilds to their church but signs
of the appropriation of the mercantile by the Bishop and Chapter.49 The signs we
have been examining come from the people themselves. At first, it might seem
in contrast to the ecclesiastical representations of the trades, which are clearly
sacred signs. While this is true, it would not follow to argue, as scholars have
since Bakhtin, that the culture of the Church was hegemonic and this mercantile
world outside was free. We must resist the tendency to romanticize 'popular' art
as the free-expression of the ‘folk’. As Richard Sennet argues, ‘The street bore
the imprint of aggressive assertion, then, it was the space left over after people
had asserted their rights and powers’.50 The ‘truie qui file’, while being an image
‘of the people’ in Bakhtin’s rather idealized sense of the term, is no less
repressive, ideologically charged, and tyrannizing than the ass playing the lyre.
It locates women in a particular place in the imaginary social order as ruthlessly
as any sculpture of the vice of ‘luxuria’, but it does so in the space not of the
sacred but of the street, with all its raucous noise and clamour. All this suggests
that the image of the spinning sow held an ambivalent attraction for the people
coming to the fish market at Chartres. It would have been, in a way. an
antithetical image, embodying a series of both positive and negative associa
tions, alimentary, social, and sexual many of them dealing with the status of
women. If the cathedral represented her in the dichotomy of either Eve or Mary,
this other building shelters a far more unstable and shifting series of signs.
For medieval people, stone was dead whereas wood was thought to be alive.51
JS See: Fournier (as in note 8), p. 92; Le Marais, mythe et réalité, ed. by J.-P. Babeion
(Paris: Picard. 1987). p. 255. who describes how Tenseigne pouvait venir d’une rumeur
selon laquelle un montreur avait dressé une truie à filer, mais l'acte parut tellement
extraordinaire qu’on le brûla avec sa truie Place de Grève, sous l'accusation de
sorcellerie’.
49 Jane Welch-Williams, Bread. Wine and Money: The Windows of the Trades at
Chartres Cathedral (Chicago: University of Chicago Press, 1993).
50 Richard Sennett, Flesh and Stone: The Body and the City in Western Civilization
(New York: Faber. 1994). p. 193: and Jean-Pierre Leguay (as in note 8). pp. 98-110.
51 M. Pastoureau, 'Introduction de la symbolique médiévale du bois’, in his L'Arbre:
Histoire naturelle et symbolique de l'arbre, du bois et du fruit au moyen âge (Paris: Le
Léopard d'Or, 1993), p. 173.
Al ¡he sign o f the 'Spinning Sow ’ 275
Like the human body, it grew old and its flesh split and withered (notice the
vertical split in our example) along with its significations. This sign is one of a
number of representations carved in wood that I am working on for a larger
project about this forgotten genre of street art in medieval France, including an
enigmatic one from a now-demolished house once on the main street in Tours
showing two women castrating a hybrid man (it was originally the corner post of
a horse gelder’s establishment), which has powerful resonances of gender. This
should not surprise us. After all, the street was always a site of transaction and
conflict rather than of transcendence and resolve. For this very reason, the
images found on the medieval street refuse that myth of plenitude organized by
the Church’s system of signs. Always fractured, always spilling out into more
multiple registers of meaning than can be contained under one rubric, the
meaning of any street sign cannot be contained by a written text or verbal Logos.
Unlike the cathedral, whose thousands of images are, by contrast, always in
danger of multiplying into meaninglessness precisely because they are so over
determined, so ruled by the thrall of textuality, the more shifting significations of
the street survive and prosper in the human imaginary through this flexibility of
signification. Perhaps this is also why we in modernity have forgotten them and
prefer to resurrect the eternal Logos from the dead stones of the cathedral rather
than feel the pulse of everyday life through the living wood of the house.
In her analysis of the role of proverbs in sixteenth-century French culture,
Natalie Zemon-Davis notes that they are often
wide-ranging in metaphor and tone, inconsistent with one another and flexible
| . ..], available for many different circumstances and for recommending different
courses of action. Indeed, the meaning of any individual proverb might vary with
the situation to which it was applied and the status of the speaker and listener.52
In the same way, the spinning sow as a ‘proverbial’ image functioned differently
in the lexical and imaginary register than an ‘official’ image on the church’s far
more textualized apparatus. The ‘Truie qui file’ marked sites of multiple and
shifting identities, places, and communities, not imposed from above but
articulated from somewhere within the teeming multiplicity of the body politic
itself.
There are enormous difficulties and dangers involved in any analysis and
study of this material. With fewer than one percent of the images today extant of
the hundreds that, as we know from documents once filled these urban spaces, a
single image on one building like the 'Maison du Saumon’ can too easily fill the
aporia and become a kind of profane cathedral of the marketplace. A single
image, like the spinning sow. comes to stand in for the hundreds of lost images
that once surrounded it. Yet in terms of the issues raised in this present volume,
52 Natalie Zemon-Davis. 'Proverbial Wisdom and Popular Errors’, in her Society and
Culture in Early Modern France (Stanford: Stanford University Press, 1975), p. 243.
what makes this material challenging for the researcher is that it demands a
totally different Art History, flexible enough and anthropological enough to
realize that visual signs work in history on A more complex level than traditional
textual-iconographic and formal-aesthetic categories. Because of these difficul
ties, I have purposely avoided giving one uniform iconographie ‘reading’ of the
image, to in any way limit the spilling out of meanings associated with the
spinning sow in Chartres. That would be to undermine what I believe was its
fruitful ambivalence as a sign on the street. Its meanings will not be better
understood simply by using the Church Fathers or classical sources. Rather its
context has to be sought within the complex notion of ‘everyday life’. While
some medieval historians—notably the Institut für Realienkunde at Krems,
along with the French Annales school, notably the work of Jacques Le Goff and
Jean-Claude Schmitt—have usefully incorporated this concept into their work,
the Anglo-American tradition of Medieval Studies in which I work is deeply
suspicious and dismissive of terms such as everyday life, popular, and even the
term folklore.51 While every stone of Chartres has been or will be decoded for its
social, political, and historical significance, wooden ‘vernacular’ ones like the
spinning sow only a few hundred yards away are left to rot heuristically. as
indecipherable emblems of an irrecoverable popular.
Although I have been contrasting Chartres cathedral with what I have called
the ‘other’ Chartres, what I hope to have shown is that these two sites are not
mutually exclusive—one sacred the other profane, one a site of ritual and the
other of the everyday. Images in both realms are highly ritualized products of
culture whose meaning changes over time and is sometimes irrecoverably lost.
The spinning pig, whether it articulates the power of women in rites of sexuality
and fertility or their dangerous status as unclean and malefic, does so in relation
not only to the Virgin Mary carved on the same house but also to all the images
of the virgins that inhabited the city, including the cathedral. Where is this pig in
the medieval imaginary of Chartres, excluded as it is from the salvific economy
of the cathedral and the many other churches in the town, marginal as it is to this
very house, but so crucial an aspect of people’s lives?5
55 Jacques Le Goff, Time. Work and Culture in the Middle Ages, trans, by Arthur
Goldhammer (Chicago: University of Chicago Press, 1980); Aron Gurevich. Medieval
Popular Culture: Problems of Belief and Perception (Cambridge: Cambridge University
Press. 1988); Peter Burke. "Popular Culture Reconsidered', in Mensch und Objekt im
Mittelalter und in der Frühen Neuzeit, ed. by Gerhard Jaritz (Vienna: Verlag der
österreichischen Akademie der Wissenschaft. 1990). pp. 181-93; and the essays collected
in Symbole des Alltags Alltag der Symbole: Festschrift für Harry Kühnei zum 65.
Geburtstag, ed. by G. Blaschnitz, H. Hundsbichler, G. Jaritz. and E. Vavra (Graz:
Akademische Druck. 1992). For the sceptical American position, see John Van Engen.
"The Christian Middle Ages as an Historiographical Problem". American Historical
Review 91 (1986). 519-52.
Les marges à drôleries des
manuscrits gothiques:
problèmes de m éthode
J ean W irth
ans la première moitié du Xllle siècle se met en place un nouveau décor
D du manuscrit enluminé, la marge à drôleries. Des motifs profanes et

humoristiques, très souvent animaliers, se multiplient à l’extérieur du
texte et de son illustration, tantôt cantonnés dans des antennes émergeant de
i’initiale, tantôt posés sur ces antennes ou flottant dans le bas de page. Après une
série de tentatives, leur présentation se codifie vers 1250 dans les ateliers
parisiens, en épousant une mise en page solidement organisée à partir de
l’initiale. Elle évolue et se diversifie en restant très structurée dans les ateliers
anglais, dans les villes du nord de la France et dans les Flandres. Vers 1300, le
genre est adopté dans le sud de la France, en Espagne, en Allemagne et en Italie.
A partir du milieu du XlVe siècle, les feuillages prolifèrent dans les marges des
manuscrits parisiens et prennent la place des drôleries. Le genre est passé de
mode. 11 connaîtra cependant des revivais successifs jusqu’à la Renaissance.
La thématique des drôleries est d’abord empruntée à la chasse, à la danse, aux
jeux, aux plaisirs galants; elle prend de plus en plus une dimension satirique
incluant l’obscénité, l’anticléricalisme et même le blasphème, quel que soit par
ailleurs l’ouvrage décoré. La majorité des manuscrits à drôlerie sont des psautiers
et des livres d'heures, ce qui laisse tout d’abord perplexe. Face à l'idée préconçue
d’un Moyen Age confit en dévotion, dominé par des inquisiteurs qui auraient
surveillé méticuleusement la production des livres, baignant nolens volens dans
l'unanimisme d'une foi profonde et naïve, le phénomène est incompréhensible,
mais tout compte fait secondaire. La plupart des chercheurs se sont longtemps
278 Jean Wirth
contentés d’une remarque rapide pour signaler la présence de drôleries dans les
manuscrits. Pourtant, d’autres domaines artistiques posaient des problèmes
comparables et attiraient de temps à autre l'attention, ainsi l’oeuvre de Jérôme
Bosch ou encore les miséricordes des stalles, tandis que la littérature en langue
vulgaire donnait de la sensibilité médiévale une idée tout aussi contrastée, si l’on
tenait compte d’oeuvres comme le Roman de Renaî t ou les fabliaux.
La redécouverte des drôleries médiévales commence au XIXe siècle. La
caricature était alors un art de prédilection, de sorte que Wright, Champfleury,
puis Maeterlinck se mirent à en faire l’histoire, avec une évidente sympathie
pour l’humour médiéval.1 Avec Emile Mâle commence timidement la discussion
scientifique sur les marges à drôleries.2 Après avoir réfuté les explications
symboliques du décor animalier des églises gothiques données par Auber, Félicie
d’Ayzac et même parfois par le Père Cahier, il introduit les drôleries des
manuscrits liturgiques, pour réfuter l’objection que ces sujets n’auraient pas été
tolérés par le clergé s’ils n’avaient eu un sens profond. Décrivant ensuite
sommairement les marges de deux livres de prières (Paris, Bibliothèque
Nationale de France, lat. 13260 et lat. 14284), il déclare: ‘A quoi bon multiplier
les exemples? Il est clair que de pareilles figures n’ont aucun rapport avec les
Heures de la Vierge ou avec les Psaumes de la Pénitence qu’elles illustrent’. Il
croit finalement pouvoir expliquer l’humour de ces oeuvres par la naïveté: ‘La
foi profonde donna à ces temps la gaieté, la sérénité de l’enfance’. Aussi, les
drôleries sont-elles chastes: ‘Au badinage des artistes il ne se mêla jamais ni
indécence, ni ironie f. . .] Ce n’est qu’au XVe siècle qu’apparaît dans l’art ce
réalisme un peu bas qui, à l’occasion, ne recule pas devant l’obscénité’.
Mâle semble vouloir clore le débat plutôt que le lancer. Mais, quel que soit
son bien-fondé, l’idée que les drôleries n’ont pas de sens sera de plus en plus
contredite au XXe siècle. Quant à l’affirmation qu’elles ne contiennent ni
indécence, ni ironie, si elle est vraie de nombreux manuscrits parisiens, elle
étonne à propos du livre d’heures lat. 14284, originaire de Thérouanne, dont
l’auteur vient de louer la beauté. Il est en effet difficile de ne pas voir d’ironie
dans la représentation de singes en clercs ou en évêques, dans une école de
singes ou devant un chevalier effrayé par un escargot ou un lapin. On peut
également se demander s’il est parfaitement décent de représenter un clerc se
faisant surprendre en train de caresser une femme ou encore un évêque bénissant
1Th. Wright, Histoire de la caricature et du grotesque dans la littérature et dans l'art

(Paris: Gamier frères, 1867) [trad, de: A History o f Caricature and Grotesque in
Literature and Art (Londres: Virtue Bros., 1865)]. Jules Fleury-Husson Champfleury,
Histoire de la caricature au moyen âge (Paris: Dentu, [1870]). L. Maeterlinck, Le genre
satirique dans la peinture flamande (Bruxelles: Van Oest. 1907).
2Emile Mâle, L'art religieux du Xllf siècle en France (Paris: Armand Colin. 1987; 1‘rc
éd. 1898), pp. 121-28.
Les marges ¿1drôleries 279
un moine, un visage féminin sur le postérieur.34
E tat d e la question
Des recherches plus poussées sur les drôleries se sont beaucoup fait attendre. Ni
Dada, ni le surréalisme ne semblent avoir connu l’existence du genre, alors que
l’oeuvre de Bosch recommençait à passionner. C’est dans une thèse allemande,
celle de Marie Theres Bergenthal, que le genre est mis pour la première fois en
relation avec la thématique littéraire du monde à l’envers et avec les adynata, les
figures rhétoriques qui le décrivent.'1 En soi, le parallèle n’explique pas grand-
chose, mais il montre que l’humour des enluminures marginales n’est pas un
phénomène isolé. Surtout, il pose le problème du rapport entre les drôleries et
des textes—fables, fatrasies, proverbes ou fabliaux—bien différents des prières
qu’elles accompagnent le plus souvent.
Le lent démarrage de la recherche est d’abord dû à la difficulté d’accès du
matériau. Encore aujourd’hui, rares sont les manuscrits publiés et on ne connaît
souvent le décor marginal d’un manuscrit que par la reproduction d’un ou deux
folios dans un catalogue. Aussi les travaux réalisés par Jurgis Baltrusaitis dans
les années cinquante constituent-ils un exploit. Après avoir montré dans Le
Moyen Age fantastique ce que les drôleries devaient à la glyptique et aux
monnaies romaines, il est parvenu, dans Réveils et prodiges, à brosser l’histoire
du genre, en montrant sa constitution à partir d’un répertoire déjà existant dans
l’art roman et persistant dans la périphérie—mais aussi au coeur—du monde
gothique, puis en suivant son développement et ses mutations jusqu’à la
Renaissance.5 A ce jour, non seulement cette synthèse n’a pas été dépassée, mais
personne n’a essayé de lui en substituer une autre. Pourtant, les travaux sur
l’enluminure gothique se sont multipliés, la connaissance des ateliers, de leur
localisation et de leur chronologie s’est beaucoup précisée. Par la force des
choses, le travail de Baltrusaitis a vieilli.
Pas plus que Mâle, Baltrusaitis n’était enclin à chercher du symbolisme dans
les drôleries, ni même à s’interroger longuement sur leur possible signification.
C’est dans cette direction que se lança quelques années plus tard Lilian Randall,
aux Etats-Unis où l’oeuvre de Panofsky avait mis ces études à l’ordre du jour et
■’ Paris, BnF, ms. lat. 14284, fol. 1v (lapin et chevalier) 8 (singe en clerc), 23 (évêque
bénissant), 31 (clerc et femme), 53v (singe en évêque), 69v (école des singes).
4 Marie Theres Bergenthal, Elemente der Drôlerie und ihre Beziehungen zur Literatur
(Berlin: Hohmann. 1934).
5 Jurgis Baltrusaitis, Le Moyen Âge fantastique. Antiquités et exotismes dans ¡'art
gothique (Paris: Flammarion, 1955; 2e éd. 1981), pp. 9-71. Idem, Réveils et prodiges. Le
gothiquefantastique (Paris: Armand Colin, 1960). en particulier pp. 70-76, 140-52, 196—
234.
280 Jüan W irth
où Horst W. Janson venait de publier un admirable ouvrage sur l’iconographie

du singe, faisant abondamment usage des marges enluminées.6 Dès 1957, elle
cherchait dans les exempla, les histoires amusantes dont les prédicateurs
médiévaux truffaient leurs sermons, une source possible des drôleries.7
Contrairement aux textes littéraires relatifs au monde à l’envers, les exempla
donnent un signification morale aux motifs comiques. Ceux que cite Randall,
comme l’histoire de Renart faisant le mort pour capturer les oiseaux, trouvent
leurs équivalents visuels dans les marges. Le prédicateur peut utiliser l’histoire
en faisant de Renart une figure du diable et lui donner ainsi un sens. Il reste à
savoir si ce sens, qui n’est certainement pas inhérent à l’épisode, doit être reporté
sur sa représentation dans les marges. L’auteur semble le croire.
Par ailleurs, Lilian Randall a compris la nécessité de rendre les marges plus
accessibles à la recherche. En 1966 paraît Images in the Margins o f Gothic
Manuscripts, où l’auteur indexe par ordre alphabétique des sujets les décors
marginaux de 238 manuscrits et publie 762 illustrations.^ L’ouvrage donne enfin
une image globale du genre et une bonne idée de sa richesse: il constitue une
mine de renseignements et une base irremplaçable pour les recherches
ultérieures. L’identification des sujets contient certes des erreurs et des
imprécisions. A titre d’exemple, l’auteur ne fait pas la différence entre les clercs
séculiers et les moines et qualifie curieusement d’abbés et d’abbesses les
chanoines et les chanoinesses reconnaissables au port de l’aumusse. En dehors
des cas où le terrier est représenté, les lapins sont qualifiés de lièvres. Mais,
comme il faudrait une connaissance universelle du Moyen Age et de la nature
pour éviter toutes les fautes de ce genre, on ne peut que saluer l’érudition et la
perspicacité qui dominent l’entreprise. Dans l’ensemble, les manuscrits sont
identifiés aussi correctement que possible en fonction des connaissances alors
disponibles. L’introduction ne fait pas l’histoire du genre, mais en fixe
solidement les caractères généraux.
Sans manquer à la reconnaissance que suscite un tel travail, il faut en marquer
les limites. Tout d’abord, les vérifications montrent que la collecte des
informations a été conduite avec un soin inégal d’un manuscrit à un autre,
souvent sur l’original, parfois sur de mauvais microfilms, voire sur des
reproductions partielles du manuscrit. Bon nombre d’indications de folios sont
fausses. De nombreuses confusions semblent s'être glissées entre la collecte des
informations et l’impression de l’index. Il faut compter avec un taux d’erreurs ou
6 H. W. Janson, Apes and Ape Lore in the Middle Ages and the Renaissance (Londres:
Warburg Institute, 1952).
L. M. C. Randall. ‘Exempla as a Source of Gothic Marginal Illustration’, Art Bulletin,
39(1957), 97-107.
s L. M. C. Randall, Images in the Margins of Gothic Manuscripts (Berkeley: Univ
ersity of California Press, 1966).
Les marges ci drôleries 281
d’omissions d’au moins 10%, de 50% environ pour certains manuscrits.9 Surtout,
la forme même de cet énorme index est déconcertante. En choisissant l’ordre
alphabétique des sujets, l’auteur avait certainement en vue l’utilisation de la
documentation par des chercheurs travaillant sur un thème iconographique, par
exemple sur l’iconographie d'un animal particulier. Pour favoriser des travaux
sur le décor du manuscrit, il aurait sans doute mieux valu faire le catalogue des
sujets manuscrit par manuscrit et folio par folio, au lieu d’extraire les motifs de
leur contexte. Pour tirer de l’index ce qui concerne un manuscrit particulier, il
faut un travail considérable. L’ouvrage ne facilite pas l’étude de l’évolution du
genre et de ses variations d’une région à l’autre. Il a peut-être même contribué à
décourager ce genre de recherches.
Depuis la grande publication de Randall, Michael Camille est le seul auteur à
avoir tenté une approche globale de la marge enluminée, dans un essai sur les
marges du Moyen Âge au sens le plus large possible.101La marge de la page est
mise en rapport, d’une part avec les emplacements comparables du décor
architectural roman puis gothique, d’autre part avec la marginalité des
personnages et des comportements représentés ou avec celle des langues
vulgaires par rapport au latin. L’intuition fondamentale de l’ouvrage est
intéressante. L'ordre hiérarchique de la page enluminée, la prédominance de son
centre, réservé au texte et aux images qui l’illustrent, sur sa périphérie qui
accueille toutes les fantaisies, justifie largement ce point de vue. Par ailleurs,
l’absence d’inhibition de Camille et la vigueur de son vocabulaire lui permettent
d'appeler un chat un chat et d’aborder franchement des sujets scabreux que ses
prédécesseurs avaient ignorés par naïveté ou par pudibonderie. La gêne qu’on
éprouve devant l’ouvrage (comme en témoigne le compte rendu assez dur de M.
Hamburger)" vient surtout d’une démarche iconographique incertaine, parce que
fondée sur une sociologie hâtive et sur une connaissance parfois approximative
du matériau. Contentons-nous de deux exemples. En soulevant M’importante
question de la différence des sexes’ ou plutôt, selon le texte original, *the
important issue o f gender' , l’auteur affirme que ‘Les images que [les marges]
contiennent sont essentiellement faites par des hommes et pour des hommes’.12 Il
serait plus opportun de noter, d’une part que nous ne connaissons absolument pas
9 Parmi les manuscrits les plus mal étudiés se trouve le psautier-heures de Jean de
Neuville (New York, Pierpont Morgan Library, MS. M. 730) dont plus de 90% des
drôleries sont passées sous silence. Or ce manuscrit, datable entre 1243 et 1246, constitue
l’un des moments essentiels dans le développement du genre.
10 Michael Camille, Images dans les marges. Aux limites de fart médiéval (Paris:
Gallimard, 1997); trad, de: Image on the Edge. The Margins of Medieva! Art (Londres:
Reaktion, 1992).
11 Cf. les observations de J. F. Hamburger, Art Bulletin. 75 (1993). 319-27.
12Camille (cf. note 10), pp. 170 et ss.
282 Jean W irtii
le pourcentage des enlumineurs de sexe féminin, d’autre part que les femmes
prédominent largement parmi les destinataires des livres de prières à drôleries,
en particulier de ceux qu’étudie Camille. Il n’est pas plus heureux d’affirmer à
propos de la scatologie que Íes marges des manuscrits en sont littéralement
pleines’.13 Camille n’a sans doute pas regardé les mêmes manuscrits que Mâle
(‘Au badinage des artistes il ne se mêla jamais ni indécence, ni ironie’). En fait,
les motifs scatologiques forment entre 1 et 2% des drôleries et sont inégalement
répandus selon les manuscrits, beaucoup d’entre eux n’en comprenant aucun. Le
jugement de Mâle et celui de Camille sont donc également erronés. L’ouvrage
plus récent consacré au psautier Luttrell est sensiblement plus épais, mais il
souffre des mêmes défauts et ne remplace pas l’excellente petite monographie de
Janet Backhouse.14
Au risque de sembler injuste, il faut admettre que peu d’études récentes ont
vraiment fait avancer la recherche sur les manuscrits à drôleries. Du point de vue
de l’histoire du genre, le compte-rendu du livre de Randall par Nordenfalk est
certainement la contribution la plus importante malgré sa brièveté.15 Il a le mérite
de poser sérieusement la question de l’origine des drôleries, en apportant du
matériau neuf. Les plus grands services sont rendus par des études dont les
drôleries ne sont pas le sujet, comme l’ouvrage de Branner sur l’enluminure
parisienne ou, plus récemment, celui de Judith Oliver sur les psautiers liégeois.16
Du point de vue iconographique, rares sont les travaux de bon sens. On marquera
d’une croix blanche l’étude du Roman de la Rose ff. 25526 de la Bibliothèque
Nationale de France par Sylvia Huot, un cas exceptionnel de rapport texte /
image analysé avec compétence et ingéniosité, même si elle n’échappe pas
toujours à la surinterprétation.17 La plupart de ces recherches sont viciées par des
a priori malheureux sur ce rapport et rappellent de près ou de loin les égarements
déjà dénoncés par Emile Mâle. Plutôt que de les écarter par une autre pétition de
principe, nous allons essayer de critiquer méthodiquement les principaux
malentendus sur la nature des drôleries et sur leur rapport aux textes.
13 Ibid. p. 152.
14 Michael Camille, Mirror in Parchment. The Luttrell Psalter and the Making o f
Medieva! England (Londres: Reaktion, 1998). J. Backhouse, The Luttrell Psalter
(Londres: British Library. 1989).
15 C. Nordenfalk, compte rendu dc Randall, Images in the Margins o f Gothic
Manuscripts, dans Burlington Magazine, 109 (1967), 418-21.
16 R. Branner. Manuscript Painting in Paris during the Reign o f Saint Louis (Berkeley:
University of California Press, 1977). J. Oliver, Gothic Manuscript Illumination in
the Diocese o f Liège (c. 1250-c. 1350) (Leuven: Peeters, 1988).
17 S. Huot, ‘Vignettes marginales comme glose marginale dans un manuscrit du
‘Roman de la rose’ au quatorzième siècle (B. N. fr. 25526)', Littérales. 2 (1987). 173-86;
The "Romance o f the Rose” and its MedievaI Readers: Interpretation. Reception.
Manuscript Transmission (Cambridge: Cambridge University Press, 1993), pp. 273-322.
Les marges à drôleries 283
L es lim ites d e l ’interprétation
La prolifération des études sur le rapport texte / image depuis une trentaine
d'années mériterait une enquête sociologique sur les stratégies universitaires. Elle
répond à l'injonction de pratiquer l'interdisciplinarité, c’est-à-dire de juxtaposer les
approches de différents spécialistes sur des sujets qui exigeraient plutôt la
pluridisciplinarité, la maîtrise par chacun de l’ensemble des méthodes requises
pour appréhender l’objet. Il est à craindre que l’interdisciplinarité soit à la
spécialisation aveugle ce qu’un emplâtre est à une jambe de bois, tout en
permettant à des historiens et à des critiques littéraires de pontifier sur les images
en ignorant tout de l'histoire de l’art et de ses méthodes.18 Elle a cependant des
origines plus lointaines dans I’iconologie panofskienne, parfois adaptée à des
genres artistiques, comme les drôleries, auxquels elle convient mal. Même pour
des gens qui savent regarder les images, faire de l’image l’illustration d’un texte
rassure sur la pertinence de l’interprétation. Dans le cas des drôleries, ou bien on
nie à la fois qu’elles aient un rapport au texte et un sens, comme l’avait fait Mâle,
ou bien on voudrait leur trouver un sens dans un hypothétique rapport au texte.
L’article de Lilian Randall, ‘Games and the Passion in Pucelle’s Hours of
Jeanne d'Evreux’, est un bon exemple d’une interprétation de motifs apparemment
profanes et anodins visant à leur donner un sens par rapport au contenu du livre
d’heures.19 Il a dans l’ensemble été accepté par la critique et ne présente pas de
grosses invraisemblances. Aux folios 15V—16 des Heures de Jeanne d’Evreux (New
York, Cloisters, 54.1.2), Jean Pucelle a placé, respectivement sous le Baiser de
Judas et l’Annonciation, deux cavaliers grotesques jouant à la quintaine avec un
tonneau et des jeunes gens jouant à la grenouille, jeu qui consiste à frapper pour
rire un personnage assis qui doit deviner l’identité de ses agresseurs (fig. 1).
Constatant que les représentations du Christ aux outrages montrent parfois une
parenté formelle avec celles du jeu de la grenouille et qu’il est fait explicitement
allusion à ce jeu dans un Livre de ta Passion à propos de la dérision du Christ,
l'auteur voit dans la scène marginale une allusion à la Passion qui compléterait la
trahison de Judas de la page précédente. En même temps, ce jeu printanier
évoquerait l’époque de l’Annonciation, fêtée le 25 mars. Quant aux cavaliers et au
18 L'interdisciplinarité étant l'alibi de la spécialisation aveugle, mettre en cause le

rapport texte/image, c’est attaquer la routine universitaire dans ses fondements. Nous en
avons fait l'expérience en soumettant au Fonds National Suisse de la Recherche le projet
en cours. L’expert, une historienne de l’art américaine à laquelle le féminisme tient lieu
de compétence, l'avait condamné parce qu’il affirmait que les marges à drôleries sont
normalement indépendantes du texte. Il est à l’honneur de l’institution académique
d’avoir accepté de financer le projet l’année suivante, après que je lui eus soumis une
réfutation en règle de l’expertise au lieu de modifier le rapport.
19 L. M. C. Randall, ‘Games and the Passion in Pucelle’s Hours o f Jeanne d ’Evreux’,
Speculum, 47 (1972), 246-57.
284 Jean W irtii
Fig. 1. Heures de Jeanne d'Evreux, New York, Cloisters, MS. 54.1.2, fols 15'-16.
‘Arrestation du Christ et Annonciation ’. © The Cloisters, New York
Fig. 2. Bréviaire de Renaud de Bar, Verdun, Bibliothèque municipale, ms. 107,

fol. 138r, 'Siège d ’un château’. © Bibliothèque municipale, Verdun
tonneau sur la quintaine. ils feraient allusion au Juif errant qui avait frappé le
Christ lors du portement de croix et au Christ lui-même, le vin du tonneau
rappelant l’eucharistie.
On ne voit pas pourquoi chercher une allusion au Juif errant, mais il n’est pas
impossible que la quintaine sous le Baiser de Judas soit un commentaire parodique
de cette image, encore que la validité de l’interprétation serait difficile à établir.
L'allusion à la Passion est en revanche très improbable dans le cas du jeu de la
grenouille. Pour que le parallèle fonctionne, il faudrait que les yeux du jeune
homme faisant office de grenouille soient voilés, comme cela est fréquemment le
cas dans les représentations de ce jeu et dans celles de la dérision du Christ. De
surcroît, on comprendrait mal pourquoi Pucelle a placé sa ‘grenouille’ sur un
coussin et lui a donné comme seuls bourreaux des jolies filles qui semblent le
titiller. S’il s’agissait vraiment d’une allusion à la Passion, le peintre aurait tout fait
pour la rendre incompréhensible. En revanche, comme d’autres bas-de-page
complètent incontestablement l’image principale dans ce petit livre, ainsi le
Massacre des Innocents au bas de l’Adoration des Mages du fol. 69, l’allusion au
printemps est très possible. De cet exemple, nous pouvons tirer une première règle
d'interprétation trop rarement appliquée: NE fa ir e L’HYPOTHESE D’UNE ALLUSION
QUE SI CELLE-CI A QUELQUE CHANCE D’ETRE PERCEPTIBLE.
La thèse de Sharon Davenport sur les manuscrits enluminés pour Renaud de
Bar contient quatre cents pages consacrées à l'analyse des drôleries et représente
la tentative la plus systématique que nous connaissions de leur donner un sens.20
On y trouve bon nombre de renseignements précieux, parce que l’auteur montre
une rare connaissance des realia, qu’il s’agisse du costume des jongleurs ou des
religieux, de l’anatomie des animaux, de leurs moeurs réelles ou légendaires et
des symbolismes sexuels qui peuvent leur être attachés. Mais elle pose le
principe d’un rapport avec le texte liturgique et interprète pratiquement chaque
drôlerie en ce sens, tout en multipliant les adverbes exprimant l’incertitude.
Voyons quelques exemples. Dans le bréviaire de Renaud de Bar (Londres,
British Library, Yates Thompson 8, fol. 92), l’auteur renchérit sur Randall en
interprétant le jeu de la grenouille et une lutte à califourchon comme des
allusions à la Passion. Mais on peut se demander si la liturgie du quatrième
dimanche de l’Avent que ces images accompagnent, toute centrée sur
l’imminence des fêtes de Noël, est vraiment celle qui appellerait le mieux ces
allusions. Au fol. 192, l’auteur a su identifier le thème du lapin, ennemi du
tailleur qui s’intéresse à sa fourrure (plutôt qu’à sa viande), mais elle croit devoir
le relier à l’abstinence durant le carême. Il arrive, ainsi au fol. 53, qu'un psaume
faisant allusion à la musique soit décoré de musiciens dans les marges. Mais, si
l’on prend en compte le fait que tous les psaumes sont des prières chantées, que
20 S. K.. Davenport, Manuscripts illustratedfo r Renaud o f Bar. bishop o f Metz (1303-

1316) (thèse de PhD inédite, Courtauld Institute of Art. University of London, 1984),
pp. 564 et ss.
2S6 Jean Wirth
beaucoup d'entre eux le rappellent explicitement et que les musiciens sont aussi
innombrables dans les marges qu’elles entourent ou non des psaumes, il est plus
raisonnable d’y voir une coïncidence. On fixera donc une seconde règle d’inter
prétation: LA POSSIBILITE D’UNE ALLUSION EST INVERSEMENT PROPORTIONNELLE
A LA FREQUENCE DU MOTIF ICONOGRAPHIQUE ET A CELLE DU MOTIF TEXTUEL
AUQUEL IL EST SUPPOSE SE RAPPORTER.
Toutes les interprétations de Davenport ne sont pas aussi incertaines. Comme
elle le remarque, il est clair que l’un des peintres, celui qu’elle identifie comme
la seconde main, a accès au texte latin, ce qui ne vaut sans doute pas pour tous
les enlumineurs. 11 remplace fréquemment les drôleries par des scènes bibliques
en rapport avec le texte et quelques-unes de ses drôleries peuvent être
légitimement soupçonnées d’y faire allusion. Dans la seconde partie du bréviaire
de Renaud (Verdun, Bibliothèque municipale, ms. 107, fol. 137v), des lapins
défendent un château attaqué par des chiens et on lit sur la même page
l’antiphone: ‘Expugna impugnantes me' (fig. 2). Une coïncidence est possible,
car le combat est un thème récurrent dans les psaumes et les prières qui en
dérivent comme dans les drôleries. Mais, bien qu’il s’agisse de la troisième main,
moins attentive au texte, l’intention est cette fois plus probable, car les marges et
l’illustration du psaume 34, dont provient l’antiphone, montrent plusieurs fois
une scène de combat.21 Dans la première partie (Yates Thompson 8), deux
allusions débusquées par l’auteur chez le second artiste sont très vraisemblables.
Aux fol. 295v-296, un singe couronné est porté par deux lapins vers une tente où
siège une guenon à son tour couronnée. Il s’agit de l’office de sainte Agathe et
on y lit le répons: 'Ipse me coronavit qui per apostolum Petrum in custodia me
confortavit'. Le motif de la guenon couronnée est suffisamment exceptionnel
pour justifier l’interprétation parodique du thème. L’intention est plus difficile à
percevoir au fol. 308v, lorsqu’un homme nu vu de dos tient un tissu rouge muni
d’un galon, à proximité de l’antiphone: ‘Qui solebat in sericis procedere
vestimentis, post in abjectis vestibus servit, pauper pauperibus', mais l’allusion
est difficilement contestable.
Il n’en reste pas moins qu’à vouloir systématiquement réduire les drôleries à
des illustrations du texte, Davenport laisse échapper leur incongruité et leur
impertinence. Les superbes marges du troisième peintre mettent l’accent sur les
jeux guerriers, la fauconnerie et les jolies femmes, ce qui ne va pas complète
ment de soi dans le bréviaire d’un évêque et devrait poser quelques problèmes
sur sa personnalité. Persuadée du sérieux des drôleries, Davenport en vient
logiquement à défendre son évêque contre la réputation d’ecclésiastique indigne
que les historiens lui avaient faite, en passant sans broncher sur des faits qui font
réfléchir. La carrière de Renaud de Bar n’est portant pas très édifiante. Vers
1300, il est entre autres archidiacre à Bruxelles et à Besançon, chanoine à
21 En particulier: Londres. BL, MS. Add. 62925 (Psautier Rutland), fol. 37: Paris, BnF,
ms. lat. 1076, fol. 42; Cambrai, Bibliothèque municipale, ms. 87, fol. 132v.
Verdun et à Metz. Evêque de Metz élu en septembre 1302, il est consacré

l’année suivante et ordonné prêtre à cette occasion. Il s’agit donc de la carrière
typique d'un prébcndier absentéiste qui ne s’est peut-être jamais assis dans les
stalles d’un choeur. A partir de 1306, il passe son temps à guerroyer contre la
ville, le chapitre et le duc de Lorraine. Rien n’indique le moindre intérêt pour les
tâches spirituelles et, lorsque les enlumineurs plaçaient dans les beaux livres
qu’il se commandait des images relatives à ces tâches, c’était probablement pour
respecter la dignité de la fonction plutôt que celle de la personne.
La volonté de mettre à tout prix les drôleries en rapport avec le texte qu’elles
décorent n’est pas le seul abus destiné à leur donner un sens. N’importe quelle
histoire médiévale se présente, comme les fables, ou peut se présenter, comme
les exempia, avec une moralisation. Chaque objet du monde extérieur, chaque
événement de la vie peut devenir pour un esprit pieux l’occasion d’un petit
sermon. Mais il va de soi que la moralisation n’est pas inhérente à l’objet ou à
l’anecdote et que le même prétexte peut servir à des moralisations différentes.
Comme on l’a déjà remarqué à propos de l’article de Randall sur les exempla, il
est douteux que le bon tour de Renart faisant le mort pour attraper les oiseaux
soit systématiquement destiné à évoquer les ruses du diable. Il est plus probable
qu’une anecdote ou qu’une image quelconque ait été vue comme ouverte à
l’interprétation, que ce même Renart ait pu évoquer à un dominicain le diable
tentant les pécheurs et à un franciscain un dominicain plumant les dévotes. Un
petit ouvrage comique comme Salomon et Marcolfe, où le roi et son fou
échangent des maximes respectivement nobles et grossières, repose tout entier
sur la possibilité d’interpréter les mêmes faits avec le sérieux du moraliste et la
dérision du bouffon.22
Deux autres articles de Randall, malgré l’excellence de leur érudition,
suscitent pour cela des réserves. L’un porte sur le guerrier terrorisé par un
escargot (fïg. 3).23 L’auteur relève l’existence d’oeuvres littéraires où les
Lombards sont supposés assez lâches pour fuir devant les escargots et interprète
donc le thème comme une satire dirigée contre eux. Elle relève aussi que les
Lombards n’avaient pas le droit de porter les armes dans le royaume de France,
alors que les marges nous présentent des guerriers, souvent des chevaliers, armés
jusqu’aux dents. Par ailleurs, le chevalier est fréquemment représenté, dans les
cycles des vertus et des vices où il illustre la lâcheté, mais aussi dans le décor
marginal, en train de laisser tomber son épée face à un lapin menaçant. On voit
mal ce qui empêcherait quelqu’un qui en veut aux usuriers lombards d’identifier
comme Lombard le chevalier au lapin aussi bien que celui à l’escargot, ou*2
~ Salomon et Marcolfus, éd. W. Benary (Heidelberg: Winter, 1914). Cf. P. Lehmann,

Die Parodie im Mittelalter (Stuttgart: Hiersemann. 1922; 2e éd. 1963), pp. 172 et ss.
2' L. M. C. Randall, ‘The Snail in Gothic Marginal Warfare", Speculum, 37 (1962),
358-67.
288 J ean W irth
Fig. 3. Psautier de Gorleston, Londres, British Library, MS. Add. 49622, fol.
162', ‘Chevalier effrayé par un escargot ’. ©British Library’, London
Fig. 4. Psautier gantois, Oxford, Bodleian Libraiy, MS. Douce 6, fol. 92'',
‘Couvade ’. © Bodleian Library. Oxford
encore ce qui empêcherait quelqu'un qui en veut, par exemple, aux barbiers
d'identifier comme barbier le chevalier à l’escargot. Traitant dans un autre article
de l’homme qui couve des oeufs (fig. 4), le même auteur présente plusieurs
possibilités d’interprétation avant de conclure à une plaisanterie anti-anglaise.24
Les Anglais sont en effet souvent supposés munis d’une queue et qualifiés de
‘coués’ depuis le Xlle siècle. Or, ‘coué’ peut faire jeu de mot avec ‘couard’,
mais aussi avec ‘couvé’. L’homme qui couve serait donc un Anglais efféminé, ce
que confirmerait l’absence du thème dans les manuscrits d’outre-Manche.
L’interprétation laisse à nouveau sceptique. Les manuscrits anglais sont aussi
remplis de personnages coués que ceux du continent, peut-être plus, alors que le
jeu de mot sur ‘couvé’, effectivement présent dans un texte anti-anglais, passe
par la désignation des Anglais comme ‘coués’, tandis que la couvaison mascu
line ou couvade apparaît dans des contextes littéraires où il n’est pas question
d’Anglais.
En cherchant des interprétations politiques, Randall tend à oublier un autre
niveau symbolique plus général et dont la mise en oeuvre est plus probable. Le
lapin qui fait fuir le chevalier est, en français médiéval un ‘conin’, ce qui fournit
une mine inépuisable de jeux de mots obscènes. Qu’on se contente d’évoquer ici
les aventures de Renart à la cour du roi Conin.25 Or l’escargot, mou et visqueux
dans sa coquille, suggère également le sexe féminin, comme l’a bien vu
Camille.26 Le tournoi et le duel sont des métaphores fréquentes du coït, dans la
littérature comme dans les marges. Dans un fabliau comme Berengier au long
cul, un chevalier fanfaron est mis au défi par sa femme, méconnaissable dans son
armure, de jouter avec elle ou de lui baiser le postérieur.27 Face au ‘long cul’ de
son redoutable adversaire, le chevalier effrayé laisse tomber son épée, exacte
ment comme s’il avait affaire à un conin ou à un limaçon. Pour ce qui est de la
couvade, elle renvoie de manière non moins évidente à l’absence de virilité et on
pense immédiatement au roi de Turelure en mal d’enfant qui scandalise le héros
dans Aucassin et Nicolette. Il faut aussi remarquer, comme le fait Randall, que
cette posture s’accompagne généralement dans les marges d’un geste caractéris
tique: le couveur lève un oeuf vers le soleil pour le mirer et vérifier sa fécondité.
Le même geste apparaît non moins fréquemment dans ce répertoire pour le
médecin qui mire les urines, une pratique qui servait de test de grossesse et qui
est appliquée à des hommes supposés enceints dans le Jeu de la Feuillée d’Adam
de la Halle.
24 L. M. C. Randall, ‘A Mediaeval Slander’, Art Bulletin, 42 (1960), 25-38.

25Le Roman de Renart, éd. A. Strubel (Paris: Gallimard, 1998), branche 23, pp.751 et
ss. (branche 22 de l’édition Martin).
26Camille (cf. note 10), pp. 46 et ss.
27 Nouveau recueil complet des fabliaux, éd. W. Noomen et N. Van den Boogaard
(Assen: Van Gorcum, 1983-98), iv, n° 34, pp. 247 et ss.
290 Jean W irtii
Pourquoi préférons-nous voir dans le chevalier qui laisse tomber son épée
devant un escargot ou dans la couvade une allusion sexuelle plutôt que politique?
Outre que, comme nous l'avons vu, l’interprétation politique est difficilement
généralisable, remarquons d’abord que les deux interprétations ne sont pas
exclusives, car elles ne se situent pas au même niveau. Pour la couvade, Randall
comprend très bien que la virilité du couveur est en cause. En fait, l’interpré
tation sexuelle de ces thèmes est la condition de possibilité d’une éventuelle
interprétation politique. Dans le cas de la couvade, elle est immédiate, puisque
l’homme adopte un comportement des femelles destiné à la reproduction de
l’espèce. Dans le cas du chevalier, elle semble bien se déduire du choix des
animaux qui le terrorisent. La souris, la grue ou l’écureuil par exemple, qui
métaphorisent le pénis dans les fabliaux, ne jouent jamais ce rôle, malgré le soin
que prennent les enlumineurs à innover en variant les acteurs d’une histoire
conventionnelle. L’interprétation sexuelle des deux thèmes est donc pertinente,
tandis que leur moralisation politique ressemble fort à la projection arbitraire
d’une moralisation. Si le peintre voulait proposer un sens moral, il aurait
certainement trouvé les moyens de le suggérer. A titre d’exemple, pour identifier
Renart prêchant aux poules à l’évêque, il lui donne une mitre. Il serait excessif
d’en déduire que les représentations d’un renard non mitré volant une poule
soient des allusions anticléricales, mais il est significatif qu’on ne donne guère
d’autres attributs à cet animal que ceux du clergé, qu’il ne revêt jamais, à notre
connaissance, la couronne royale ou le bonnet à grelots du jongleur. 11 serait
arbitraire de faire entrer tous les renards dans le clergé, mais il est raisonnable
d’affirmer que cet animal rusé évoque en premier lieu le clergé. On peut ainsi
poser une troisième règle d'interprétation: LA SIGNIFICATION D’UN MOTIF NE
PEUT ÊTRE ÉTABLIE QUE SUR L’EXAMEN DE SES DIFFÉRENTES OCCURRENCES.
L’état de la recherche nous oblige à formuler cette règle qui, comme les
précédentes, devrait aller de soi.
Méthodologiquement proche du travail de Randall, l'article consacré par
Howard Helsinger à l’illustration du psaume Beatus vir qui sert de frontispice au
psautier est d’une rigueur inhabituelle.28 Grâce à l’index de cette chercheuse, il
peut faire intervenir des comptages et prendre en compte la fréquence des
thèmes. L’analyse la plus longue et la plus minutieuse concerne la chasse au cerf.
Il repère le thème dans huit frontispices du psautier et se demande si sa
fréquence à cet emplacement est significative. Le cerf apparaît dans un tiers des
manuscrits décrits par Randall, la chasse au cerf dans trois quarts de ces cas. 11
considère néanmoins que la présence du thème au frontispice est significative,
car il n’accompagne avec la même fréquence aucun autre psaume enluminé. Le
psaume 41 (‘De même que le cerf désire la fontaine des eaux, de même mon âme
te désire mon Dieu...’) et le Bestiaire qui signale l’hostilité du cerf envers le
28 H. Helsinger, ‘Images on the Beatus Page of Medieval Psalters’, Art Bulletin, 53

(1971), 161-76^
serpent, inspirent la moralisation du cerf dans le commentaire biblique: il repré

sente Yhomo viator, le chrétien pérégrinant exilé dans ce monde. Les chiens de
chasse qui poursuivent le cerf sont interprétés par les prédicateurs comme des
figures du diable. L’iconographie reprend ces moralisations, déjà dans le psautier
d’Utrecht (v. 820) où les chiens poursuivant le cerf illustrent le psaume 41. Si le
thème décore le Beatus vir des psautiers gothiques, c’est parce que cette page
sert de frontispice au recueil tout entier et que sa thématique n’est donc pas
étroitement liée au contenu du premier psaume.
La démonstration semble à première vue indiscutable, car l’objection qu’on
aurait tirée de la banalité de la chasse au cerf est réfutée par le repérage d’une
fréquence anormale au frontispice des psaumes. Vérifions tout de même les
chiffres: nous avons enregistré 356 cerfs dont 125 dans les psautiers et 60 dans
les livres non liturgiques. Cela fait 2,9 % des fiches dans les psautiers comme
dans les livres non liturgiques. 11 est peu probable que l’égale fréquence du cerf
dans les manuscrits non liturgiques et en particulier dans les manuscrits profanes
soit due à une pieuse moralisation, mais cela ne contredit pas formellement la
thèse d’Helsinger. 11 est plus ennuyeux de constater que les psautiers ne se
distinguent pas par l’abondance de ces animaux. Les singes, par exemple, font
13,6 % des fiches dans les psautiers et 12,6 % dans les livres non liturgiques. Les
singes ont donc un rapport plus étroit avec le psautier que les cerfs. Surtout,
l’auteur a oublié un paramètre essentiel: le Beatus vir, frontispice du psautier, est
beaucoup plus souvent et plus abondamment décoré qu’aucun autre psaume. A
l’exception du frontispice de la Bible, il s’agit de la page manuscrite la plus
susceptible de recevoir un thème quelconque. Même si nous ne pouvons traduire
ce dernier point en donnée chiffrée, il suffit pour considérer la chasse au cerf,
décorant 8 frontispices sur les 102 psautiers et bréviaires pris en compte par
Randall, comme assez peu fréquente à cet emplacement. Sa répartition dans les
différents types de manuscrits laisse au contraire prévoir qu’elle est motivée par
les goûts cynégétiques de la clientèle.
Helsinger repère 14 psautiers au moins où David et Goliath occupent le bas de
page du Beatus vir. Il s’agit d’un pourcentage exceptionnel, mais le thème est
motivé de manière cette fois évidente par l’attribution du psautier à David et par
le contenu du psaume. 11 note également neuf scènes de tournois sur cette page,
ce qui n’est pas trivial. Le tournoi est déjà là dans le psautier de Saint-Albans
(entre 1120 et 1130) où un commentaire, dû à l’enlumineur lui-même, en fait un
symbole du combat spirituel, exégèse que Helsinger applique aux autres
occurrences. Vérification faite, le tournoi occupe 0,36 % de nos fiches, 0,42 %
pour les psautiers et 0,67 % pour les manuscrits non liturgiques. La part des
manuscrits non liturgiques s’explique par le report en marge de scènes illustrant
le texte dans les romans chevaleresques. Il reste que le tournoi est significative
ment plus présent dans le psautier que dans la moyenne des manuscrits. Mais
faut-il nécessairement y voir une figure du combat spirituel? Dans le contexte
monastique du psautier de Saint-Albans, cette moralisation explique vraisem-
292 Jean W irth
blablement la présence du thème. Or les psautiers gothiques qui nous occupent

ont souvent une ambiance très profane. Dans un psautier pris en compte par
Helsinger (Oxford, Bodleian Library, Laud. lat. 84, fol. 19), le tournoi oppose un
chevalier à une femme (fig. 5), tandis que dans le psautier-heures de Jean de
Neuville (New York. Pierpont Morgan Library, M. 730), cette variante orne les
bouts-de-lignes de manière récurrente, en compagnie de drôleries obscènes et
scatologiques. Il s’agit à l’évidence d’une allusion aux ébats amoureux et il serait
difficile, même à une oie blanche, de lui appliquer le sens d’un combat spirituel.
A défaut de convaincre, les études de Randall et d’Helsinger ne heurtent pas le
bon sens. Il n’en va pas de même de l’article récent de Madeline Caviness sur les
heures de Jeanne d’Evreux.29 Selon toutes probabilités, le chef-d’oeuvre de Jean
Pucellc a été commandé par Charles IV pour sa jeune et royale épouse en 1324.
L’auteur part de ce fait et des présupposés ordinaires sur l’oppression des femmes
au Moyen Age. pour interpréter le décor marginal comme une série d’exemples
moraux destinés à faire de la reine une épouse fidèle et une reproductrice soumise,
une intention bien compréhensible après l’affaire de la Tour de Nesles où trois des
brus de Philippe le Bel avaient beaucoup fait parler d’elles. Analysant la page
consacrée à l’Annonciation qui avait déjà intéressé Randall (fig. 1), Caviness fait la
pétition de principe que la reine prie dans l’initiale pour avoir un héritier mâle. La
chandelle (en réalité une masse d’arme) que tient le garde qui la protège est donc
un symbole phallique. En lisant la prière ‘Domine labia mea aperies', le mot
'labia' occupant le milieu de la page, Jeanne pense aux lèvres de son vagin que son
maître et époux se chargera d’ouvrir, sans toutefois lui procurer les plaisirs de la
jeunesse auxquels renvoie le bas de page. Le chat dans le feuillage en dessous de
l’initiale (si c’est bien un chat) est le pussy de la reine; la souris sous le jeune
homme du bas de page (en réalité un lapin) représente à nouveau le pénis, le chien
la fidélité conjugale et le reste à l’avenant. La page se partagerait entre les
symboles des vertus demandées à la reine et ceux des vices dont elle doit se
détourner. La même clé d’interprétation est imperturbablement appliquée au reste
du manuscrit et l’auteur n’a pas de mal à conclure sur le caractère révolutionnaire
de ses résultats. Considérant que ses conceptions de la recherche doivent être
adoptées par l’ensemble du monde académique, elle s’insurge dans l’une des
dernières notes de son texte contre un collègue qui avait osé proposer un cours sur
l’art du XIXe siècle, fondé sur une liste d’artistes mâles de race blanche, en grande
partie français, et se félicite que la menace d’un procès ait empêché cette infamie.
L’étude du statut social des deux sexes à travers l’histoire est un sujet qui
mérite davantage l’intérêt qu’on lui porte aujourd’hui que l’indifférence dont les
historiens ont trop souvent fait preuve à son égard. Mais cela n’autorise pas à se
comporter comme si les opinions féministes garantissaient par avance la validité
de l’interprétation. La méthode utilisée par Madeline Caviness permettrait de
29 M. H. Caviness, ’Patron or Matron? A Capctian Bride and a Vade Mecum for Her
Marriage Bed’, Speculum, 68 (1993), 333-62.
Les marges ¿¡ drôleries 293
Fig. 5. Psautier, Oxford, Bodleian Library, MS. Laud. lat. 84, fol. 19, 'Initiale
Beatus © Bodleian Libraiy, Oxford
tirer les mêmes conclusions de quelque oeuvre picturale ou littéraire que ce soit.
Toute histoire peut en effet être considérée comme un exemple à suivre ou à ne
pas suivre, dans un livre d’heures comme dans un western, de sorte que les
mêmes intentions didactiques et phallocratiques peuvent être infailliblement
prêtées à n’importe laquelle. Appliquée au bréviaire de Renaud de Bar, cette
méthode permettrait de prouver sans la moindre difficulté que l’évêque était une
fille à marier. Il s’agit de ce que les anthropologues anglo-saxons appellent un
self-confirming system et qu’ils croient trop souvent caractéristique des sociétés
autres que la nôtre.
En appliquant la seconde règle d’interprétation que nous proposons, Camille
aurait pu éviter l’une des hypothèses les plus discutables de son essai sur les
marges.30 Il imagine que certaines drôleries obscènes ont pu être suggérées au
peintre par des jeux de mots sur le texte latin, comme la devinette bien connue
qui fait de *conculcavit’ le mot le plus poilu du psautier, puisqu’on peut y lire les
mots français ‘con’, ‘cul’, ‘cas’ et ‘vit’. C’est ainsi qu’un enlumineur ignorant le
latin aurait pu lire l’anglais 'legs' pour le latin 'lege' et placer dans la marge une
tête sur jambes. Pareillement, 'juvencularum', décomposé dans le latin 'juvenis'
et le français ‘cul’, aurait pu inspirer à l’artiste du psautier Rutland le jeune
homme nu à tête de Christ qui offre son derrière à l’agression armée d’un singe
chevauchant une autruche (Londres, BL, MS. Add. 62925, fols 66'—67). On ne
peut rejeter formellement la possibilité qu’un peintre se soit un jour permis de
rêvasser ainsi sur le texte du psautier et on conviendra avec Camille que le goût
du jeu de mots n’a pas beaucoup de limites au Moyen Age. Malheureusement, la
langue latine est ainsi faite que les syllabes incriminées y prolifèrent et que le
petit jeu lasserait rapidement l’obsédé sexuel le plus invétéré. Dans la seule
double page du psautier Rutland qu’on vient d’évoquer, il y a, outre le mot
'juvencularum', cinq occurrences du préfixe 'con-' et, si l’enlumineur avait été
'°Camille (cf. note 10), pp. 55 et ss., 61.

294 Jean Wirth
au courant des vilains mots aujourd’hui en usage, il aurait pu mettre à profit deux
occurrences du suffixe ‘-bit’. Autant qu’à faire, ne serait-il pas plus vraisem
blable que le bas de page ait été inspiré par le latin du psaume 67, verset 14: ‘e/
posteriora dorsi eius'? Mais, s’il avait vraiment procédé ainsi, le peintre aurait
probablement réservé son bas de page blasphématoire pour le psaume 77, verset
73: ‘Et percussit inimicos suos in posteriora; opprobrium sempiternum dedit
illis'.
L’éventualité que les drôleries reposent sur des jeux de mots ou, plus
généralement, sur des allusions textuelles, dépend de l’accès que le peintre avait
au texte, soit directement, par la connaissance de la langue écrite, soit in
directement, si quelqu’un lui apportait de l’aide. Il n’est généralement pas
possible de trancher formellement entre l’accès direct ou indirect au texte et la
prudence voudrait qu’on parlât, à la manière de Beat Brenk,31 du concepteur des
drôleries. Mais, dès qu’on s’est familiarisé avec le contenu d’un manuscrit, il
devient le plus souvent évident que l’enlumineur possédait ou non une idée du
texte. Voyons quelques exemples.
Le cas du bréviaire de Renaud de Bar a déjà été évoqué: l’un au moins des
peintres l’illustre sans difficultés lorsqu’il choisit de le faire. Plusieurs psautiers
ont été décorés dans les initiales à l’aide d’une liste qui circulait en latin et en
français et qui était donc destinée à des enlumineurs capables de lire respective
ment l’une et l’autre de ces langues.32 Dans le cas du lat. 10435 de la
Bibliothèque nationale de France, enluminé à Arras vers 1280, l’enlumineur
utilise une liste française et ne connaît pas bien la Bible. Lorsqu’il lit l’indication
‘moyses et aaron drescent l’arche’, il ne pense pas à l’arche d’alliance et fait
tenir l’arche de Noé à ces deux personnages, interprétant le français ‘drescer’
dans le sens de ‘brandir’ et non pas de ‘bâtir’, que suggère mieux le latin
'erigunt'. Les indications de la liste ont été reproduites à l’encre rouge dans la
marge, après le travail de l’enlumineur, jusqu’au folio 155v. Il s’agissait d’une
liste incomplète, à moins qu’elle ait été reprise à l’enlumineur avant la fin de son
travail, car l’iconographie des initiales change alors de caractère et devient
passe-partout: aux scènes bibliques succèdent des personnages en prière. On
serait tenté d’en conclure que le peintre n’avait pas accès au texte. Pourtant, au
psaume 127, 'Beati omnes qui timent Dominum', il place des dominicains en
prière dans l’initiale et un franciscain en conversation avec une femme dans la
marge, laissant entendre que les premiers craignent le Seigneur plus que le
second (fol. 158).
31 B. Brenk, ‘Le texte et l’image dans la vie des saints au Moyen-Âge’, in Texte et
Image. Actes du Colloque international de ChantUle (Paris: Belles Lettres, 1984), pp. 31-
39.
32 E. A. Peterson, ‘The Textual Basis for Visual Errors in French Gothic Psalter
Illustration’, in The Early Medieval Bible. Its Production. Decoration and Use, éd. R.
Gameson (Cambridge: Cambridge University Press, 1994), pp. 177-204.
Les marges ci drôleries 295
Le peintre ou son éventuel conseiller lit donc le français et assez de latin pour
identifier l’incipit des psaumes, mais il est incapable de concevoir tout seul un
programme d’exégèse par l’image. Son exemple paraît assez caractéristique de la
culture moyenne des enlumineurs, mais les écarts peuvent être considérables des
uns aux autres. Le livre d’heures lat. 13260 de la Bibliothèque nationale de
France, à l’usage de Saint-Amand, semble avoir été décoré à Cambrai dans la
dernière décennie du xme siècle.33 Le peintre a illustré les psaumes 1, 32, 45, 69,
73, 75, 101, 109 et 137, ce qui est inusuel, sauf pour les psaumes 1, 101 et 109.
Comme l’a vu François Avril, l’iconographie est celle qui prévaut en France
pour les psaumes majeurs. Or ce n’est pas le peintre, mais le scribe, qui réserve
les emplacements destinés aux grandes initiales. On avait donc certainement
prévu une subdivision originale du psautier, mais le peintre, faute de connaître
ou de pouvoir inventer des iconographies conformes à ce choix, s’est contenté de
reproduire celles dont il disposait, indifférent à leur emplacement.
Dans un livre d’heures à l’usage de Saint-Omer produit après 1318 et
faussement identifié comme celui d’une Marguerite de Beaujeu (New York,
Pierpont Morgan Library, MS. M. 754 et Londres, BL, MS. Add. 36684), trois
plaisanteries blasphématoires montrent que le peintre (ou son éventuel
conseiller) lisait et comprenait parfaitement les prières. Au fol. 27 du manuscrit
Morgan, un singe tire le vin du derrière d’un manchot avec un fausset de
barrique pour remplir un calice. Un peu plus loin, au fol. 35, un singe défèque
dans un calice devant une dame et, au folio 48v, un singe boulanger enfourne un
derrière, tandis que son compagnon ramène des pains sur une planche tout en
désignant le texte de l’index (fig. 6). Ces plaisanteries ornent l’office du Saint
Sacrement et accompagnent pour deux d’entre elles le texte de la consécration
eucharistique. La troisième correspond à l’hymne Pange lingua, le singe attirant
l'attention sur les paroles ‘mysterium corporis et sanguinis sui’. Plus générale
ment, les marges de cet office sont décorées de nombreux calices comme celles
de l’office des morts le sont de crânes et de squelettes. Le peintre ne devait pas
être beaucoup plus pieux que la dame à laquelle le livre était destiné mais, à
moins qu’un clerc ait veillé à côté de lui pour lui suggérer les mauvais coups, on
admettra qu’il comprenait parfaitement la liturgie.
L’obsession du rapport texte / image a souvent fait ignorer aux chercheurs les
rapports qui se tissent entre les marges et les images principales. Dans bien des
cas où, à première vue, la marge se rapporte au texte, il est bien plus probable
que l’artiste commente cette image ou s’inspire du répertoire auquel elle
appartient. A titre d'exemple, l’illustration du psaume 52, 'Dixit insipiens’, se
prolonge souvent dans la marge de manière évidente. On y trouve parfois un fou
(Londres, British Library, MS. Burney 345, fol. 70; Copenhague, Kongelige
Bibliotek, MS. 3384, fol.98). Dans le psautier de Louis le Hutin, conservé à la
" L 'art au temps des rois maudits. Philippe le Bal et sesfils (1285-1328) (Paris, 1998:
Exposition. Grand Palais), n° 212, p. 313.
296 Jean W irtii
Fig. 6. Livre d ’heures, New York, Pierpont Morgan Library. MS. M. 754,
fol. 48'', ‘Singes boulangers ’. © Pierpont Morgan Library, New York
Fig. 7. Psautier gantois, Oxford, Bodleian Library. MS. Douce 5, foi 146, ‘Ecole
des singes © Bodleian Library, Oxford
Fig. 8. Psautier-Heures de Geoffroy d ’Aspremont et d'Isabelle de Kievraing,

Oxford. Bodleian Library, MS. Douce 118, fol. ¡01'. ‘David en singe couronné'.
© Bodleian Libraiy, Oxford
cathédrale de Tournai (fol. 92), un lapin mange un objet rond comme le fait
souvent l’insensé du psaume. Dans le psautier gantois Douce 5-6, l’école des
singes occupe le bas de page (Oxford, Bodleian Library, Douce 5, fol. 146)
(fig. 7); il ne s’agit certainement pas d’une coïncidence, car l’enlumineur de ce
livre a trouvé une iconographie marginale appropriée pour d’autres grandes
divisions du psautier. L’exemple du psautier d’Ormesby (Oxford, Bodleian
Library, Douce 366, fol. 72r), est l’un des plus éclairants. En choisissant de
placer les tentations du Christ dans l’initiale, l’artiste renonçait à y mettre le
couple plus usuel du roi et du bouffon, mais il peint dans la marge une aventure
du roi Salomon et de son bouffon Marcolfe.
Les motifs provenant des initiales du psautier envahissent souvent le psautier
tout entier et, au delà, peuvent être utilisés pour décorer n’importe quel type
d’ouvrages. C’est le cas du fou du psaume 52, mais aussi celui du roi David
jouant de la harpe qui peut devenir un singe couronné, comme dans le psautier de
Geoffroy d’Aspremont et d’Isabelle de Kievraing (Oxford, Bodleian Library,
Douce 118, fol. 101v) (fig. 8), tandis que sa victoire sur Goliath est parodiée par
celle d’un lapin terrassant un chevalier avec une fronde dans le pontifical de
Renaud de Bar (Cambridge, Fitzwilliam Museum, MS. 298, fol. 7). L’apparition
de tels sujets hors contexte dans ces deux derniers exemples témoigne d’une
forme d'indifférence au texte. Du reste, le peintre n’a pas besoin de s’intéresser
au contenu des psaumes ou même de le connaître pour parodier dans la marge un
répertoire d’initiales qui lui était de toute manière familier. Mais on comprend
que ces pratiques puissent égarer les chercheurs: la thématique des psaumes étant
répétitive (musique, prière, triomphe sur l’ennemi, angoisse face à ses succès,
péché et pénitence), la tentation est grande de chercher une allusion chaque fois,
par exemple, qu’il est question de combat dans le texte, tandis que des animaux
s’affrontent dans les marges.
Face à ce risque, il faudrait non seulement vérifier les contextes dans lesquels
un motif est récurrent, comme y invite notre troisième règle, mais encore la
récurrence des procédés d’illustration à l’intérieur d’un même ouvrage. On ne
peut pourtant faire de cette vérification une quatrième règle, car ce serait
présupposer une consistance du décor qui ne se vérifie que rarement. 11 y a dans
les manières de faire médiévales une exigence de systématicité et un manque de
systématicité dont la présence simultanée nous déconcerte. Le paradoxe prend
souvent la forme d’un écart entre la cohérence du projet et les aléas de la
réalisation. En l’occurrence, le décor marginal met en oeuvre des procédés
systématiques et répétitifs, mais il est rare que le décor d’un manuscrit se
poursuive de bout en bout selon les principes adoptés dans les premières pages.
Les peintres avaient certainement des difficultés à planifier correctement leur
travail, à prévoir le temps nécessaire à une réalisation et à la mener à terme sans
se laisser distraire par d’autres tâches, afin de respecter les délais. 11 se peut aussi
que l’impatience des commanditaires ait contribué à des simplifications de
programme en cours de route. Nombre de décors marginaux s’arrêtent
298 Jean W irth
brutalement, se raréfient progressivement ou deviennent de plus en plus répétitifs

et sommaires vers la fin du livre.
Il est encore plus étonnant de constater que la programmation du décor
marginal devait souvent être très lâche. Il est clair qu’on décidait au départ de la
fréquence des marges enluminées. 11 s’agissait soit de les décorer toutes, soit de
concentrer le décor marginal à un certain niveau d’articulation du texte, par
exemple aux grandes divisions du psautier ou aux pages initiales des livres de la
Bible. En revanche, dans une majorité de cas, aucun choix ne semble avoir été
effectué entre des drôleries indépendantes du texte, comme des scènes de chasse
répétitives, des illustrations du texte qui pourraient tout aussi bien se trouver
dans l’initiale, comme David et Goliath au début du psautier, des parodies de
l’image principale, comme des animaux associant leur chant à celui des chantres
du psaume 97, ‘Cantate Domino’, des compléments de l’image principale,
comme des bergers complétant ceux de l'Annonce aux bergers dans les Heures
de la Vierge, ou enfin des scènes renvoyant à des textes autres que celui du
manuscrit, par exemple au Bestiaire ou aux fables.
On pourrait y voir une inconséquence ou supposer que l’emplacement des
marges en faisait une tâche secondaire et relativement indifférente, de l'ordre du
remplissage. Il est vrai que l’organisation de la page subordonne naturellement la
marge et que les programmes des images principales sont en général nettement
plus codifiés. Inversement, il faut admettre que dans la grande majorité des cas,
le ou les mêmes peintres s'occupaient des deux parties du décor: la qualité des
drôleries est rarement inférieure à celle des images principales et le peintre
semble le plus souvent le même, si l’on évite de supposer des mains différentes
chaque fois qu’une réduction de l’échelle entraîne une simplification des formes.
Plutôt qu’une tâche de rang inférieur, le décor de la marge est un exercice plus
libre, une occasion de faire place au caprice, de laisser libre cours à sa fantaisie
et de stimuler celle du lecteur. En conséquence de quoi, une drôlerie faisant
allusion au texte ou à l’image principale dans un manuscrit n’implique
absolument pas que d’autres le fassent et, inversement, une longue série de
drôleries dépourvues de signification n’exclut pas un soudain clin d’oeil à un
passage qu’on désire mettre en valeur.
Le problème de l’interprétation des drôleries est encore compliqué par un
ingrédient qui défie le raisonnement iconographique: le recours délibéré au non-
sens. Cela ne leur est pas propre, puisque les genres poétiques du fatras et de la
fatrasies reposent entièrement sur le non-sens.34 Dans les marges comme dans
ces poèmes, aucun procédé n’est plus courant que Vadynaton, la représentation
d’objets inexistants, comme les êtres hybrides, d’événements impossibles,
comme la poursuite d’un chien de chasse par un lapin, et d’actes absurdes tels
que l’utilisation d’une mâchoire d'âne comme archet pour jouer de la vièle. Ces
’4Cf. L. C. Porter. Lafatrasie et le fatras. Essai sur la poésie irrationnelle en France

au Moyen Age (Genève: Droz, 1960).
monstruosités peuvent être produites par substitution d’un objet à un autre, ou

encore par inversion des rôles, ainsi lorsque le gibier chasse le chasseur. Il n’y
aurait pas de difficulté, s’il suffisait de considérer le non-sens comme un non-
sens, mais il n’en va pas ainsi, car ce qui n’a pas de sens littéral peut avoir un
sens symbolique. Dans certains cas, il faut admettre que la conjonction de
l’inconsistance du rapport texte / image et de la propension au non-sens rend
impossible de décider avec certitude si telle rencontre entre le texte et l’image est
intentionnelle ou relève d’une coïncidence. Nous avons donné un exemple de ce
genre, celui des lapins attaquant un château dans le bréviaire de Renaud de Bar,
que Davenport met en rapport avec le texte: ‘Expugna expugnantes me'. Le
problème ne concerne pas que le rapport texte / image. Comme nous l’avons déjà
remarqué, le nom de ‘conin’ donné au lapin permet un jeu de mot obscène qui
explique certainement le motif du chevalier laissant tomber son épée, terrifié par
un lapin. Mais faut-il prendre chacun des milliers de conins qui habitent les
marges comme une allusion sexuelle?
De fait, le conin apparaît le plus souvent dans des scènes de chasse et la
chasse est à son tour une métaphore fréquente du rapport entre les sexes. En
même temps, il serait impensable que la chasse et les conins ne soient pensés que
dans cette dimension et que leur représentation évoque systématiquement la
sexualité plutôt que l’exercice physique en plein air ou le civet. Le simple bon
sens fait supposer que d’une part le contexte plus ou moins érotique de la
représentation, d’autre part la tournure d’esprit du lecteur du manuscrit, grivois
ou naïf, influenceront la compréhension du motif. Mais ce constat ne mène pas
loin. L’idée que la signification d’un thème iconographique n’existe pas absolu
ment, ne dépend pas de l’intention de l’artiste, mais varie selon le contexte et
selon la perception du spectateur est aujourd’hui courante; elle a été adroitement
théorisée pour l’image liturgique chrétienne par Staale Sinding-Larsen.35 Elle
nous semble avoir le désavantage de conduire trop aisément au présupposé que
toutes les lectures se valent et sont également pertinentes pour une oeuvre, qu’il
n’y en a pas de bonnes et de mauvaises. Fait-il même privilégier la lecture de la
personne à laquelle l’oeuvre était destinée? Et faudrait-il accepter aussi celle que
fait aujourd’hui l’étudiant débutant, lorsqu’il confond les attributs des saints?
Cette conception de l’iconographie pourrait être baptisée, de manière un peu
malveillante, le point de vue du consommateur. Elle témoigne en tout cas d’un
rapport à l’oeuvre qui privilégie le spectateur et la réduit à son utilisation, à sa
consommation en somme. Deux ouvrages influents ont beaucoup fait, chacun à
sa manière, pour légitimer cette attitude: la Phénoménologie de la perception de
Merleau-Ponty et Le musée imaginaire de Malraux. Dans les deux cas, l’oeuvre
est réduite à la perception d’un Français du milieu du siècle que son éducation
pousse à fréquenter les musées.
3> S. Sinding-Larsen, Iconography and Ritual. A Study of Analytical Perspectives

(Oslo: Universitetsforlaget, 1984), pp. 147 et ss.
300 Jean Wirth
Même s’il était possible de spéculer avec quelques chances de succès sur la
psychologie du spectateur médiéval, ce serait de toute manière secondaire par
rapport à la compréhension de l'oeuvre comme un dispositif pensé par son auteur
à l’aide de codes, pour être compris par des destinataires compétents. Cela ne
signifie pas que l’artiste ait nécessairement prévu une lecture unique et univoque
de l’oeuvre. Une représentation littérale du sacrifice d’Abraham doit être
identifiée comme telle pour avoir un sens, mais elle peut se prêter à plusieurs
lectures typologiques également acceptables. Le sculpteur d’un crucifix n'attend
probablement pas les mêmes réactions des différents publics auquel il sera
exposé. Pourtant, l’image est précisément codée et l'identification du supplicié
comme le bon ou le mauvais larron serait une erreur inexcusable de la part d’un
chrétien jouissant de toutes ses facultés. Cette affirmation ne nous contraint pas à
considérer que chaque lapin est ou n'est pas un symbole sexuel. 11 faudrait pour
cela qu’un code ait été mis en oeuvre pour lever l’ambiguïté, par exemple qu’un
attribut particulier soit affectée au lapin lorsqu’il joue ce rôle. On remarquera au
contraire deux choses: d’une part, que l’attribution d’une valeur symbolique,
sérieuse ou ludique, à un motif marginal, se faisait généralement sans crier gare;
d’autre part, que la notion de symbole sexuel n’existait pas au Moyen Age et que
l’artiste ne risquait pas plus de signaler de tels symboles que de mesurer en
centimètres les dimensions d’une enluminure.
C onclusion
Entre la négation du sens des drôleries par Mâle et la recherche d’un sens
spirituel à chaque drôlerie par Davenport, il y a place pour une interprétation
satisfaisante. Il faut partir du constat que la densité iconographique de la marge
est le plus souvent faible. La grande majorité des motifs qui la peuplent évoquent
le monde de la chasse, de la musique, de la danse et des jeux, dans le but de
flatter les goûts d’une clientèle aristocratique dont ce sont les passe-temps. Dans
certains cas, comme le psautier d’Isabelle de France, reine d'Angleterre
(Munich, Bayerische Staatsbibliothek, Cod. gall. 16), le commanditaire semble
avoir désiré pour un livre liturgique un décor plus conforme à sa fonction et
demandé le remplacement total ou partiel des drôleries par des illustrations
marginales. Dans d’autres cas, comme dans le psautier-heures de Jeoffroy
d’Aspremont et d’Isabelle de Kievraing, une allusion au texte sacré de temps à
autre semble servir de légitimation aux drôleries. Mais la majorité des allusions
qu’on peut déceler dans les marges n'ont aucun caractère dévot et sont destinées
à provoquer le rire ou le sourire, ainsi lorsqu’on se moque de la couardise du
chevalier, de la lubricité du moine ou même du sacrement de la messe.
‘Primitive’ paintings:
the visual world o f populus rusticus
H elena E dgren
ost medieval pictures dealt with in (art) historical research and writing
M are bound together by one common trait: They are all part of the same
European artistic tradition. The pictures may vary as to the degree of
skill with which they have been executed, but they can all be thought of as fruits
of the same tradition. The meaning of the pictures, their inner content, is not
always clear and obvious, but each individual image can be studied by
comparing it with other pictorial products of the same age, or through the use of
literary sources. All of these works of art represent a common, Pan-European
culture rooted in the civilizations of Antiquity and fostered by, or connected to,
the Catholic Church.
Nevertheless, in the Scandinavian countries, as elsewhere in Europe, there
also flourished another kind of medieval art, the so-called ‘primitive’ paintings,
whose creators were not professional artists trained in the European painting
tradition (fig. 1). This was a widespread phenomenon: Finland has about 100
surviving medieval churches, and roughly half of these contain ‘primitive’
paintings. In Denmark, ‘primitive’ paintings have been recorded from some 150
churches to date.' In both Finland and Denmark, ‘primitive’ paintings were
1 Ullu Haastrup. 'Danske kalkmalerier 1475-1500'. in Danske kalkmalerier. Sengotik

1475-1500. ed. by Ulla Haastrup (Copenhagen: Nationalmuseet, 1991), p. 26; Katrin
Kroll. 'Die Komik des grotesken Körpers in der christlichen Bildkunst des Mittelalters’,
in Mein ganzer Körper ist Gesicht. Groteske Darstellungen in der Europäischen Kunst
und Literatur des Mittelalters, ed. by Katrin Kroll and Hugo Steger Steen (Freiburg im
302 H elena Edgren
Fig. I. Church of St Mary (Maaria), Southern Finland, 'Primitive paintings’.

© Author
probably more numerous early on, but in the course of time they have been
replaced by newer and more professionally executed illustrations. Many
medieval church paintings are still to be found beneath the plaster. Accurate
figures are not available for Swedish paintings, since they have not been studied
as systematically as their Finnish and Danish counterparts. The ‘primitive’
paintings still suffer from the label of ‘second-rate art’ that they received a
hundred years ago, and consequently they have not been able to compete with
finer paintings for the attention of art historians. They are also much more
difficult to interpret than so-called finer art. As a Swedish example of the genre
one could mention St Birgitta’s own convent church in Vadstena, where
paintings classifiable as ‘primitive’ adorn the high vaults.
‘Primitive’ art, however, is not limited to paintings on the plastered walls of
stone churches. A large number of carvings, found on the walls of Norway’s
wooden stave churches, belong to the same class of art (fig. 2). The Norwegian
carvings have not been sketched on the wall with pigment but, rather, have been
cut into the wall boards with a sharp utensil such as a knife. As the walls have
become soiled and darkened during the centuries, the carvings have become very
Breisgau: Rombach, 1994), pp. 11-105; Steen Schjödt Christensen. 'Mysterious

Images—Grimacing. Grotesques. Obscene, Popular: Anti- or Commentary Images?’,
Medium Aevum Quotidianum. 39 (1998), 55-75.
Primitive paintings 303
Fig. 2. Church of Borgund, Norway, ‘Graffiti’, from Martin Blindheim, Grafitò

in Norwegian Stave churches c. 1150-c. 1350 (Oslo: Universitetsforlaget, 1985)
obscure. Their existence only became generally known in art-historical circles in
1985, when Martin Blindheim published his exhaustive study: Graffiti in
Norwegian Stave Churches.2
2 Martin Blindheim, Graffiti in Norwegian Stave Churches c. / 150-c.1350. Medieval

Art in Norway (Oslo: Universitetsforlaget, 1985).
304 H elena Eixiren
A large number of similar carvings have been recorded in the churches of

Gotland, where the figures have been scratched in the wall plaster. Carvings
have been found mainly on the interior surfaces of the walls, but in some cases
also on exterior surfaces.'
What, then, are these ‘primitive’ paintings and carvings? Who made them,
and on whose order? What is their content? What do they tell us about the
people who made them and perceived them, about how they conceived
themselves and their world? Many questions concerning these pictures have yet
to be satisfactorily answered. From the beginning, art historians have been
totally baffled by these paintings. They have always been thought of as repre
senting the ‘others’, the 'populus rusticus’ who, in the Middle Ages, formed a
group of their own. unreachable by ordinary research methods because they left
such scant traces of themselves in written documents.
In Finland, the first ‘primitive’ paintings were discovered in 1880 (fig. 3).
The study of medieval wall paintings had begun about ten years earlier and had
produced some impressive results: paintings, which could not be compared to
anything known thus far from the other Nordic countries. When paintings of a
totally different kind appeared from under the plaster in 1880, art historians were
both surprised and helpless; they felt totally incapable of interpreting and
understanding this new art form (fig. 4). In his research report, one art historian
described how impossible it was to describe what he observed: ‘Formulating the
description turned out to be somewhat more problematic than I had expected.
Due to the extraordinary ornaments that differ from anything that one normally
sees in in secco painting. [ ...] I had to try to develop a new kind of terminology
in order to be able to give the reader at least some form of verbal description of
what the eye perceived’.1 Here was, in an extreme form, the art historian’s
perpetual problem how to render the visual world into words.
The first reaction of the art historians was not a clear rejection but, rather, a
hopeful curiosity. Emil Nervander, one of Finland's leading art historians in the
late nineteenth century, recorded the paintings carefully and even planned to
publish them as a series of lithograph plates. He saw in them a trace of ‘Irish
style mixed with recollections of Swedish rock carvings’ and dated them to the
thirteenth century, when the Catholic Church finally had secured its position in
Finland.3*5 However, just to be on the safe side, he also wrote to the celebrated
3 Uaininn O’Meadhra, ‘Medeltida ristningar i Gotlands kyrkor. Hikuin 24 (1997),

229-31.
J Emil Nervander's 1900 report (on the wall-paintings of Pemâ church) lodged in the
topographical archive of the Helsinki National Board of Antiquities Department of
History: Berättelse om verkställ undersökning cif Penici kyrkcis kalkmálningar afgifven af
E. Nervander 15.1.1900: Helena Edgren. Primitiva mãlningar? Bild och känsla frein antik
till nyantik. Picta Nr. 3 (Âbo: Ábo Akademi. 1994).
5Edgren (as in note 4).
Primitive' paintings 305
Fig. 3. Church o f Nousiainen, Southern Finland, ‘Primitive painting’.

© National Board of Antiquities, Picture Archives
director of the Danish National Museum, Sophus Miiller, and asked for his
opinion on the paintings. Müllers answer was crushing. He saw no evidence of
Celtic or thirteenth-century influence; instead, he thought the paintings were
considerably later and totally unsuitable for publication or study. A few might be
recorded as examples of the art form, but the rest were not worth wasting one’s
time on. Müller thought the paintings were extremely primitive, an appellation
they have borne from that day onwards. Nervander, honest scientist that he was,
published his correspondence with Müller and abandoned his project.'’ Müllers
negative stance came to dominate the attitudes of later generations of art
historians toward this genre, and in many cases has continued to do so to this6
6 Emil Nervander’s report on Sophus Müllers 1903 comments on the wall-paintings

of Nousis church, lodged at the Suomen museum in Helsinki: Uttalande af Sophus Miiller
om kalkmálningarna i Nousis kyrka (Suomen museo. 1903).
306 H elena E dorkn
Fig. 4. Church o f Nousiainen, 'Primitive paintings’. ©Author
Fig. 5. Church o f St Mary,‘St Olav o f Norway or St Erik o f Sweden’. © Author

'Primilive ' paintings 307
Fig. 6. Church of St Mary, 'Devil, primitive painting’. © National Board of

Antiquities, Picture Archives
Fig. 7. Church of Korppoo, Southern Finland, ‘Ship ’, probably depicting the

journey o f St Olav o f Norway to Finland. © Author
308 H elena Edgren
day. Many of the 'primitive' paintings uncovered during the nineteenth century
were actually recorded on paper or film but were afterwards covered up again.7*9
Only towards the end of the twentieth century have we begun to uncover them
once more from underneath the plaster. The 'primitive' paintings of Finland now
have been added to an iconographie catalogue based on the standard Nordic
model, though their description and classification continue to pose great
problems. The first Finnish study on 'primitive' paintings was published in the
1970s,s but serious general art-historical interest in the genre has caught on only
during the last decade. Previously, 'primitive' paintings received just a passing
mention, if even that.
The motifs in these ‘primitive’ paintings and carvings were largely the same,
but certain differences also can be noted. The differences are partly caused by
the wide age distribution of the paintings and carvings. The carvings in the stave
churches date from between 1150 and 1350.''The Gotland carvings date from the
thirteenth century to the end of the Middle Ages and the beginning of the
Modern Era,"' and the Finnish and Danish paintings primarily stem from the late
Fifteenth century."
The edging ornaments found on vault ribs appear only in painted form. These
ornaments include different kinds of plant, cross, and roundel patterns. The
diverse individual motifs, on the other hand, are shared by both the paintings and
the carvings. These motifs include various kinds of human figures, such as clear
representations of saints (fig. 5). musicians, soldiers, artisans, and other less
easily interpreted figures. The individual 'personages' portrayed also include
several unmistakable devils (fig. 6). The numerous animal figures include
horses, birds, dogs, foxes, lions, and dragons. Of these, the horses and lions are
especially typical in the stave church carvings. In Norway, horses have been
thought to refer to the old Nordic horse cult, while the lions have been
considered to reflect both the domestic tradition and a strong influence of the
Romanesque art of continental Europe.12The motifs also include various kinds of
ornaments and magical signs known from later folk art, including looped crosses
and pentagrams. Other common motifs are labyrinth patterns and various kinds
of ships and boats (fig. 7).
7C. Frankenhaeuser, 'Räntämäen kirkko ii. Maalaukset", Suomen museo (1910). 9-12.
s Anna-Lisa Stigcll. ‘Kyrkans lecken och ârets gang. Tideräkningen och Finlands
primitiva medeltidsmâlningar'. Filistea Fomminnesßreningens Tidskrifi 77 (1974).
9Blindheim (as in note 2).
'"O'Meadhra (as in note 3), p. 227.
" Ulla Haastrup. "Danske kalkmalerier 1475-1500’. in Danske kalkmalerier. Sengotik
1475-1500, ed. by Ulta Haastrup and Robert Egevang (Copenhagen: Nationalmuseet.
1991). p. 26: Edgren (as in note4): Steen Schjpdt Christensen (as in note 1).
12Blindheim (as in note 2). pp. 60-63.
'Primitive ’paintings 309
Rune texts also are found in Denmark, Sweden, and Norway. Some of these
are clear textual wholes, often with a religious content. Others are individual
runes, which have been considered magical marks used to ward off evil. In
Finland, where runes were not used, known texts were written in the Latin
alphabet. These also included both actual texts and others that try to look like
text but are unreadable, as well as individual characters, such as Virgin Mary
monograms (fig. 8).
‘Primitive- art also employs a large selection of Christian symbols, such as
the Tree of Life and various types of crosses. Coats of arms are also common
(fig. 9). All of these have been studied to some extent using art-historical
methods, that is. analyses of content, contacts with other forms of visual art. and
style, so I will not dwell upon this aspect any further. 1 will instead attempt to set
myself outside the sphere of actual art-historical research and use the paintings
as historical sources. In other words. 1 shall present a few attempts at interpreting
whose worldview this form of art reflected, who painted them, who determined
their content, and what sort of spiritual contacts the pictures describe. In order to
achieve this goal, one must first reject the tenacious preconceptions concerning
the reception of the paintings during the era in which they were originally
painted, in other words, what the medieval person would have thought of them.
The traditionally doubtful and negative attitude and the strangeness or
‘otherness’ of the ‘primitive’ paintings have the effect that art historians
sometimes, even nowadays, find it hard to believe that they have not always
been looked upon with a similar feeling of disdain. As late as 1997, a publication
on the Gotland material poses questions like ‘Who has doodled and sketched on
the church walls during the Middle Ages, and why? How could it have been
allowed? Was it done in secrecy, or with the permission of the priest?’1314When
one looks at ‘primitive’ paintings without preconceptions, however, it seems
clear that they were as accepted in their own time as any other form of
ecclesiastical art. This, at least, was the situation in Finland.1'’
The situation can be illustrated by a few examples. Perhaps the most
convincing of these is the fact that ‘primitive’ paintings are found even in
Finland’s central shrine, the bishop’s own cathedral in Turku (fig. 10). It is quite
inconceivable that images not approved by ecclesiastical authority could have
been painted in a cathedral where the bishop and his clergy celebrated mass
daily, and where devotional activities continued at the numerous altars almost
without break from dawn till dusk. If any inappropriate paintings had appeared
on the cathedral walls, one cannot doubt that they would have been promptly
removed. The ‘primitive’ paintings, however, are still there.
There are also cases where the ‘primitive’ paintings that had been acquired in
13O’Meadhra (as in note 3). p. 234.

14Edgren (as in note 4).
310 H elena E dgren
Fig. 8. Church o f St Mary, ‘M a r y © A u t h o r

'Primitive 'paintings 311
Fig. 9. Church o f St Mary, 'Coat o f arms’. © Author

312 H elena E dgren
Fig. 10. Cathedral o f Turku. 'Primitive painting'. ©Author

'Primitive ' paintings 313
Fig. y/. Church o f Karjaa, Southern Finland, ‘Apostle ’ from a credo series.
© Author
314 H elena Edgre.n
connection with the building of the church or the erection of the ceiling vaults
remained uncovered when so-called finer paintings were ordered. This is the
case in, for example, the Karjaa church on the south coast, where ‘primitive’
paintings adorn the vaults while finer paintings cover the walls. The situation at
Karjaa is exceptional in the sense that both series of paintings have been
executed on the same plaster surface. Their age difference cannot therefore be
very great, and both in fact have been dated to the 1470s based on a painted date
that appears among the ‘primitive’ paintings. We are thus dealing with a case
where someone, perhaps a member of the building team, finished the work by
decorating the vaults with ‘primitive’ paintings. As soon as the congregation’s
finances allowed it or a suitable painter was found, the walls were decorated
with an impressive credo mural (fig. 11). The ‘primitive’ paintings, however,
suffered no harm because of this operation; instead, the two series of paintings
form a single entity. The local patrons thus accepted both equally as decorations
to their church.
Danish art historians have wondered why the churches should have suddenly
been content to have in secco paintings composed of scattered, apparently
meaningless ornaments and figures, even though instructive and content-charged
murals were being created at the same time. How could such pictures have
satisfied patrons whose taste, under the influence of a lengthy tradition, had
become accustomed to images with a clearly defined theological content?15One
possible answer lies in recalling the generally known fact that the primary value
of the paintings lay not in their aesthetic appearance but. rather, in their symbolic
meaning. One also can ask, what was the aesthetic training of the medieval
person? Most people, including the clergy and the lay public, had none at all.
The visual arts were not represented in any way in the curricula of medieval
schools and universities.16The wealthy, and those who had studied abroad, might
have seen fine art, but this hardly applied to the domestically trained rural clergy
or the lower nobility. Educational books were normally cheap copies with no
illustrations. Illustrated versions were available only to the wealthy few. For
most people, the only contact with the visual arts was through cheap woodcuts,
which were also often extremely clumsy. At least in Finland, only the richest and
best-trained classes had permanent access to the products of the European
painting tradition. Judging from heraldic church paintings (coats of arms)
(fig. 12), they were also the people who ordered most of the high-quality
paintings.17
15Haastrup (as in note 11). pp. 27-28.

16 Jussi Nuorteva, Suomalaisten ulkomainen opinkäynti amen Turun akatemian
perustamista 1640 Suomen kirkkohistoriallinen seura (Helsinki: Suomen kirkko-
historiallinen seura, 1997).
1 Camilla Ahlström-Taavitsainen, Vapenbilder bland kalkmâlningarna i Finlande
medeltidskyrkor in Publikationer utgivna av konsthistoriska Institutionen vid Helsingfors
'Primitive ' paintings 315
The ‘primitive’ paintings, however, also include a large number of coats of

arms and ownership marks. That indicates that both the clergy and the local
persons of note wanted to be remembered for their part in building and
decorating the churches and, as Ulla Haastrup has noted, to be symbolically
present at the celebration of Mass.*18The paintings therefore cannot be connected
with only the lowest class. Rather, they represent the common aspirations of the
whole congregation.
There are cases from both Finland and Denmark where a construction team
that had been working on a church painted ‘primitive’ decorations on the walls
and vaults as a parting shot, as it were.19 By the late fifteenth century, the
paintings had become such an essential part of the church interior that the church
was not considered ‘finished’ without them. Both countries also had identifiable
groups of painters who worked on several different churches. The Danes have
even started to refer to ‘primitive’ paintings as ‘murermesterbemalinger’ (master
mason paintings).20 In Finland, however, this explanation does not cover the
whole field. For instance, some churches have ‘primitive’ paintings dating from
periods that cannot be connected with any notable construction episode. In one
case, several images of ships have been painted on the same place on the church
wall, so that the older figures were superimposed.21 These are votive paintings,
forerunners of the later custom of giving ship-models to the church as votive
offerings.
Although a local building team might have executed the paintings, this does
not mean that the team or the master builder made the final decision on the
choice of motifs and the meaning content. Even when the paintings are purely
ornamental decorations devoid of figurative representations, the paintings may
nevertheless have a deep religious significance.22This is the case in, for example,
the church of Finström in the Àland Islands, where the clumsy stars in the vaults
portray the history of the world, from the Creation to the Last Judgement.22
Even in cases where it is difficult to see any coherent plan and the placement
of the paintings appears totally fortuitous, one must take into account the general
rules and regulations concerning the acquisition of pictures by churches. One
urtiversitet. 8 (Helsingfors: Helsingfors universitet. 1984).

18Haastrup (as in note 11). p. 26.
19 In Denmark, see Haastrup (as in note 11), p. 28. In Finland, e.g.. Karjaa church.
Southern Finland.
20e.g. Haastrup (as in note 11), p. 26.
21Maaria (St. Mary) church in Finland proper.
22 Compare Niels M. Saxtorph. ‘Markblomstmaleren. nogle værksteder’. in Danske
kalkmalerier. Sengotik 1475-1500 (as in note 11), p. 89.
22 Valdemar Nyman. Âlands medeltida kyrkokonst. Det äländska folkets historia 1,2.
(Mariehamn: Âlands kulturstiftelse. 1980). p. 98.
316 H elena Eix '.ren
Fig. 12. Church o f Taivassalo, Southern Finland. 'Coat o f arms'. ©Author

Primitive ’ paintings 317
Fig. 13. Church of St Mary. ‘Master mason and his assistant'. © Author
Fig. 14. Church o f St Mary, ‘Dominican fria r’. ©Author

318 H elena E dgren
Finnish example of control by ecclesiastical authorities dates from the 1480s,

when the bishop and chapter of Turku forbad the churches to hire artisans,
including painters, without the chapter’s approval. Not even votive offerings
could be brought to the church without the local priest’s permission.24This order
sometimes has been considered to reflect the clergy’s negative attitude towards
‘primitive’ paintings,25 but it is, in fact, just one of a series of medieval statutes
also known from other countries besides Finland. Whatever the nature of the
‘primitive’ paintings, we must assume that they were approved at least by the
local priest or clergy, who also, most likely, had a say as to their content. The
background of the rural parish priest was probably only a little different from
that of his parishioners, and their worldview would have been largely identical,
characterized by the same traditional beliefs and notions.
Finally, I would like to present two Finnish churches as examples of the wide
range and variability of ‘primitive’ paintings. One is the church of Maaria in
western Finland, dedicated to St Mary, and the other is the church of Pyhtää in
eastern Finland, consecrated to St Henry of Finland. Maaria church lies only
some ten kilometres from Turku, the country’s medieval centre, and thus was
located within the immediate sphere of influence of the bishop himself. The
church building has been thought to emulate Turku Cathedral in its vault
structure. Nevertheless, the paintings in this church represent the ‘primitive’
style in its most disorganized form. The walls and vaults are full of individual
figures. Most of these have a Christian content, but their mutual placement does
not appear to be governed by any clear principle. Some of the figures also
represent a pictorial world that has no connection with Christian tradition.
The decoration of the third vault of the north aisle has the clearest structure.
The vault panels contain both texts and monograms that refer to the Virgin
Mary, and in all probability the vault has been the site of an altar consecrated to
Our Lady. On the north wall, two human figures are clearly benefactor portraits
(fig. 13). They are probably the master builder of the church and his assistant,
who have placed themselves under the perpetual protection of the Holy Virgin.26
Another vault contains a different ‘portrait’, a Dominican friar standing in front
of a small church and holding a cross in his hand (fig. 14). The Dominicans were
the most prominent ecclesiastical order in Finland, and the sight must have been
familiar to all parishioners.27 The paintings also include several obvious
24 Finlands medeltidsurkunder, ed. by Reinhold Hausen, 8 vols (Helsingfors:

Statsarkivet. 1910-35), ¡V: 1451-1480. document 3853.
25 Tove Riska, Keskiajan maalaustaide. Ars. Suomen laide 1 (Helsinki: Weilin &
Göös, 1987), p. 144.
26Helena Edgren, Donatorsfrâgor (Helsinki: Suomen museo, 1989).
27 Helena Edgren, ‘Dominikanbrodern i S. Marie kyrka’, in Den monastiska
bildvärlden i Norden. Föredrag vid det 10:e nordiska symposiet för ikonografiska studier
Primitive ' paintings 319
Fig. 15. Church of St Mary, 'Devil'. © National Board o f Antiquities, Picture

Archives
references to saints, including a ship with crew presumably portraying the
journey of St Olaf. A song describing this journey was a part of the standard
repertoire of pilgrims visiting the grave of St Olaf in Trondheim.25 Perhaps the
most striking of the figures nevertheless may be the Devil (fig. 15), who
undoubtedly has been included as a warning against the greatest temptations of
the male world: the excessive craving for money and sexual pleasures. In 1910,
when the paintings were studied for the first time, the figure was considered so
indecent that it was immediately covered up.-'
In the church of Pyhtää the situation is totally different. In spite of their
‘primitive’ nature, the paintings form a coherent whole with a lucid theological
content. In this case, the guiding influence of a trained theologian, the local vicar,
is indisputable. Among the ‘primitive’ coats of arms, one might have belonged
to the vicar who was in office when the paintings were made. The Pyhtää church289
; Vadstena 1986, ed. by Ann Catherine Bonnier. Mereth Lindgren, and Marian Ullén
(Stockholm: ICO/RiksantikvarieUmbetei. 1990). pp. 11-16.
28 Lindgren, Mereth, ‘St. Olav in der mittelalterlichen Malerei Mittelschwedens. St
Olav, seine Zeit und sein Kult’, Acta vishyensia vi. Vishysymposiet för historiska
vetenskaper ( 1979).
29Frankenhaeuser. Räntämäen kirkko (as in note 7).
320 Helena E dgrf.n
it; it
Fig. 16. Church of Pyhtää, Southern Finland. 'Figures from a credo series and
Christ on the Cross'. © Finnish National Board of Antiquities, Picture Archives
'Primitive' paintings 321
Fig. 17. Church of Pyhtää, ‘The Apocalyptic Madonna and other paintings’.
© Finnish National Board o f Antiquities, Picture Archives
322 H elena E dgren
also contains finer paintings in addition to the 'primitive' set. The most
extensive mural in the Pyhtää church is the so-called credo suite, the twelve
apostles portrayed on the church wall. In accordance with Church tradition, the
first in the series is St Peter, the leader of the apostles, who at Pyhtää has been
placed next to the most important painting, the figure of the crucified Christ
(fig. 16). The apostles have been placed next to the consecration crosses, as was
the custom. There are also other saints depicted on the walls, including an almost
six-metre-high representation of St Christopher, protector from the calamity of
sudden death.
The crucified Christ is portrayed with bleeding wounds in the tradition of the
late medieval Passion mystery. The emphasis on the role of the Virgin Mary,
both as the mourning mother at the foot of the Cross and as the Apocalyptic
Madonna (fig. 17) in the ceiling vault, is totally in accordance with the spirit of
that period.30 The face of the Saviour portrayed in the adjacent vault resembles
the indulgence icon popular in St Birgitta convents.'1 The battle scene placed
next to the Apocalyptic Madonna may symbolize one of the medieval cardinal
sins, Ira (anger), so it also conveyed a clear moral message. Certain other
figures, such as the trumpeter—known also from other churches—are more
difficult to interpret. As a whole, however, the decorations of the Pyhtää church
definitely represent a striving towards the same overall effect as in more
artistically decorated churches, even though the style was what we term as
‘primitive’.
As we have seen, the variability of the ‘primitive’ paintings is great, and they
reflect the spiritual influences through which individuals have constructed their
worldviews in very different ways. The Pyhtää suite is dominated by a lucid and
comprehensive theological idea, which shows that the designer was well in touch
with the theological currents of the times. In the Maaria church, in contrast,
motifs from folk traditions and beliefs are much more evident. Their pictorial
world can be compared to the world known from Finnish oral tradition. Folk
epics and spells; a world based on the same unique admixture of pre-Christian
beliefs and Christian influence, even though the folklore was not recorded until
almost four hundred years after the paintings were made. The common people,
including the priests who often came from their ranks, changed their conceptions
of the world extremely slowly, especially in the road-less forests of Finland. The
inhabitants of these backwoods would undoubtedly have nodded in common
understanding of many paintings that remain unreadable mysteries to us.10
10 See. e.g., Helena Edgren. 'Mercy and Justice. Miracles of the Virgin Mary in
Finnish Medieval Wall Paintings’, Finska Fomminnesföreningens Tidskrift 100 (1993),
28-33
11 Sixten Ringbom. 'Bild och aviat’, Iconographisk Post. Nordic Review of Icono
graphy. 3—4 (1983). voi. 3. pp. 8-16: voi. 4. pp. 1-14.
M an and picture: on the function o f
wall paintings in medieval churches
ANNA NILSÉN
he interiors of medieval churches were often covered with paintings,
T sometimes at considerable expense. In Denmark, Sweden, and Finland

we are lucky to possess a great many series of pictures, a majority of
them from the late Middle Ages and some of them still in a very good condition
(fig. 1).'
Continuous picture programmes covering the whole of the church interior are
seldom found in the larger churches, such as cathedrals and monastic churches.1
1 For Sweden, see, e.g., Henrik Cornell and Sigurd Wallin, Schwedische Kirchen
malereien des 16. Jahrhunderts (Stockholm: Humanistiska Sällskapet, 1954); idem.
Schwedische Kirchenmalereien des 16. Jahrhunderts 2 (Stockholm: Humanistiska
Sällskapet, I960); idem, Stockholmer Malerschulen des 15. Jahrhunderts (Stockholm:
Humanistiska Sällskapet, 1961); Äke Nisbeth, Bildernas predikan (Stockholm: Gidlund,
1986); Anna Nilsén, Program och funktion i senmedeltida kalkmäleri. Kyrkmälningar i
Svealandskapen och Finland 1400-1534. Kungliga Vitterhets Historie och Antikvitets
Akademien, Diss. (Stockholm: Almqvist & Wikseil, 1986). For Danish wall paintings,
see, e.g., A Catalogue of Wall Paintings in the Churches of Medieval Denmark 1100-
1600. ed. by Knud Banning. 4 vols (Copenhagen: Akademisk Forlag, 1976-82); Danske
kalkmalerier. i (1080-1175)-. n (1175-1275); ill (1275-1375); tv (1375-1475); v (1475-
1500); vi (1500-1536). ed. by Ulla Haastrup and Robert Egevang; VII (1536-1700). ed.
by Eva Louise Lillie (Copenhagen: Nationalmuseet, 1985-92). Overviews of wall
paintings in Finland may be studied in Nilsén (as above); and in Tove Riska, ‘Keskiajan
maalaustaide’, voi. I of Suomen Taide (Helsinki: Kustannusosakeytio Otava. 1987), pp.
116-78.
324 A nna N ilsén
Fig. 1. Ärentuna church. Uppland, Sweden. The paintings from c. 1435 are
among the best preserved in Sweden. The Passion series on the north wall, today
damaged by the widening o f the wall for the present large eighteenth-century
windows, was probably well lit by one or two windows on the southern side. It
has a devotional character. However, the picture programme in Ärentuna is, as
a whole, of a mainly descriptive and didactic type. © Author
Usually there are different programmes for each chapel or section of the interior,
whereas in country churches the whole interior normally was covered with
paintings in a single campaign.2
Since there are both architectural and liturgical differences between these
churches on the one hand and the country churches on the other, and since there
is a larger number of paintings preserved in the latter type of churches, their
paintings thus forming a more homogeneous material, I shall base my discussion
mainly on these country churches.
A natural question in connection with medieval Scandinavian wall paintings
is whether they had more than purely decorative functions. Where any
inscriptions, connected with the paintings and mentioning their purpose, survive,
these inscriptions state—I think without exception—that the paintings were
’ The picture programmes of the late medieval wall paintings in Middle Sweden and
Finland have been studied in Nilsén (as in note I). All programmes seem well conceived
from a theological point of view. They are, up to c. 1450, mostly descriptive and epic in
character, and after 1450 mostly composed with a leading idea, a main theme. Here there
is a difference between Sweden and, e.g., northern Gennany.
Man and picture 325
Fig. 2. Roslagsbro church, Uppland, Sweden. The original church consisted of

the present nave, built in the latter part of the thirteenth century. The choir and
the porch were added around 1400. Up to c. 1470, the probable year when the
brick vaults were built, the church had a wooden ceiling. The original, very
small, thirteenth-century windows, walled up in the fifteenth century, when a
couple of somewhat larger ones were made, may still be seen on the southern
wall. The fifteenth-century windows—one of them visible on the wall, close to the
porch—were replaced in the eighteenth century by the present windows. On the
northern wall there were no windows in the Middle Ages. © Author
made to the honour of God and to embellish the church. But, of course, we need
not and should not accept these statements as the only truth. They may just
represent a generally accepted standard formulation. A widespread opinion is
that medieval wall paintings were mainly intended as a pedagogic aid to be used
by the clergy in their sermons. Hans Belting, for instance, concludes that since
the sermons mostly took place inside the churches, where both priest and
congregation could see the paintings, the priest must have taken advantage of
that situation and referred to them in his sermons.3 The opinion that the function
of the wall paintings in medieval churches was mainly instructive derives from
general statements by early Christian authorities, as for instance Gregory the
' Hans Belting, Das Bild und sein Publicum im Mittelalter. Form und Funktion früher
Bildtafeln der Passion (Berlin: Mann. 1981). pp. 244-51.
326 A nna N ilsén
Great (c. 540-604), repeated by later authors through time. However, throughout
the Middle Ages there were many and varying reasons for providing churches
with paintings, which, as we shall see, also entailed many and varying functions.
In some quarters, though, there was no enthusiasm at all for providing churches
with pictures. St Bernard and St Birgitta of Sweden are well-known examples of
this attitude."
The idea that the medieval parish priest referred to the pictures around him
when preaching is anachronistic and based on a modern idea of preaching and
teaching.5 Against such an opinion there are several important arguments, of
which the impossible pedagogic situation is among the weightiest. It is hard to
imagine a good pedagogue trying to make his audience understand which one of
all the pictures on walls and vaults he is referring to, in a badly lit and probably
also crowded nave, especially when he was preaching in many cases far away
from the picture referred to, and only a few persons would have a chance to
come close enough to the pictures to be able to see them. The Swedish country
churches of the Middle Ages had few, and small, windows, in order not to let in
the cold during winter (fig. 2). This meant that the interiors of the churches were
much too dark for the whole body of paintings to be identified by the
congregation. Only in the porches and a few places inside the churches was there
good light, and there are very often to be found didactic pictures, mostly of a
moralising character. Devotional pictures, too, and series of the Passion of Christ
were in general placed on well-lit walls (figs. 1 and 3).6
" Bernard of Clairvaux, Apologia ad Guillelmum, Sancti Bernardi Opera, vol. 3, ed.
by J. Leclercq and H. M. Rocháis (Rome: Editiones Cistercienses, 1963), p. 106; Birgitta
of Sweden, Revelationes extravagantes 31; cf. Lennart Hollinan, Den Heliga Birgittas
Reiielaciones extrauagantes, Samlinga utgivna av Svenska Fomskrifts-Sällskapet, ser. 2.,
Latinska skrifter, band 5. Diss. (Uppsala: Almqvist & Wikscll, 1956), p. 145. A similar
attitude towards paintings in churches is discernible in ‘Pictor in Carmine’ by an
unknown author of the Romanesque era, possibly Adam of Dore, an English Cistercian;
see M. R. James, ‘Pictor in Carmine’, Archaeologia, 94 (1951), 141^2. For the reference
to the article by Dr. James I thank Dr. Phillip Lindley, Leicester.
s This was first argued in Nilsén (as in note 1), pp. 507-15, where the functions of the
paintings in relation to the beholder and other relevant parties are discussed (Gemían
summary, pp. 529-30).
6 Candles were expensive and used with economy. Indirect information on the use of
candles is sometimes given in the provincial laws. Both the older and the younger version
of the law of Västergötland. for instance, mention candles in two places, at the front and
at the back of the church, and probably concerning those of the main altar and the one at
the baptismal font near the exit; see Svenska landskapslagar, ed. by Ake Holmbäck and
Elias Wessen, 5 vols (Stockholm: H. Geber, 1933-46), IV, pp. 14 (note 27), 206. But
even if candles had been used more lavishly, this would rather have added to the mystical
atmosphere than given a good overall light.
327
Man and picture
Fig. 3. Roslagsbro church. The visitor was met in the porch by the picture oj
Mater omnium with an accompanying prayer in Latin. © Author
328 A nna N ilskn
Fig. 4. Ärentuna church, c. 1435. ‘The seven deadly sins with their roots’. The
headline over the picture says ‘Septem peccata mortedia cum radice meaning
‘The seven deadly sins with their roots The instructive picture shows us the sins
and their places in the body of man. their 'roots'. Each sin is represented by a
monstrous head with a sinner in its mouth and is also accompanied by its name
in Swedish. The right portion o f the painting is damaged. From the figure 's left
hand came Invidia, and from his backside the sin considered as the ugliest sin of
all, Superbia. The representation is on the west wall at the end of a series
depicting the Ten Commandments. © Author
Man anti pic ture 329
Fig. 5. Hiirkeberga church. Uppland, Sweden, C. I4S5. ‘The Wheel of Life’, a

'memento mori’ motif showing the inevitable destiny of all and also the vanity of
earthly endeavour. © Author
Fig. 6. Hdrnevi church. Uppland. Sweden, c. 1485. 'True and false prayer'. How
you should pray is illustrated in the porch o f this church and many others. The
old. simply dressed, barefoot man on Christ’s right has all his thoughts on
Christ’s wounds when praying, while the thoughts of the fashionably-dressed
young man on Christ’s left go to worldly goods, to his riches, to food and to
drink: it follows that Christ turns his face away from him. © Author
330 A nna N ilsên
There are indeed instructive pictures in our medieval churches, but these
pictures are overly explicit in character and need no intermediary to be
understood (figs. 4-6). Nor do there seem to be any references to paintings in the
sermons handed down to posterity. Very much of this material, of course, has
disappeared, but if references to pictures had had any importance or been
general, it certainly would not have been difficult to find a few examples of such
sermons. Nor do the ecclesiastical sections of the provincial laws of the Middle
Ages mention paintings among the things necessary to a church, and there are,
indeed, churches, which have never had any paintings as well as churches with
mostly or only ornamental decoration. The closest we come to any mention of
paintings in medieval churches at all is a document from 1480 concerning the
diocese of Turku (Âbo) in Finland, saying that the parishes were not allowed to
engage any painter (or other craftsmen) without first asking the bishop and
chapter.' Many of the Swedish parish priests were of rural origin, with just a
rudimentary education.*8 These priests can hardly be believed to have practised
any more advanced pedagogical methods to capture their audience.
Michael Baxandall has, before Belting, touched upon the function of pictures
in churches during the Middle Ages. In his book Painting and Experience in
Fifteenth Century Italy (1972), he devoted a chapter to the function of religious
pictures, where he refers to the thirteenth-century author, Giovanni Balbi,9 who
in his Catholicon gives three reasons for providing churches with pictures:
1. They were for the instruction of simple people, because these are
instructed by the paintings, as if from books;10*
2. They should revive the memory of Christ’s passion and the examples set
by the saints; and
3. They were painted to excite the feeling of devotion.
Finlands medeltidsurkunder, ed. by R. Hausen, 8 vols (Helsingfors: s.n., 1910-35),

vol. IV, 1451-1480 (1924), pp. 525-26, document 3853. It says: . . . ‘Item statuimus,
quod nullus curatorum assumat sibi muratores, vitrifiées, pictores, carpentarios aut
ceteros officiales ad notabile edificium ecclesie spectantes sine speciali nostra vel nostri
capituli licencia, sub pena xl marcharían’.
8 Svenska kyrkans historia, ti: Den señare medeltiden, 1274-1521, ed. by Yngve
Brilioth (Stockholm: Svenska kyrkans diakonistyrelses bokforlag, 1941), p. 726; Bertil
Nilsson, ‘Om rättigheten och skyldigheten att predika’, in Predikohistoriska perspektiv,
ed. Alf Härdelin (Uppsala: Skeab Verbum, 1982), pp. 56-57; and Sven-Erik Pender,
‘Predikan ad populum under svensk medeltid’, in ibid., pp. 81-82.
Michael Baxandall, Painting and Experience in Fifteenth-Century Italy (Oxford:
Oxford University Press, 1972), p. 41.
0This was certainly true for some pictures, e.g.. the moralizing pictures (figs. 4-6).
However a good many pictures were too complicated to be interpreted by people without
learning, or could not be well distinguished because of the distance or lack of light.
Man and picture 331
Baxandall finds it tempting to see the mentioned points as a brief for a painter,"
and he concludes that ‘preacher and picture were both part of the apparatus of
the church, and each took notice of the other’.12 Baxandall has seen a connection
between painting and sermon, and he is right to do so. There is a connection. But
this connection is mutual only to a certain degree. Thus, there is an instructive
factor but just for a part of the picture programme: the one that was painted
where it could be seen, consisting of pictures that could be understood by
everybody. There is certainly also a mnemonic function, but of course not for
pictures that could be neither seen nor understood; and there is finally also the
emotional aspect, but it is applicable only to some of the pictures. Thus all three
arguments are true only to a certain degree. A serious problem is that so many of
the pictures painted on walls and vaults could not be seen or interpreted by the
congregation. Those pictures could not function in any of the three roles
mentioned.
I tend to see the connection between preaching and painting in another way.
In my opinion the church paintings are not sermons in themselves but reflections
of sermons, glimpses of what was being preached in the churches.13 The series of
events represented on the walls often covers the whole of the church year. It
sometimes is possible also to trace certain preferences or tendencies, stressed by
the choice of pictures or by the character of the pictures, reminding one of the
leitmotifs of thematic sermons. In one church you may find the Virgin Mary’s
importance for the Salvation clearly outlined among the paintings. Another
programme may stress the necessity of prayer. A third one will focus on the
Holy Eucharist. Still another one may be moralising, warning against sin and
contrasting good examples with bad ones and so on. There is a great variety
among the picture programmes of the late Middle Ages, much more than during
the Romanesque era or the High Gothic period, when the content of the picture
suites had a more descriptive character and mainly represented biblical
material.14
In Sweden, the change in character of the sermon, in favour of a thematic ap
proach to preaching, takes place in the middle of the fifteenth century and
coincides with the change of composition of the ensembles of mural paintings.15
" Baxandall (as in note 9), p. 41.

Ibid. p. 48
15 This aspect has been studied in Anna Nilsén. ‘Paintings as a Reflection of Sermon’,
in Ars Ecclesiastica. The Church as a Contextfor Visual Arts, ed. by Arja-Leena Paavola
(Helsinki: University of Helsinki. Department of Theology, 1996), pp. 14-33.
1J Nilsén (as in note 1), particularly pp. 459-60; see also pp. 527-58 (Zusammen
fassung).
15On the development of the sermon in Sweden, see Roger Andersson. PostiUor och
predikan, Diss. (Stockholm: Riksantikvarieämbetet. 1993). The above-mentioned change
332 A nna N ilsén
The thematic sermon had developed centuries earlier. But this—often very
elaborate—type of sermon mostly was addressed to a learned audience, though
the same structure could be applied also in sermons to an unlearned audience
with, however, a less sophisticated degree of complication than was used for a
learned audience.
Let us return to the church paintings. As I have already pointed out, their
functions cannot possibly have been the same throughout the Middle Ages, first
of all since the cultural and spiritual climate was not the same all the time. One
well-known change is that there was, during the Gothic period, from the
thirteenth century onwards, a successively growing tendency towards a more
emotional approach in worship. This certainly made the educational aspect of
the pictures less predominant than it may have been earlier, while the third of the
functions, mentioned by Giovanni Balbi, the one of inviting empathy, must have
stood out as more and more important. This is reflected in the choice of motifs
and in the way of placing them. New types of devotional pictures such as Mary
of the Seven Sorrows, Mary of the Rosary, Mary as Mater omnium (fig. 3), the
Man of Sorrows, the Anna Christi, God with his sacrificed Son (fig. 7), and
other motifs—all with an emotive quality—appear in connection with altars, in
porches, and wherever the congregation could have easy access to them. There
had been, of course, devotional pictures earlier, but they were there to function
more exclusively as receivers of prayers; in the late Middle Ages the function of
devotional pictures was also to appeal to the onlooker’s empathy, to arouse
emotion.16
The three functions mentioned by Giovanni Balbi are certainly important, but
they do not constitute the stimuli for providing churches with paintings. And
how would one be able to explain all the pictures that could not be seen by
anybody? First of all, 1 think we ought to believe what the inscriptions tell us
about the intention and function of the paintings: that they were painted in the
honour of God and to embellish the church. If the whole of the ensemble of wall
paintings could not be discerned or interpreted, it could undeniably be appre
ciated for the beauty of its colours, which may have had the effect that abbot
Suger of Saint-Denis rated so highly, an anagogical function, meaning that the
contemplation of beauty could lift the soul from its dull ordinary state and give it
access to the Divine.17 However, paintings, like other adornments, could be
is discussed on pp. 192-93. For the change in the picture programmes, see Nilsén (as in
note 1), pp. 459-60, 527; and Nilsén (as in note 13), pp. 22-24.
"■ On the development of the devotional image, see Sixten Ringbom. Icon to
Narrative. The Rise of the Dramatic Close-Up in Fifteenth-Century Devotional Painting
(Âbo: Âbo akademi, 1965).
Erwin Panofsky, Abbot Suger on the abbey church of St Denis and its art treasures
(Princeton: Princeton University Press, 1946). p. 21.
Man ami picture 333
Fig. 7. Vende! church, Uppland, Sweden, painring in the porch, c. 1452. 'The
Trinity’. © Author
334 A n n a N ilsén
Fig. 8. Tensta church, Uppland, Sweden. End thirteenth century. ‘Exterior from
south ’. This was no ordinary parish church but probably built by some among
the foremost families o f thirteenth-century Sweden. © Author
Fig. 9. Tensta church. Paintings from 1437-38. The painter was Johannes
Rosenrod. ©Author
Man and picture 335
Fig. 10. Tensta church. ‘Beugt Jönsson, the donor o f the paintings, at the feet of
St Peter’. © Author
336 A nna N ilsén
given to the church for many reasons, very often for personal ones. In addition
to the above-mentioned decorative function, which we should accept as the
primary one, applicable to the whole ensemble of paintings, we may assume
underlying functions, which are not always easily discovered. There are, as I see
it, several possible types of such function, all of them referring to people
involved: to commissioners, priest and congregation, and sometimes also to the
painter. The officially documented generosity of the person who paid for the
execution of paintings and sometimes also commissioned them could be useful
in a personal career. It could add to a person’s reputation or mend it. The votive
aspect was probably very important. Many precious objects were given to
churches as gifts to God to thank him for a success or in the hope of success.
This certainly also applied to paintings.IS
Let us study a case where we know a little more than usual of the paintings
and the persons responsible for them and where more than one possible function
may be assumed, though this is not evident at first sight. The paintings of Tensta
church (fig. 8 and 9) north of Uppsala in middle Sweden constitute a good
example.1819 We know the year the choir paintings were executed, which was
1437.20 We know who paid for the paintings. The donor was one of the foremost
men in Sweden of that time, Bengt Jönsson of the Oxenstierna family (fig. 10).21
We also know the name of the painter, Johannes Rosenrod probably a German.
For once we even know a little of the parish priest, whose name was Olof,
documented as a member of the Corpus Christi guild of Stockholm.22 First, a few
words about the paintings and the picture programme. The figures and motifs are
comparably large and distributed sparsely all over the interior, kept together in a
tapestry-like unity by a vegetal ornamentation. Both the outline of the figures
and the application of the colours are mostly of high quality, sometimes
exquisite (fig. 11). Johannes Rosenrod was no doubt a skilled painter. The
programme is thematic. Besides an evident reverence for the Virgin Mary, there
are two clearly discernible main themes: one stressing the Eucharist and one
focusing on the Swedish saint, Birgitta (c. 1303-73).
18Nilsén (as in note 1), pp. 507, 514, 515. 529-30 (Zusammenfassung).
19 The programme of Tensta church has been studied in Anna Nilsén, ‘Johannes
Rosenrod—ein deutscher Maler in Schweden’, in Austausch und Verbindungen in der
Kunstgeschichte des Ostseeraums. Homburger Gespräche 9: 22-25.Ì 1.1987, ed. by Lars
Olof Larsson and Jan von Bonsdorff (Kiel: Martin-Carl-Adolf-Böcklcr-Stiftung, 1988),
pp. 43-60.
The nave was probably painted the following year.
■' See Hans Gillingstam, Àtterna Oxenstierna och Vasa under medelliden (Diss.)
(Stockholm: Hæggstrôm. 1952). pp. 259-332.
” ‘Handlingar rörandc Helga Lekamcns gille i Stockholm i: Gillesbokcn 1393-1487’,
in Kungl. Bibliotekets handlingar. Bilagor. 2: 1-3, ed. by Isak Collijn (Stockholm:
Kungl. Bibliotek. 1921), pp. 48, 169.
Man and picture 337
Fig. II. Tensta church. 'The Annunciation'. ©Author

338 A nna N ilsén
The programme is well composed from a theological point of view and

probably worked out by the parish priest Olof, who, in this church, under the
patronage of Bengt Jönsson, was assuredly no ordinary, unlearned parish priest
of rural origin but was an educated man. Many second sons among the nobility
chose a career within the Church. The stress on the Eucharist in this programme
may reflect Olof s own type of religiosity, indicated by his membership of the
Corpus Christi guild. On the northern side of the nave, well lit, was (before the
opening of the wall for a window in the place) a picture of the Man of Sorrows,
surrounded by the Anna Christi, a devotional picture one would expect to find in
connection with the Corpus Christi cult and certainly painted to serve those who
wanted to exert their devotion and pray in front of it.
To judge from the content of the programme, it is also obvious that Bengt
Jönsson himself influenced it, which one would expect, considering not only that
he had spent a large sum of money on the project but also that he was an
educated man and still striving towards the peak of his career. In 1437—the year
of the choir paintings—he was already a member of the government council.
From 1439 he was to reside over the provincial court of Uppland,2324and in 1448
he was, together with his brother, for a short period governor of Sweden. During
the same period his son Jöns Bengtsson made a rapid career within the Church.
Jöns returned in 1437 or 1438 from Leipzig,31 where he had studied at the
university from 1434, and was almost immediately appointed dean of Uppsala
Cathedral.25*In 1448, finally, he was appointed archbishop of Sweden, the same
year as his father was ruling over Sweden.
For a person with Bengt Jönsson’s position and ambitions, we may suspect
the paintings to have a propagandist function, and it is in such a connection that
the remarkable series of pictures in Tensta, representing the most important
events of Birgitta’s life and deeds, should be considered. The series even
contains a defence of her revelations, which at the time were under critical
debate at a Church council in Basel, where it was maintained with some
insistence that the visions of Birgitta were more likely to be inspired by the devil
than by Christ and Mary (fig. 12). Birgitta was to be one of the most renowned
saints of the Catholic Church. The importance of her person and of her revela
tions hardly can be exaggerated. In 1391 she had been canonized, and only a few
years later she was mentioned among the Swedish national saints.
To call in an important painter and adorn a large church at one’s own expense
was doubtless regarded as an honourable deed and thus was certainly
The Swedish term for this office was lagman, provincial judge.
24 Gillingstam (as in note 21), p. 337 (s.a. 1438); Sven Tunberg, ‘Jöns Bengtsson
"Oxenstiema” in Nordisk Familjebok (Stockholm: Nordisk familjeboks förlags
aktiebolag. 1914), col. 1149:.?.a. 1437.
:s Gillingstam (as in note 21). p. 337.
Man and picture 339
Fig. 12. Tensta church. ‘Revelations o f St Birgitta We here see a most effective
defence of the Revelations o f Saint Birgitta. The devil tries to instil in Birgitta a
false revelation, but an Angel comes to her rescue and crushes his skull with a
mace. How Christ and Mary inspire Birgitta's Revelations is shown in the
following picture o f the Birgitta series. © Author
propaganda in itself, but in the case of the Tensta paintings and particularly the
Birgitta series the propagandizing function is, as we shall see, more explicit than
this. From 1437 Birgitta was adopted among the patrons of Tensta church,26 and
this was certainly no coincidence. Bengt Jönsson was distantly related to
Birgitta,2 who was already in his time a national symbol. The connection with
Birgitta indeed must have been something to manifest and propagate for a man
with great ambitions, for both himself and his son. The depicted reference to the
attack on her revelations, fresh in 1437, and the defence of them among the
paintings, has no known precedents. This applies also to the rest of the pictures
within the series. It thus may be concluded that it was specially composed for
Tensta church, and there is little doubt that it was personally ordered by Bengt
A medieval church bell, still in the church, mentions Birgitta among the patrons of
the church; See Nils Sundquist, ‘Tensta kyrka’, in Upplands kyrkor, 42 (Strängnäs:
Ärkestiftets Stiftsrâd, 1970), p. 37.
Nilsén (as in note 19), p. 46. See also Sigurd Rahmqvist, Sätesgärd och gods (Diss.)
(Stockholm: Historika institutionem Stockholms universitet, 1996), pp. 242-43.
340 A nna N ilsén
Jönsson to establish and proclaim his own and his family’s connection with the
national saint.
It is very seldom that the main themes of a picture programme are so easily
explained with reference to identifiable contemporaries as in Tensta. We may
conclude from the paintings of Tensta church that personal ambition sometimes
may have resulted in the most beautiful pictures and that medieval wall paintings
may have had other than instructive, devotional, or mnemonic functions—in
other words, that there may have been functions of a less obvious character with
reference to other parties than the congregation, such as the donor or the parish
priest.
Representations o f Jews in
Danish medieval art— can images
be used as source material on their own?
U lla H a a s t r u p '
his article deals with representations of Jews in medieval art in Denmark.
T I have selected a few examples from the wall paintings and altarpieces
from the extensive material available. The images will show us how Jews
were characterized as good, neutral, or evil in the period between the late 1080s
and the Reformation in 1536. Jews never were depicted with photographic
accuracy or through abstract symbolism. Their clothing, in particular their hats,
always followed the norms of the medieval Jews themselves or, in particular, the
norms demanded by Christian society. Jews had to differ from other citizens and
must not be mistaken for Christians.
The images derive mainly from the art of the churches; they were com
missioned by the Church and/or by secular donors. Since Jews were depicted in
an attire similar to that of the contemporary European Jew, one must presume
that the didactic and momentous pictorial message of the Church was understood
by the spectator; otherwise the images would carry no meaning. If there were no
Jews in Danish society, how would the congregations and the clergy immedi
ately understand that a given figure was a Jew, depicted in the midst of an often
densely populated scene? Would it be merely because of his hat, or because his
face was in profile, denoting evil? Did all knowledge of Jews, their aspect, and
dress derive from travels or from pictorial models? After presenting the various
kinds of images, I shall discuss whether minority groups of Jews may have lived
Translated into English by Elisabeth Kofod-Hansen.

342 U lla H aastrup
Fig. 1. Jelling church, fresco, J080-1100, 'St John the Baptist preaching to a
gathering of Jews’, Watercolour by the restorer J. Magnus-Petersen 1874.
© Nationalmuseet, Copenhagen
Representations o f Jews in Danish medieval art 343
within the Danish realm during the Middle Ages. Information on this matter is,
however, almost completely lacking in the few existing Danish written sources.
In Jelling church in Jutland, some extremely well preserved frescoes showing
the Life of John the Baptist and the Baptism of Christ were discovered in 1874—
75. Unfortunately, they were destroyed by humidity and were re-painted.
However, the very carefully-executed water-colours by the restorer J. Magnus-
Petersen show us, among other scenes. John the Baptist preaching to a gathering
of Jews (fig. 1). The text scroll reads ‘. . . vox clamantes. . .’. The Jews are
standing in rows, wearing brimmed hats with alternately round or pointed
crowns. The position of the faces is neutral. From stylistic criteria, I would date
the frescoes of the Jelling workshop to around 1080-1100.' The same
characteristic hats are found in Reichenau, in the frescoes in the apse of the
church of St Peter and Paul. The row of prophets in the lower parts of these
paintings wear similar types of hat. Dendrochronological investigations have
dated the frescoes to around 1104-157 The early Romanesque stained-glass
windows in the nave of the cathedral in Augsburg, Bavaria, show us, among
other motifs, the prophet Daniel with a similar Jewish hat.
The frescoes from around 1125 in Râsted church in Jutland include the oldest
extant depiction of an evil Jew in Denmark (fig. 2). He is shown in profile with a
hooked nose, a short beard, and a Jew’s hat. He is part of a painting showing,
among other things, the Magi, who during their journey pay a visit to King
Herod and his court in order to enquire about the newborn child. This picture and
several other motifs in the extremely carefully prepared pictorial programme are
directly in keeping with the liturgical plays about the Three Magi known abroad.
At King Herod’s court, the sage Judex Barbatus is asked what the scriptures tell
about the king who is to come. In these particular Danish Romanesque frescoes,
he stands out singularly in profile. All other figures in the church are painted in a
three-quarter profile, and it follows that this is a marked characterisation of him
as an evil Jew.'*2
Poul N0rlund and Egmont Lind. Danmarks romanske kalkmalerier (Copenhagen:

Selskabet til Udgivelse af Danske Mindesmærker, 1944), pp. 50-63, cat. 70; and Ulla
Haastrup. 'Johannes i Jelling', Skalk 2 (1972) pp. 18-28. See also the following items in
Danske kalkmalerier. Romansk tid ¡080-1175. ed. by Ulla Haastrup and Robert Egevang
(Copenhagen: Nationalmuseet. 1986): Ulla Haastrup, Introduction, pp. 24-26 and 39;
Knud J. Krogh, ‘En meget sorgelig historie', pp. 64-65; Ulla Haastrup, ‘Johannes
Dpberens historie', pp. 66-69; Kirsten Trampedach. 'Garnie billeder pâ nye vægge'. pp.
70-71; and Ulla Haastrup. ‘Jellingværkstedets forbilleder’. pp. 78-79.
2 Wolfgang Erdmann. Die Reichenau im Bodensee. Geschichte und Kunst. 10th rev.
edn. (Königstein im Taunus: Die blauen Bücher. 1993), pp. 22-25.
Ulla Haastrup, 'Die romanischen Wandmalereien in Râsted. Ikonographie,
Bildprogram und Theater', Hafnia (1972). pp. 69-138; Lise Gotfredsen. Râsted Kirke—
Spil og Billede (Copenhagen: Akademisk Forlag, 1975); Ulla Haastrup, ‘Liturgisk,
344 Ul.I.a H aastrup
Fig. 2. Raasted church, fresco, about 1125, ‘judex barbatus in the scene of the
Magi at the court o f King Herod. © Author
Around 1275, the chancel vault in Str0by church on the principal Danish
island of Sjælland was decorated with paintings. Prophets are depicted on the
pendentives. One of them has obvious Jewish facial traits, but he is shown in a
neutral three-quarter profile. In the paintings of the Passion of Christ at Strpby,
the motifs are not framed but, rather, are intertwined with plant stalks and leaves.
Here one finds in contrast to the Jewish prophet the oldest Danish examples of
cruel Jewish executioners.*4 Their dress includes yellow coats or coats with broad
stripes. On their Jews’ hats there are stalks, their heads are mainly shown in
profile with open mouths and tongues protruding, and they have hooked noses.
They beat and torture Christ. Simon of Cyrene is pushing Him and does not
assist in carrying the cross. The Jews in these pictures show strong hatred, as is
the case in the rest of Europe at the end of the thirteenth century. A slightly later
parallel is found in the large illustrated Speculum húmame salvationis from
kirkeligt teater og Râsteds kalkmalerier'. in Danske kalkmalerier. Romansk lid 1080-

1175 (as in note 1). pp. 84-87; Ruth Mellinkoff, Outcasts: Signs of Otherness in Northern
European Art in the Late Middle Ages (Berkeley: University of California Press, 1993).
4Kirsten Trampedach, ‘Nye iagttagelscr, Stroby'. in Danske kalkmalerier. Senromansk
lid 1175-1275. ed. by Ulla Haastrup and Robert Egevang (Copenhagen: Nationalmuseet.
1987). pp. 184-85: and Ulla Haastrup. ‘Representations of Jews in Danish Medieval Art",
in Danish Jewish Art: Jews in Danish Art. ed. Mirjam Gelfer-Jprgensen (Copenhagen:
Rhodos. 1999), pp. 153-54.
Fig. 3. Altarpiece from Boeslimde church, now at the National Museum in

Copenhagen, about 1425-35, Lübeck. 'The Virgin Mary's Training in Reading’.
© Nationalmuseet, Copenhagen
346 U lla H aastrup
Kremsmünster of approximately 1325/ The executioners holding Christ before

Pilate and the three tormentors in the scene of the Flagellation are also caricature
Jews, in profile with open mouths and protruding tongues.
In 1980, wall paintings from the second quarter of the fifteenth century were
discovered in Brönnestad church in Skâne, which belonged to medieval
Denmark and now is part of Sweden." From an artistic point of view, they rank
among the best of late medieval Danish frescoes. They include several depictions
of good as well as bad Jews. In the representation of the Infancy of Christ there
are two motifs on the pendentives showing the Women with the infant Christ and
Simeon and his assistant, in the Presentation in the Temple. Simeon is wearing
the mandatory Jewish synagogue garb, the assistant wears a Jew’s hat. In the
depiction of the Miracle of the Cornfield during the flight into Egypt, the farmer
wears a tall pointed hat—a peasant’s hat or a Jew’s hat? Pilate is wearing a
similar hat in the depiction of Christ before Pilate. Herod wears a crown and
Caiaphas a bishop’s mitre. Pilate is often represented with a Jew’s hat and not as
a Roman governor. Brönnestad displays a rare motif where Judas receives the
thirty silver coins from the high priests. Two Jews are shown, the high priest
handing over the money and Judas opening his moneybag to receive them. Both
are painted in evil profile, with Jewish traits. Some of the pendentives also
display caricature-masks in profile. The one shown here must represent a spitting
Jew if compared with Judas’s profile.7 However, even worse pictures of Jews
occur in the main scenes of The Passion of Christ. The Flagellation is conducted
by three tormentors, two of whom are shown as Jews with strongly caricatured
traits. They are painted not in profile but with open mouths showing their teeth,
which was considered highly indecent during the Middle Ages. Sadism and
hatred radiate from their ugly faces.
The representations of Jews vary widely in character even in one pictorial
programme. One cannot help but wonder at the vast shift in attitude from the
depictions of the old Simeon to those of the tormentors of Christ.
Pictures of Jews are not found solely among the Danish wall paintings: they5
5Speculum humanae salvationis. Codex cremifanensis 243. Kremsmünster. Faksimili-

Ausgabe. ed. by Willibrod Neumüller (Graz: s.n.. 1972).
‘ Torkel Eriksson, ‘Nyframtagna medeltidsmälningar i Brönnestad', Ale: Historisk
tidskrift for Skáneland 2 (1982), pp. 1-19: ‘To usædvanlige motiver', ibid.; ‘Bebudelsen
og Jomfru Maria med bamet’. in Danske kalkmalerier. Gotik 1375-1475, ed. by Ulla
Haastrup and Robert Egevang (Copenhagen: Nationalmuseet, 1985), article 29. 30.
Ulla Haastrup, ‘Groteske Masken en face und im Profil. Die Bedeutung des
Gesichtsstellung in der mittelalterlichen Symbolsprache', in Mein ganzen Körper ist
Gesicht—Groteske Darstellungen in der europäischer Kunst und Literatur des
Mittelalters, ed. by Katrin Kroll and Hugo Steger (Freiburg im Breisgau: Rombach,
1994). pp. 313-24: Haastrup. ‘Representations of Jews in Danish Medieval Art’ (as in
note 4), pp. 132-37, et passim.
are also found on the winged, sculpted wooden altarpieces that furnished Danish
churches in great numbers during the later Middle Ages. One of the major
examples, the great winged altarpiece from Boeslunde church on Sjaelland, can
be seen at the National Museum in Copenhagen.* It is an imported work of art
from Lübeck, dated about 1425-35. Following re-conservation in 1980, it is now
to be seen with its original paint both on the sculptures and in the paintings. This
altarpiece relates to the stories of Joachim and Anna and those of the Virgin
Mary's visit to the Temple and her training in reading there (fig. 3). The teacher
on the throne-like chair wears a Jew’s hat. The paintings on the wings include
the Passion where The Kiss of Judas shows a red-haired Jewish Judas in a yellow
coat and Malchus, who is also depicted in the evil Jewish profile. The paintings
on the outer wings contain detailed representations of Jews with different kinds
of Jew’s hat and the fur-lined garments of the wealthy.
Between 1460 to 1480, a large painting workshop was active in Sjaelland. It
worked with speed and therefore presumably made cheap wall paintings,
covering the interiors in their entirety from chancel to tower bay in about thirty
village churches in the northern part of Sjaelland." It worked from models, which
we so far unfortunately have not been able to identify among preserved
woodcuts. In this anonymous workshop, the ‘Isefjords workshop’, there was
great consistency in the symbolism of the facial pose.'” Three-quarter profiles
alone are seen in depictions of ordinary people—and of Christ in epic scenes.
Some people are shown in the evil profile: these are almost exclusively the cruel
Jewish tormentors or. for example, Judas.
In Nprre Herlev church, the Jews flagellate Christ savagely. They are shown
in profile with open mouths and hooked noses, and one has his chest bare in
further mockery of Christ. In Mprkpv church one finds a painting of a similar
flagellation, but in this case the executioner has loosened his hoses to mock with
his bare behind. A similar detail is found in The Crowning with Thorns in
Mprkpv, and in the same motif in N0rre Herlev the evil sneering profiles are the
predominant feature. In Jörg Preus’s paintings from 1502 in Stiftung Melk in
* Francis Beckett, Altertavler i Danmark fra den senere Middelalder (Copenhagen:

Ministeriel for Kirke- og Undervisningsvæsenet, 1895): Ulla Haastrup. Mariatavlen fra
Boeslunde (Copenhagen: Nationalmuseet. 1984).
9 See the following entries in Danske kalkmalerier. Gotik 1375-1475 (as in note 6):
Niels-Knud Liebgott. ‘S. Laurentius' legende’, and ‘S. Katharinas legende': Bodil Franck,
'Moraliteter’; Dorthe Falcon Möller. 'Hyrdemes sækkepiber'; Ebbe Nyborg,
‘Dommedag’; Ulla Haastrup, 'Resiaureringsprincipper'. 'Det bibelske verdensforlpb'. and
'0delagte kalkmalerier og restaureringshistorie’. See also Niels M. Saxtorp, Danmarks
kalkmalerier, 2nd edn. (Copenhagen: Politikcns forlag. 1997), pp. 22-23 et passim; and
Jorgen Aabenhus, En kalkmalers bibelhistorie (Copenhagen: Religionspædagogisk forlag.
1999).
Haastrup, 'Representations of Jews in Danish Medieval Art' (as in note 4), p. 132.
348 U lla H aastrui*
Fig. 4. Altarpiece in St Mary in Helsingborg, 1449-52. ‘The Circumcision'.

© Author
Fig. 5. Tuse church, wall painting, 1460-80. ‘The Circumcision ’.

© www.kalkmalerier.dk
Fig. 6. Fanefjord church, wall painting, about 1500. ‘Jesus among the Doctors’.
© Author
350 U lla H aastrup
Austria, we see the same type of mockery. Jews, however, were represented in a
more neutral manner, as well-dressed wealthy men with a variety of types of
Jewish hats. For instance, there is less drama in the rendering of the Expulsion of
the Moneychangers from the Temple in Mprkpv church.
There were often many different types of Jewish hats depicted within one
decoration or altarpiece. In the pictorial arts of the Middle Ages, each figure in
the depictions of episodes from the Bible had to be painted in contemporary
dress, so as to ensure that it could be easily identified. For that reason, even the
dress of Jews changed and varied. The central panel of the winged altarpiece in
Gudme church is taken up by a densely populated crucifixion scene. The group
of Jews is shown with several different kinds of hats and beards. In the carved
relief with secondary paint from the altarpiece in Brendekilde church, Funen,
from about 1475, there is for instance a considerable number of Jewish hats in
the scene with the Crowning of Thorns. Several Jews are wearing hoods or other
pointed hats, with large upturned brims.
One particular motif is of special interest in this context; namely, the Circum
cision and Naming of Christ on the eighth day as described in the Bible. In the
Middle Ages, making a picture of the event as it had once occurred posed a
problem. Should it be depicted as an analogy to the common practice of
baptizing children inside the church, as an operation performed by a high priest
looking somewhat similar to a bishop, or should it be represented in the manner
in which the Jewish ritual of circumcision was carried out in the synagogue?
On the winged altarpiece in the church of Mary in Helsingborg, which is
dated to 1449-52 on the basis of the coat of arms of the donor, the feudal vassal
Joachim Flemming, the Circumcision is shown in relief on the central panel (fig.
4)." The small naked child is placed on a prominently placed altar. The infant
Jesus is held by two women, while a bearded man and his assistant carry out the
operation. Thus it is performed by two Jewish men with the holy women as
participants. The altar receives the blood of Christ for the first time—a highly
significant theological detail. The Circumcision by the Isefjords workshop in
Tuse church, which is probably about ten years later, also depicts the child laid
on a table. The Virgin Mary is holding the infant Jesus; behind her stand two
women. The high priest stands on the opposite side of the altar with a knife in his
hand. He is wearing a bishop's mitre, but his garb is nevertheless secular, with
fur-trimmings on the borders of the knee-length coat. He carries a purse or bag in
his belt (fig. 5).
Fanefjord church on M0n has an exceptionally large and detailed pictorial
programme painted by the so-called Elmelunde workshop around 1500. Several
of the motifs are copied directly from the forty-page edition of the Biblia*186
" Lena Liepe, Den medeltida träskulpturen i Shane. Produktion och forvärv. pp. 154,
186, Den medeltida träskulpturen i Skäne. En bilddokumentation, fig. 153, Skänsk
senmedeltid och renässans, vols 14-15 (Lund: Lund University Press, 1995).
Representaiions o f Jews in Danish medieval art 351
Fig. 7. Haggadah. German, c. 1460, London, British Library, MS. Add. 14762,
fol. 7. 'German Jews’ (from Metzger: Jewish Life , fig. 175)
Pauperum, as for instance David and Goliath and Samson slaying the Lion.11 In
the depiction of Christ, aged twelve, among the doctors in the temple there is no
doubt as to who are the Jews. The two Pharisees wear tall hats with round
crowns; the evil one shown in profile has a small pointed top on his hat, and the
other one carrying a scroll wears a plushy grey hat. Similar plushy tall hats are
worn by other evidently Jewish figures in Fanefjord, and such hats are found also
i: Ulla Haastrup, 'Pictures in Both Books and Churches. Biblia pauperum and
Speculum humanae salvationis as reflected in Danish ecclesiastical art’, in Living Words
and Luminous Pictures. Medieval Book Culture in Denmark, ed. by Erik Petersen
(Copenhagen: Det Kongelige Bibliotek. 1999), pp. 127—44; Haastrup. 'Representations of
Jews in Danish Medieval Art' (as in note 4). p. 132.
352 Ulla H aastrup
in a German Haggadah manuscript of 1460 (figs 6 and 7)." The Circumcision is

painted in Fanetjord church as one of the scenes in the Childhood of Christ
(fig. 8). A man wearing a red coat with a fur collar and brim is sitting on a turned
wooden chair with the small naked child in his lap, Christ with the cruciform
halo. The high priest is wearing not a bishop’s mitre but a plushy Jew’s hat. A
kneeling young man with no headgear is performing the circumcision. Two
women with haloes stand next to them. From an iconographical point of view,
this is an unusual motif. The all-important altar is missing; instead, the Jewish
high priest or godfather is sitting on a chair with the infant Jesus. In the forty-
page edition of the Biblia Pauperum, the scene is missing. In the slightly more
recent fifty-page edition, the Circumcision is depicted with a bishop at the altar
where Christ is laid, exactly like the depiction in Tuse church."
Where did the painters on Mpn find their model? Turning to illustrated
Hebrew manuscripts one finds close parallels (fig. 9). In a Hebrew manuscript
from Naples, dating to the third quarter of the fifteenth century, there is an
illustration containing the same elements as the mural in Fanefjord church.
Godfather Sandaq or the high priest is sitting in a chair with the naked child in
his lap, and the Jewish officiant, the Mohel. is kneeling in front of them while
performing the operation.I! On the altarpiece in Nürnberg, the Tuscher altar from
the mid-fifteenth century, the Circumcision is painted almost as a 'photographic’
rendering of the Jewish ritual (fig. 1 0 ) .The Virgin Mary and Joseph are seen at
the furthermost back of the picture, looking in over a curtain.
Depictions with such realistic details cannot be found in Danish murals: they
belong to quite another genre with other sets of rules for the symbolic rendering
and with no use of spatial effects. When seeking depictions of Jews, one must
make up one’s mind as to what kind of pictures one should expect to find. Now
we are approaching the core of my hypothesis. How could one depict the naming
or baptism of Christ in a large Danish parish church as if it took place in a Jewish
synagogue? In Nürnberg there were a lot of Jews, and one would understand the
‘caricature’ painting of the Circumcision; but what about in Denmark?
I have pointed out earlier the manner in which all biblical events were
rendered in contemporary garb and with everyday objects. They should be quite*145
Térèse and Mendel Metzger, Jewish Life in the Middle Ages. Illuminated Hebrew
Manuscripts of the Thirteenth to the Sixteenth Centuries (Secaucus. NJ: Chartwell. 1982).
p. 123, fig. 175.
14 Knud Banning, 'Jesu omskærelse. En gennemgang af billedsiden i den 50-bladede
Biblia Pauperum', in Genesis Profeta. Studier i gammeltestamentlig ikonograß, 6.
Nordiska symposiet fór ikonografiska studier. Gimo 1978. ed. by Patrik Reuterswärd
(Stockholm : Almqvist& Wiksell. 1980). pp. 135-39.
15Térèse and Mendel Metzger (as in note 13), p. 225, fig. 333.
Mellinkoff (as in note 3). voi. II, fig. 11.23.
Fig. 8. Fanefjord church, wall painting, c. 1500. 'The Circumcision’. ©Author
Fig. 9. Hebrew manuscript, Naples, Third quarter o f the fifteenth century.

Nîmes, Bibliothèque municipale, ms. 13, fol. 18F. ‘The Circumcision’ (from
Metzger: Jewish Life./?#. 333)
354 U lla H aastrup
comprehensible to the spectator. Nobody must be at a loss as to the meaning of

the scenes or have any doubts about the identity of the participants. The clergy
who approved and probably also determined the pictorial programs in the big
rich church of Fanefjord must have had an immediate understanding of the
depiction of the Jewish ritual of circumcision with the small Jesus-child with his
cruciform halo, and so must their congregations. Could they fathom it, if they
themselves were ignorant of contemporary Jews, their clothing and their reli
gious customs? Would one for instance have to sail the short distance from
Gedser to Rostock to see Jews?'7 Or might one in that period around 1500, so
abundant in herring during the summer, have met Jews on Mpn, Falster, and
Lolland among the other foreign tradesmen from Germany, Norway, and the
Baltic countries? If we seek such minorities in the Danish realm, we must expect
to find them close to the king and close to centres of international commerce.
In Stege on Mpn, where the king had a castle, the church was given a solid
extension and was equipped with vaults in 1494 (fig. 11).**IS A series of names is
painted on the western vault of the main aisle, and \ . . vi mormester alle
evindelich meth Gudh’ [. .. we the master-masons are all for ever with GodJ. The
names of the guardians of the church are painted elsewhere, and on the northern
wall of the second pier one finds some emblems and the designation ‘Kristen lude’
(Christian Jew) next to a depiction of a man and a woman, each carrying a beaker
and holding a jug between them. The man is wearing a pointed hood, and the
caricature-masks on the pendenti ves are wearing extreme Jew’s hats. Could
Kristen lude perhaps be a recently baptized Jewish artisan who took part in the
building activities? In another significant Danish town, Nyborg on Funen, an
account of about 1537 mentions the taxation with eleven marks of the booths in
Jpdestræde (Jews Lane).''’ The royal court at Nyborg Castle and the extensive sea
trade from Nyborg counted among the most vital preconditions for a Jewish
community to live here in the same manner as in other towns of Europe.
If historians read the old Danish written sources anew, then possibly several
of the names with the surname spelled iude / jyde could be transformed from
17Meklenburgisches Urkundenbuch, vols 3-4 (Schwerin: Verein für meklenburgische

Geschichte und Altertumskunde, 1865). For example, in 1279, the city of Rostock donates
a burial-ground for Jews outside the town walls. In 1283, the Jew Salathiel’s house and
his children are mentioned. The following year, the city council receives a loan of four
hundred marks from the same Jew and grants him the city's protection. 1 am indebted to
Associate Professor Kurt Villads Jensen for drawing my attention to the publication of the
Mecklenburg diplomas.
IS Haastrup, ‘Representations of Jews in Danish Medieval Art’ (as in note 4), p. 161,
notes 132-33.
Lars Bisgaard, ‘Gildernes religiose rolle i det middelalderlige Svendborg—om en
genfunden liste over altre i Set. Nicolai og Vor Frue'. in Ârbogfor Svendborg og Omegns
Museum (Svendborg: Museumsforeningen Svendborg. 1996). pp. 42-63.
Representations o f Jews in Danish medieval an 355
Fig. 10. Master of The Tuscher-altar, Nürnberg, c. 1450, ‘The Circumcision’.

Aachen, Suermondt-Museum, from Mellinkoff, Outcasts, voi. ii, fig. ii.23
Jutlander to the minority, Jews.™ The manner of spelling was the same until the
early sixteenth century, only then i0de [‘Jew’] was spelled with the o with a slash
like the modern 0. When reading the religious medieval texts in the Danish
language, there is never any doubt that the word iude designates Jews, even
when spelled like ‘Jutlander’. Medieval names as source material could possibly
™Haastrup. 'Representations of Jews in Danish Medieval Art' (as in note 4), p. 160.
note 130.
356 U lla H aastrup
Fig. 11. Stege church, wall painting 1494. ‘Kristen iude and figures with Jewish
hats’. ©Author
be exploited further. Old Testament names in Europe were usually names for
Jews. Yet, accounts of Jewish history in Denmark always have as their point of
departure that King Christian IV in 1622 summoned Portuguese Jewish gold
smiths to Altona and granted them freedom of religious practice.
After years of preoccupation with images of Jews in the Danish Middle Ages,
I have found so many significant ones that I am no longer apt to believe the
traditional statement of the Danish and Scandinavian historian that ‘there were
no Jews in Medieval Scandinavia’. I have been surprised by the many interesting
depictions of Jews which neither I nor others had recognized before. Based on
the pictorial evidence, historians must re-read the few written sources. Perhaps
then accounts about Jews in Denmark may surface.
Why should there not, indeed, have been Jews in the Nordic countries'? On all
other levels, Denmark was an integrated part of the Catholic European society.
Where are the arguments that permit one, based on a scanty and random written
source-material, to claim that Denmark did not house a Jewish minority?
Anti-semitism, image desecration,
and the problem o f ‘Jewish execution’
N orbert S chnitzler
his essay deals with a sample of late medieval panel-paintings and
T woodcuts depicting the narrative of a miracle that allegedly took place in

fourteenth-century Flanders, in the county of Hainaut. These pictures will
be discussed from a historian’s point of view, that is to say, they will be used as
sources providing the basis for an investigation of medieval anti-Semitism, and
especially of the impact of medieval legal theory on visual strategies of
stigmatization and discrimination. From a quick glance at current debates on
issues such as the modes of representation of reality or the relation between
content and context, inaugurated by art historians and social historians as well, it
seems clear that those debates which gave rise to a re-evaluation of pictures as
historical sources require new methodological and systematic considerations,
especially in the field of legal history.' More precisely, my argument
1 N. Schnitzler, ‘Picturing the Law—Introductory Remarks on the Medieval

Iconography of Judgement and Punishment', in Law and its Images in Medieval Europe
(special issue of Medieval History Journal. 3.1). ed. by N. Schnitzler (London: Sage,
2000), pp. 1-15: Der Blick auf die Bilder: Kunstgeschichte und Geschichte im Gespräch,
ed. by Otto Gerhard Oexle (Göttingen: Vandenhoeck & Ruprecht. 1997); G. Signori.
‘Wörter. Sachen und Bilder—oder: die Mehrdeutigkeit des scheinbar Eindeutigen', in
Mundus in imagine, ed. by A. Löther et alii (Munich: Fink, 1996). pp. 11-33: H.
Talkenberger, 'Von der Illustration zur Interpretation. Methodische Überlegungen zur
Historischen Bildkunde', in Zeitschrift für historische Forschung, 21 (Berlin: Duncker &
Humblot. 1994). pp. 289-313: G. Kocher. Zeichen und Symbole des Rechts (Munich:
Beck. 1992): W. Schild. ‘Nutzen und Wert von Rechtsarchäologie und Rechts
ikonographie für die mittelalterliche Rechtsgeschichte', in Rechtsarchäologie und
358 N orbert Schnitzler
Fig. I. Engraving representing the panel-painting with the Miracle o f Notre-

Dame de Cambrón, c. 1890, J. van Péteghem, Brussels
concerns the ongoing differentiation of the system of law in the later Middle
Ages (i.e., the reception of Roman law, the generalization of legal norms, the
regulations of proceeding) and its implication for the changing contents of
images. This is to say the impact of penal law or the impact of juridical concepts
such as ‘deterrence’ (or so-called juridical ‘rituals of exclusion’) on Christian
iconography, especially on the representation of Jews. To a certain extent one
Rechtsikonographie. Eine Annäherung. Akten des Brüsseler Kolloquiums. Iuris Scripta

Historica, 5, ed. by P. de Win (Brussels: Wetenschappelijk comité voor rechts-
geschiedenis koninklijke academie voor wetenschappen, letteren en schone kunsten van
België. 1990), pp. 59-74: N. H. Ott. 'Der Körper als konkrete Hülle des Abstrakten. Zur
Wandel der Rechtsgebärde im Spätmittelalter’, in Gepeinigt, begehrt, vergessen—
Symbolik und Sozialbezug des Körpers im späten Mittelalter und in der Frühen Neuzeit.
ed. by N. Schnitzler and K. Schreiner (Munich: Fink, 1992), pp. 223-41; idem, Rechts
praxis und Heilsgeschehen. Zu Überlieferung, Ikonographie und Gebrauchssituation des
deutschen 'Belial'. MTU. 80 (Tübingen: Niemeyer. 1983): Text-Bild-Interpretationen.
Untersuchungen zu den Bilderhandschriften des Sachsenspiegels. MMS. 55.1. ed. by R.
Schmidt-Wiegand (Munich: Fink, 1986).
Antisemitism, image desecration, and the problem o f ‘Jewish execution ' 359
could argue that reality occupies the sphere of pictorial tradition.

Our starting point will be a painting situated today in a small chapel about
five miles away from the city of Mons (fig. 1). The chapel is dedicated to Our
Lady of Cambrón, which was the site of a former Cistercian convent another five
miles to the north of Mons. The painting has been dated to the first half of the
sixteenth century. Thus, most probably, it was part of the original inventory of
the chapel, which was consecrated in 1483. The panel is sub-divided into twelve
scenes illustrating the legend of an image-desecration fashioned as a late
medieval curing-miracle.2If we confine ourselves only to the first frame, we may
swiftly assume that it provides the pivotal scene (fig. 2). The panel shows the
violation of the aforesaid cultic image of the Madonna and its moral and social
impacts. The offender has to do penance for it, which means by his life. Just
behind him a man, a carver we are told, already has brandished his axe to carry
out this act of moral justice. But he is prevented from doing so, and this is the
point where the actual version of the miraculous legend begins. Moreover, the
following eleven scenes are dominated by the cultural strategies of honour and
shame, as a closer examination will indicate.
In the field of medieval piety, the relationship between image and donor had
developed in a specific direction. Obviously image and saint were almost
completely identified with one another, that is to say image and saint were
believed to be bound through heavenly ties.’ This relation was conceptualized in
2 Th. Lejeune, ‘Notre-Dame de Cambrón et sa légende’. Revue de l'art chrétien. 7

(1863), 457-81; A. Klassen. ‘Entehrung Maria durch die Juden’, Jahrbuch für
Geschichte. Sprache und Literatur Elsass-Lothringens, 21 (1905), 78-155. For the abbey
of Cambrón, see Monuments pour servir à l'histoire des provinces de Nanuir, de Hainaut
et de Luxembourg, cd. by L. Devillers (Brussels: Publications de la Commission royale
d'histoire, 1896), vol. ut. A slightly extended German version of this article has been
published in Bilder, Texte. Rituale, Zeitschrift für historische Forschung, Beihefte 24. ed.
by K. Schreiner and G. Signori (Berlin: Duncker & Humboli. 2000). pp. 111-38. For the
history of the Jewish communities in Flanders, see C. Cluse, Studien zur Geschichte der
Juden in den mittelalterlichen Niederlanden. Forschungen zur Geschichte der Juden, Abt.
A Abhandlungen (Hannover: Hahnsche Buchhandlung, 2001).
' Some scholars have suggested the theological phrase ‘real presence’ to characterize
the relationship between image and saint according to medieval pious imagination. This
seems also to be true for the image of Notre-Dame de Cambrón. Thus the notion of ‘real
presence’ is addressed by the gesture of the Madonna in the first scene representing the
image desecration: She points to the Christ-child who ‘turns away’ to escape the violent
attack. Directly beneath that scene the Madonna appears in a vision (cf. fig. 3);
interestingly enough, she is depicted not as a ‘bleeding- sculpture but as a ‘real- figure. P.
Dinzelbacher, ‘Die 'Realpräsenz’ der Heiligen in ihren Rcliquiaren und Gräbern nach
mittelalterlichen Quellen’, in Heiligenverehrung in Geschichte und Gegenwart, ed. by P.
Dinzelbacher and D. R. Bauer (Ostfildern: Schwabenverlag. 1990), pp. 115-74; Robert
W. Scribner, ‘Das Visuelle in der Volksfrömmigkei’, in Bilder und Bildersturm int Spät-
f t lu g M s f a t < M ã m t a n n i a x iM ^ a S S t e o f^ c s o s n h n p ^ ñ n »
==¡f s d ì orepratf a ajflriììcoifo^ \

Fig. 2. First field of the panel-painting: The image desecration, detail, c. 1890,
J. van Péteghem, Brussels
terms of a moral economy depending on personal relations, based on an accumu-*
mittelalter und in der frühen Neuzeit. Wolfenbütteler Forschungen, 46. ed. by R. W.

Scribner and M. Warnke (Wiesbaden: Harrassowitz. 1990), pp. 9-20: G. P. Marchai.
‘Bildersturm im Mittelalter'. Historisches Jahrbuch. 113 (1993). 255-82: J. Hamburger,
The Visual and the Visionary: Art and female Spirituality in Medieval Germany (New
York: Zone. 1998). For a critical examination of the concept of ‘real presence-, see N.
Schnitzler, ‘Illusion. Täuschung und schöner Schein. Probleme der Bilderverehrung im
späten Mittelalter", in Frömmigkeit im Mittelalter. Politisch-soziale Kontexte, visuelle
Praxis, körperliche Ausdrucksformen, ed. by K. Schreiner (Munich: Fink. 2002). pp. 221 —
42.
Antisemitism, image desecration, and the problem o f 'Jewish execution ' 361
mvit saparaUu. Safe« Jetan If

Sty.ûrus par puîâaceVroine* úp M kTuierqe pour me -¿Here
Fig. 3. Fifth field o f the panel-painting: Mary appears in a vision to the

blacksmith, detail, c. 1890, J. van Péteghem, Brussels
lation o f honour and prestige. According to this, one could suggest a triangular
model covering saint, donor, and the image that determines the interaction of
individual Christian believers.4
4 H. Belting. Bild und Kult (Munich: Beck. 1990): R. C. Trexler. ‘Florentine Religious
Experience: The Sacred Image', in idem. Church and Community 1200-1600 (Rome:
Edizioni di storiae letteratura. 1987), pp. 37-74; idem. ‘Der Heiligen neue Kleider—Zum
Be- und Entkleiden von Statuen im Mittelalter', in Gepeinigt, begehrt, vergessen—
Symbolik und Sozialbezug des Körpers im späten Mittelalter und in der Frühen Neuzeit.
ed. by N. Schnitzler and K. Schreiner (Munich: Fink, 1992), pp. 365-88: Jeffrey
Hamburger, ‘The ‘Liber miraculorum’ of Unterlinden: An Icon and Its Convent Setting’,
362 NORBERT SCHNITZLER
As the panel of the legend demonstrates, the Madonna of Cambrón had been
'blasphemously violated' by a disbelieving offender, that is to say her ‘virtue'
[virtus]—the usual Latin term—had been tested. Besides the immediate effect
(the image produces blood), the Madonna responded as an insulted medieval
mistress would have done: she calls for a ‘knightly’ follower to take revenge for
the degrading outrage.5 Indeed, the fifth panel shows a male person—a black
smith we are told, actually depicted as a knight—who is to play the part of the
avenger (fig. 3). In a vision the Madonna urges the blacksmith to become her
‘soldier’ and to take revenge for the iconoclastic attack. The impact of our lady's
claim is rather archaic, even in the social context of late medieval Europe. The
‘knightly’ blacksmith is designated to fight a duel, which in the legend is
labelled as ordeal [orciai] that means a judicium dei, against the offender.6
The next scene from the cycle confronts the spectator with a terrifying
representation of an execution (fig. 4). In the very centre of the scene, the
blasphemous offender has been hanged from the gallows, with a pyre beneath
him and two dogs hanging beside him. On the left, a group of noblemen, with the
count in their midst, is watching the execution; in the background we detect
traces of a city-skyline. The painted scene is accompanied by a verse quatrain,
which reads as follows: 'In order to punish his most gruesome outrage / between
two ravenous dogs / he was hanged to avenge the image / upside-down like a
convicted criminal’.7
The term avenge might remind us once more of the suggestion that the former
miraculous legend is mediated and at the same time transformed through the
visual narrative into a story of honour and shame. Such an ‘ethnographic’ view is
underlined by the extraordinary way the blasphemous offender was executed.**5678
in The Sacred Image East and West. ed. by R. Ousterhout and L. Brubaker (Urbana:
University of Illinois Press. 1995). pp. 147-90.
5 Lejeune (as in note 2), pp. 462-63. A similar case of denigrating an image of the
Virgin is discussed in my article "Geschmähte Symbole', in Verletzte Ehre—Ehrkonflikte
in Gesellschaften des Mittelalters und derfrühen Neuzeit. Norm und Struktur. 5, cd. by K.
Schreiner and G. Schwerhoff (Cologne: Böhlau. 1995). pp. 279-302. In general, see: K.
Schreiner. Maria—Jungfrau. Mutter, Herrscherin (Munich: Hanser. 1994); G. Signori.
Maria zwischen Kloster, Kathedrale und Welt. Hagiographische und historiographische
Annäherung an eine hochmittelalterliche Wunderpredigt (Sigmaringen: Thorbecke.
1995).
6 Lejeune (as in note 2). pp. 464-66.
7 Lejeune (as in note 2), p. 481: "Pour punir son tres-cruel outrage/Entre deux chiens
affamés/Fut pendu pour vengier l'image/Les pieds en haut comme meschant aprouvé".
8 The image represents an accumulation of the worst punishments of medieval penal
law. The malefactor has been hanged from the gallows, the very act of earthly jurisdic
tion: this is accompanied by the stake, the most severe form of penance according to
Church law. After this, to increase the degree of shame to its ultimate level, he has been
Fig. 4. Tenth field of the panel painting: Execution of the baptized Jew capite
traverso , detail, c. 1890, J. van Péteghem, Brussels
As we know from legal history, hanging was perceived as a shameful punish
ment, which was avoided and changed to beheading by sword if possible. Thus it
must have been even more shameful to be hanged upside-down. My observations
and remarks contain methodological implications and suggest mutual and corn-
hanged upside-down, with two hungry dogs beside him. On medieval penal law in
general, sec: W. Schild. Alte Gerichtsbarkeit (Munich: Callwey, 19S0). and by the same
author, ‘Verstümmelung des menschlichen Körpers. Zur Bedeutung der Glieder und
Organe des Menschen’, in Erfindung des Menschen—Schöpfungsträume und Körper
bilder 1500-2000. cd. by R. van Dülmen (Cologne: Böhlau. 1998). pp. 261-80.
plex interrelations between a highly elaborated code of late medieval society

(i.e., 'honour’) and the iconographie representation of a miracle legend. As we
will see, the painting from the chapel of Notre-Dame shaped the former legend
and the historical events in a peculiar way. And. after all, there are further
references to other contexts or discourses that we are compelled to reconstruct if
we want to reach an historical level of interpretation.
The L eg en d o f Cam brón, o r The M aking o f C ui tic Im agery by

A d d ressin g A n ti-S em itic A ctivity
Leaving the horrific details aside, I will concentrate now on another important
aspect of the legend. As we learn from the verses beneath the scenes, the icono
clast is identified as a Jew. The visual representations of the panel seem to ignore
this fact, because only the pointed hat, the pileus cornutus, refers to his Jewish
identity. Yet the written sources are absolutely clear about it, and as other
representations of this legend will indicate, it was precisely this anti-Semitic
background that made this narrative attractive for authors and painters of the
fifteenth and sixteenth centuries.9
The legend of the miraculous image of Cambrón already had been established
one hundred years earlier, in the mid-fourteenth century, most probably as a
reaction to the first expulsions of Jews from France, in 1306 and 1321
respectively.10*According to a Dutch chronicle from the 1350s, the iconoclast was
a converted Jew from Mons named Guillaume who had been baptized by the
count of Hainaut himself and who became one of his high ranking vassals."
During a visit to the monastery of Cambrón (dated to the year 1322), he
allegedly felt provoked to attack verbally as well as physically the image of the
Virgin, yet he was able to escape the place unrecognized. It was only four years
later that he was denounced by a Christian blacksmith. As we have learned
already from the panel painting, he was defeated in a duel and consequently
sentenced to death on the gallows. Thus, the written sources render the anti-
Semitic origins of the legend perfectly clear.
9 On the anti-Jewish image-desecration libel in general, see: N. Schnitzler, Ikono-

klasmus—Bildersturm. Theologischer Bilderstreit und ikanokìastisches Handeln während
des 15. und 16. Jahrhunderts (Munich: Fink, 1996). pp. 104-25.
10 See G. Mentgen, ‘Die Vertreibungen der Juden aus England und Frankreich im
Mittelalter', Aschkenaz. 7 (1997). 11-53.
" Johannes de Beke, Croniken van den stichle van Utrecht ende van Hollant, ed. by H.
Bruch. Rijks Geschiedkundigc Publicatien. GS ISO (The Hague: Nijhoff, 1982). pp. 177—
78. Some years earlier, about 1330. Wilhelm of Egmond briefly mentioned the Cambrón
event: Wilhelmi cappelani in Brederode postea monachi et procuratoris Egmodensis
chronicon, ed. by Pijnacker Hordijk (Amsterdam: Mueller. 1904). p. 167.
Antisemitism, image desecration, and the problem of Jewish execution ' 365
This tendency also can be confirmed by two other arguments. First, we should
keep in mind that during the fourteenth century the anti-Jewish image-
desecration libel became an effective strategy to overcome Jewish adversaries.i:
Second, later commentators of the Cambrón legend never forgot to add that the
papal chamber hesitated to grant an indulgence to the monastery because its
commercial aims were too obvious." It therefore becomes clear that the attempt
to establish the cult image of Cambrón had incited as well as increased anti-
Semitic feeling in the county of Hainaut. It was because of these anti-Semitic
origins, as Eric Zafran has argued, that the Cambrón legend reverberated (in
written sources as well as in art) throughout the sixteenth and seventeenth
centuries. Wilhelm Gumppenberg, for instance, the compiler of a voluminous
collection of image-miracles, recounts the story in detail.*14
In his study of the ‘Iconography of anti-Semitism’15*in the Middle Ages,
Zafran took up an older reference to an early book-print of the first decade of the
sixteenth century, entitled ‘Dishonour and Disgrace displayed to Mary by the
Jews’, vaguely ascribed to the Strasbourg humanist Thomas Murner (f 1537).lf’
Compared to the panel of the chapel of Notre-Dame the anti-Semitic elements of
these woodcuts appear even more drastic. As on the panel, the legend is
represented in twelve individual woodcuts, which illustrate more or less the same
episodes of the story. Besides stylistic aspects the woodcuts differ with regard to
the number of offenders, which has now grown to five (fig. 2).17 In other words,
the insult has become an anti-Christian collective action. Furthermore, the scenes
referring to the foundation of a pilgrimage are now replaced by two scenes
presenting legal acts: an oath-taking and an application of torture.'* Stigmatizing
effects are evoked also by Hebrew letters on the garment, by headgear, and by
the yellow badge on the clothes of the Jews.
i; Besides the legends of violated statues of the Madonna, the legend of a crucifix
attacked by a Jew was also very popular. See Schnitzler (as in note 8), pp. 107-10, 113-
22: E. Zafran. "An Alleged Case of Image Desecration by the Jews and its Representation
in Art: The Virgin of Cambrón". Journal of Jewish Art. 2 (1975), 62-71.
" Johannes Trithemius (f 1516). Chronica monasterii Hirsaugiensis, Pars it (Frankfurt
am Main. 1601). p. 216: Wilhelm Gumppenberg. Marianischer Atlas von Anfang und
Ursprung zwölfhundert wunderthätiger Marien-Bilder [. . .] durch R. P. Ma.ximillianum
Wartenberg in das Tettiseli übersetzt (Munich. 1673), p. 66.
14W. Gumppenberg (as in note 13).
15E. Zafran. The Iconography of anti-Semitism (D.Phil. diss.. New York. 1973).
1,1 Anonymous, Enderung und Schmach Mariae von den juden bewissen (Strasbourg,
1515?), ed. by Klassen (as in note 2), pp. 83-86: see also W. List. 'Ein seltener Druck aus
der Offizin des Matthias Hüpfuff, Centralblatt für Bibliothekswesen. 4 ( 1887). 290-93.
17 Klassen (as in note 2). p. 98-99.
"‘Ibid., p. 105-06.
m ^ n ; <5rauf Dm filmiyDt uno

fttuber fteíjõrcrvnb cpbúrcf fy vo: 9cm upr.
Fig. 5. Oath-Taking, from Enderung vnd Schmach der Bildung Mariae von den
Zafran’s main argument concerns the propagandist goal of the booklet to
popularize the myth of Jewish iconoclasm, which he traced back to the early
medieval period (figs 5 and 6). According to him, the booklet had been printed to
strengthen the popular anti-Semitic sentiments of a wide audience. In addition.
Zafran underlines an important remark by Murner stating that Emperor
Maximilian had ordered the painting of these twelve woodcut-scenes on the
walls of the Franciscan refectory in Colmar with the intention of strengthening
the image desecration libel. Probably this was also a measure destined to
legitimize the expulsion of Jews, which at the same time had taken place in the
3g3if ftpr (5rauf Drn mimi ray ter

p c in li'c^ irc g frag c^ D n S fr ju6ficff< in3f«inpjf& crbobt.
Fig. 6. Torture, from Enderung vnd Schmach der Bildung Mariae von den Juden
bewiesen, Strasbourg c. /575
city of Colmar.19
In my view, Zafran has missed a central point of the printed legend. Of
course, the woodcut-illustrations follow a narrative structure similar to the panel.
But in contrast to the panel, the plot is not structured according to the contem
porary discourse of ‘honour’. The woodcuts are, rather, arranged according to
the course of legal procedure, that is to say they affirm the logic of penal law.
19 Zafran (as in note 12), pp. 64-65; For the history of the Jewish community in
Colmar see Germania Judaica, 3.1, ed. by A. Maimón (Tübingen: J. C. B. Mohr, 1987),
pp. 657-62.
368 Norber t Schni tzler
The series starts with a crime and ends with the execution of the (main) offender.
The different stations of legal procedures are systematically transferred to the
field of visual narrative. Thus, I would assume that these variations were an
effect of structural changes in the legal system itself.20 The execution scene,
above all, now is clearly intended not only as the story’s climax but also as a
climax of anti-Semitic propaganda (fig. 7). These arguments might become all
the more convincing if we bear in mind that Murner, the presumed author of the
booklet, was a well-known scholar of law and member of the emperor’s
'Academy of Poets’.21
The text itself uncovers further dimensions of meaning, especially in relation
to the unusually harsh form of punishment. It points to an anti-Semitic tendency
of medieval penal law in general, which is confirmed by the application of a
special sort of punishment against Jews. The heading of the image gives little
more than a short description of the image: 'How the count ordered the Jew to be
hanged together with two male dogs and to be burned by a fire beneath the
gallows’. But the corresponding parts of the poem give a more extended version.
The presumed reader is reminded that this horrific representation should be
‘taken as an example / How all Jews are outlawed who commit such abominable
outrage’. Moreover, it is said that the sacrilegious offence was no misdeed of an
individual at all but, rather, an outcome of Jewish stubbornness: ‘Although they
are witness to such important miracles / They never learn from that / Therefore
he should receive this penance / Burned and hanged as well’.22 This is a clear
statement. The author approves of the principle intended by penal law, which
means he approves of the idea of deterrence.
In contrast the chronicles of the fourteenth century say that the Jew had ‘only’
2,1 E. Wadle, ‘Die Entstehung der öffentlichen Strafe", in Perspektiven der

Strafrechtsentwicklung (Ringvorlesung im Sommersemester 1994 an der Universität des
Saarlandes), ed. by H. Jung, H. Müller-Dietz and U. Neumann (Baden-Baden: Nomos-
Verlags-Gesellschaft, 1995); Die Entstehung des öffentlichen Strafrechts, ed. by D.
Willoweit (Cologne: Böhlau, 1999); G. Schwerhoff. ‘Devianz in der alteuropäischen
Gesellschaft. Umrisse einer historischen Kriminalitätsforschung', Zeitschrift für
historische Forschung. 19 (1992). 385-414; R. van Dülmen. Verbrechen. Strafen und
soziale Kontrolle (Frankfurt am Main: Fischer, 1990).
21 Thomas Murner—Elsässischer Theologe und Humanist 1475-1537. Eine Aus
stellung (Karlsruhe: Badische Landesbibliothek, 1987).
22The complete passage reads (my translation): "The sentence was read out by the count /
That the Jew should be put to death / As well as alienated from earth / by hanging upside-
down / with two enormous dogs beside him / which should be taken as an example / How all
Jews are outlawed / who commit such abominable outrage / And shame our religion day
after day. Although they are witness to such importam miracles / They never learn from that.
/ Therefore he should receive this penance / Burned and hanged as well / On account of his
abominable crime’, cf. Klassert (as in note 2), pp. 126-27.
Fig. 7. Execution of the baptized Jew, from Enderung vnd Schmach der Bildung
Mariae von den Juden bewiesen, Strasbourg c. 1515
been burned at the stake. There is no mention of any degrading form of
execution. However, hanging someone upside-down was not just a late medieval
anti-Semitic fantasy. As we know from contemporary court-house records or
legal ordinances, it became a common practice from the thirteenth century
onwards, lasting well into the seventeenth century.’1
11 See N. Schnitzler. ‘Juden vor Gericht—Soziale Ausgrenzung durch Sanktionen', in

Formen und Entwicklungen herrschaftlichen Strafens. ed. by H. Schlosser. R. Sprandel,
and D. Willoweit (Cologne: Böhlau. 2002). pp. 285-307; G. Kisch. ‘The Jewish
Execution in Medieval Germany and the Reception of Roman Law", in L'Europa e il
The R eality o f P enal L aw A g a in st Jew s
Evidence shows that the earliest testimonies of the application of hanging

upside-down can be found in fourteenth-century France as well as in Aragon.
The papal registers contain letters of remission, for instance one of a canon of the
chapter of Beauvais from 1326 who applied to the pope for absolution since he
had participated in stoning an executed Jew hanging upside-down on the
gallows.*24Earlier, in 1296, the annals of the Alsatian town of Colmar recount the
case of a Jewish thief who had been sentenced to death, noting that he had been
hanged upside-down on the gallows. Incredibly, the offender stayed alive for
eight days and even was able to free himself, but because of his injuries he was
unable to escape.25
A number of other cases demonstrate that the application of this form of
punishment during the fourteenth and fifteenth centuries all over medieval
Europe, including the imperial lands of Germany. But it should be noticed that in
the majority of cases the punishment had been ordered against thieves, not
against religious offenders.2'’ This renders the case of the image-desecrator of
Cambrón unique.
Of course, there are other facts that might indicate a lingering tendency to
enforce the rules of penal law against Jews. It was Ulrich Tengler (t 1511?), a
contemporary of the Alsatian humanist Murner and author of a highly influential
book of law (‘The Leyenspiegel’ dating from 1511) who was the first to give a
relevant (and explicit) definition of this special form of punishment: ‘Jewish
execution’, he wrote, ‘means to pull or to drag a Jew accompanied by two mangy
dogs by chains or ropes to the common gallows place where he has to be hanged
upside-down’. Such a description was quite common, but then Tengler added:
*... and in this way [some] one who sticks to his Jewish heresy is put from life
to death’.27 From a juridical point of view, Tengler labelled it as a so-called
‘arbitrary’ measure. But due to the fact that he dedicated a complete paragraph to
this form of punishment, it became an almost institutionalized element of the
diritto romano. Studi in memoria di Paolo Koschaker, il (Milan: A. Giuffrè, 1954), pp.
65-93.
24 S. Simonsohn, The Apostolic See and the Jews. Studies and Texts. 94 (Toronto:
Pontifical Institute of Mediaeval Studies, 1988), pp. 348-49.
25Annales Colmarienses Maiores. Monumenta Germaniae Historiae. Series Scriptores
XVII, ed. by G. H. Pertz (Hannover: Hahnsche Buchhandlung. 1861 ), p. 222.
26Schnitzler (as in note 23).
27 Ulrich Tengler. Leyenspiegel (Strasbourg. 1560). p. 118'. It is also worth mentioning
the term iudicium iudaeorum as it is used in medieval canon law; cf. W. Pakter, Medieval
Canon Law and the Jews. Abhandlungen zur rechtswissenschaftlichen Grundlagen
forschung, 68 (Ebelsbach: Rolf Gremer, 1988). pp. 47-67.
Antisemitism, image desecration, and the problem o f 'Jewish execution ’ 371
penal law code of the late Middle Ages. Tengler’s comment emphasizes two
aspects: first, he declared it a form of punishment exclusively reserved for Jews.
Second, in his view it was an act of penance, which means a measure of Church
law. resulting from the ‘stubborn' Jewish ‘heresy’. Therefore he never discussed
the aspects of honour or shame.
It has been suggested that the reception of ancient Roman law was one of the
main reasons for the transformation of the social status of Jews in late medieval
times.2*The present version of the image desecration legend and its presumed
author Murner seems to confirm this suggestion. The original version of the
image-miracle has been restructured in favour of legal aspects of the story. That
also means a change in its visual narrative. The woodcut—a medium of the new
printing technology—transformed the chronicler’s story as well as the former
visual narrative. Instead of these, it now is conceptualized from its end, the scene
of execution. That means the author tried to resolve a problem, which may be
phrased by the question: How to apply penal law against Jews?
The booklet marks some critical points of official law which—allegedly!—
conceded a lot of privileges to Jews and saved them from being accused in court.
At different levels in the legal procedure, the author demonstrates this prob
lematic issue. His main intention can be derived from one central phrase of the
lines: ‘No Jew will be caught / due to the accusation by only one (Christian)
man’.2’ This verse also is quoted in an abridged edition of Mumer’s booklet
published some years later by Pamphilus Gengenbach in Basel.30 Thus one can
conclude that the booklet of the Cambrón legend carries a clear didactic purpose
which was at least influenced by the reception of Roman law. Its aim was to
teach the audience how to apply public penal law to the members of Jewish
communities.
R itualistic F ram ew ork— The ‘W orld Turned U p sid e-D o w n ’
Another important aspect of the visual representation of hanging capite traverso

can be detected if we take a look at iconographical parallels. As we know from
Italian sources of the same period, this type of symbolic execution was also the
usual way in which the Florentine city-state acted against traitors to the com
munity. They were stigmatized as effeminate and dishonoured enemies of the
commune, and this also was demonstrated on a symbolic level through propa28
28 Kisch (as in note 23), pp. 65-93.

Klassen (as in note 2). p. 117. ('So grifft man nit den iuden an / vff reden eines
eintzigen man’].
" Pamphilus Gengenbach. Da: ist ein erschrockenliche historij von fünff schnöden
juden [. ..] (Basel, 1517).
Fig. 8. Study o f a man hanging upside-down, fifteenth century, Andrea del Sarto.
© Gabinetto di disegni U ff zìi, Florence
ganda strategies: the figures of the traitors were painted on an outside wall of the
Palazzo Vecchio, hanging upside-down (fig. 8).”
As we know from studies of medieval popular culture, such a practice of
symbolic inversion formed a general pattern of collective activity questioning the
real impact of hierarchical order as well as the constraints of social roles.
Nonetheless, the iconography of symbolic inversion received special importance
or the members of medieval nobility. According to the customs for these high-
ranking nobles, a shield with the arms of the family concerned was buried along
side its former bearer whenever the branch died out with him.“ Regarding the
shields hanging on the inner walls of churches, it had also become a useful and
effective practice to turn round those owned by nobles who died a shameful
11 W. Brückner, Bildnis und Brauch—Studien zur Bildfunktion der Effigies (Berlin: E.

Schmidt, 1966), pp. 205-16: G. Ortalli. '(.. .) pingatur in Palatino’—La pittura infamante
nei secoli xiii-xvi (Rome: Jouvence. 1979); S. Y. Edgerton. Pictures and Punishment. Art
and criminal Prosecution during the Florentine Renaissance (Ithaca: Cornell University
Press, 1987).
“ V. Filip, ‘Wappen’, in Lexikon des Mittelalters (Munich: Artemis-Lexma. 1977-2000).
vin, cols 2031-34: R. Heydenreuter, 'Wappenrecht', in Handwörterbuch zur deutschen
Rechtsgeschichte 5. ed. by A. Erler and E. Kaufmann (Berlin: E. Schmidt, 1998), cols 1139-
44.
Antisemitism, imane desecration, and the problem o f 'Jewish execution ' 373
Fig. 9. So-called ‘Schartdbrief by the nobleman Alexander of Oberg addressed

to five noblemen who had answered for Duke Erich I of Calenberg-Göttingen as
debtors; their arms are shown upside-down. About 1540, Niedersächsisches
Staatsarchiv, Wolfenbüttel, Sig.: 1 Alt. 26. © Niedersächsisches Staatsarchiv,
Wolfenbüttel
death on the gallows. That meant that they were even socially dead, without any
memoria kept alive by family or by friends. To remove shields from their proper
place on church walls or to publish illustrated letters with a nobleman’s arms
painted upside-down, proved to be an effective strategy. By its victims, it was
perceived as a severe attack on family honour.33
The same is true for another form of medieval symbolic defamation, the so-
called libelli famosi, which had been used to urge unreliable debtors to pay what
was owing. From the fifteenth century onwards, through to the seventeenth
century, the threat of becoming the subject of such a offending letter was ob
viously very effective in northern Europe (fig. 9). But—and this is of special
importance—it was the visual rhetoric of these libelli, their use of denigrating
symbols and metaphors, which made them work. The strategy of symbolic
J‘ R. Quanter. Die Sellami- ami Ehrenstrafen in der deutschen Rechtspflege (Dresden:

|s.n.], 1901). p. 149.
inversion could achieve its goal because the ties of honour that connected the
families were strong. As the present example shows, the authors used different
kinds of symbols, frequently highly-valued personal signs, as for instance seals,
vestments, or the heraldic elements, to stigmatize their adversaries as disgraceful
and unworthy debtors.”
Guido Kisch, the well-known scholar of medieval Jewry Law, already has
pointed to the same conclusion. He likewise refers to the genre of libelli famosi
as a key to explain historically the so-called ‘Jewish execution’. In addition, he
suggests an important link to the medieval legend of Judas Iscariot. As far as I
know, the medium of pittura infamanti makes the only explicit reference to the
avaricious traitor as personification of the Jews in general.'5
Thus, it seems obvious that the code of honour marks a central point. It must
be emphasized though that this was not a specifically anti-Semitic element but a
symbolic strategy widely used to stigmatize an individual as an outcast deprived
of any reputation or honour. Methodologically speaking, Mumer’s woodcuts
refer to two sometimes conflicting discourses of late medieval society: on the
one hand a highly elaborated academic discourse on law: and on the other hand a
likewise elaborated elitist symbolic system of honour and shame.'“ Both of them
dominated and transformed the traditional iconography of image miracles and
image worship.
The M eaning o f the D og a n d M ed ieva l P roto-R acism
A most interesting part of the debate on the so-called ‘Jewish execution’ focused
on the symbolic value of the dogs. It has been argued that the execution of an
offender along with animals was intended to express an affinity of both subjects,
that is to say they both—man and beast—possessed the same bad and brutish
character. This assumption has led to the suggestion that the intention to shame
and to stigmatize Jewish offenders developed in consequence of the hegemony
” O. Hupp, Scheltbriefe und Schandbilder—Ein Rechtsbehelf aus dein 15. und 16.
Jahrhundert (Munich: J. C. Manz, 1930); G. Schmidt. Libelli famosi. Zur Bedeutung der
Schmähschriften. Schehbriefe, Schandgemälde und Pasquille in der deutschen
Rechtsgeschichte (Cologne: Lang, 1985): M. Lentz, ‘Ehrverletzung als Rechtsbehelf? Zur
Erforschung spätmittelalterlicher und frühneuzeitlicher Schmähbriefe und Schandbilder
im Rahmen eines Entwicklungsgeschichte des Strafrechts", in Neue Wege strafrechts
geschichtlicher Forschung, ed. by D. Willoweit and H. Schlosser (Cologne: Böhlau.
1999). pp. 55-81.
'5 Kisch (as in note 23). pp. 88-90.
M. Dinges, ‘Ehrenhändel als ‘kommunikative Gattungen", Archiv für Kultur
geschichte. 75 (1993), 359-93.
Antisemitism, imane desecration, and the problem o f 'Jewish execution' 375
of Christian ideas in Europe, especially the idea of ‘otherness’ in respect of the

Jewish population.”
With obviously denigrating words, a visitor at the Basel Council in 1430,
named Gattaro, commented on the execution of two Jewish thieves by the city’s
hangman. ‘They’d better ask for baptism’, he wrote, ‘unless they want to die like
beasts’.5* This was the same sort of remark that the chronicler of the city of
Dortmund, Dietrich Westhoff, made when, in 1486, a Jewish thief was put to
death outside the city walls, hanging upside-down between two dogs. Westhoff
closed his report with the following words: ‘Thus the three [i/'c!| dogs were
hanging together on the gallows and in this way they also ended their lives’.59
But we also can find strong arguments that deny such an unambiguous
interpretation of the symbolic meaning of dogs. In the context of penal law. for
instance, the symbolic meaning of the dog was evident in the punishment of
‘sacking’ (in Latin: poena cullei). Perpetrators who had been placed in sacks
together with dogs, cats, or snakes usually had been sentenced for the killing of
their parents (in Latin: parricidia). Medieval glossatores who commented on
Roman law gave the following explanation: the dog is attached to symbolize the
offender, because he had violently attacked his parents like a dog, ‘which is still
blind for some days after birth, unable to recognize his parents’. What seems all
the more noteworthy is the fact that this punishment was applied to both
Christians and Jews.*
Another aspect of meaning can be gathered from the anti-Jewish polemics of
medieval theology. Medieval churchmen and preachers used the metaphor of the
dog to criticize Jewish ‘stubbornness’ and to stigmatize baptized Jews as alleged
renegades, because they behaved like ‘young dogs’ which would not recognize
their ‘master’ and always returned to eat their own vomit.*401 However this notion
clearly referred to the religious differences between Christians and Jews and not
to racial differences.42
57 R. Glanz. 'The “Jewish Execution“ in Medieval Germany’, Jewish Social Studies. 5

(1943), 3-26. From a critical perspective see G. Kisch. ‘Nationalism and Race in
Medieval Law’, Seminar, 1 (1943). 65-73.
58 R. Wackernagel. ‘Andrea Gattaro von Padua. Tagebuch der Venetianischen
Gesandten beim Concil zu Basel. 1433-35’, Basler Jahrhuch (1885). 49-52.
59 D. Aschoff, ‘Eine jüdische Hinrichtung in Westfalen I486’, in Jüdisches Leben
(Aust.-Kat.) (Gütersloh. 1989), pp. 67-78.
40Tengler (as in note 27), 118V.
41 See, for instance, Frutolfi et Ekkehardi Chronica. Monumenta Germaniae Historiae,
Series Scriptores vi. p. 208. For a revised edition (including German translation), see
Frutolfi et Ekkehardi Chronica necnon Anonymi Chronica imperatori, Freiherr von Stein
Gedächtnisausgabc. 15. cd. by F.-J. Schmale and I. Schmale-Ott (Darmstadt: Wissen
schaftliche Buchgesellschaft. 1972). pp. 107-09.
42 Kisch (as in note 37); cf. D. Nirenberg. ‘Race as an Analytical Concept in the Study
Perhaps this discussion of the different symbolic meanings of dogs can also
provide a starting point for developing a more pragmatic explanation for apply
ing this unusual form of execution against Jews. Reports of Jewish executions
that failed were frequently inserted into miracle collections. These legends gave
accounts of Jewish convicts hanging upside-down who miraculously stayed alive
for hours or even days and sometimes asked for baptism in order to die as
Christians—which meant being executed by sword or dying in hospital. Nowa
days, those fantasies may sound more or less cynical, but they may reflect the
real intentions of contemporary judges and clerics to offer Jewish convicts a last
opportunity to convert and to die as Christians.4 435*
The medieval ritual of execution was thus not only expected to take effect on the
audience in terms of ‘deterrence’, for instance, as our modem designation ‘theatre of
horrors' might suggest.44 On the contrary, it was intended to affect the repentant
sinner inwardly: during the execution his inner life should undergo a fundamental
transformation. The goal of these forms of ritual acting was therefore either to bind
or to dissolve, either to reconcile the repentant sinner or to dehumanize the
perpetrator, and to exclude him from the binding ties of social community.45
However, we should keep in mind that in the context of the iconography of
justice and law, the symbol of the dog always could convey multifold meanings.
The ‘sleeping dog’ in medieval courthouse scenes, for instance, implied rather
positive connotations. This is also true for these real-life dogs which were carried
during rituals of submission by their masters, who walked in a sack-dress and
with ashes on their heads, as symbols of penance.45
Visualizing Justice
As 1 have tried to demonstrate. Murner reshaped the story of the image-miracle

in order to construct a new narrative grounded in the discourse of justice and
law. Thus the focus of anti-Semitic implications also shifted from the image-
desecration libel to the problem of legal privileges of Jews and how to overcome
them. Therefore, the new (propagandistic) medium of printed woodcuts was
of Medieval Iberian Anti-Judaism’, in Jüdische Gemeinden und ihr christlicher Kontext in

kulturräumlich vergleichender Betrachtung (5.-IS. Jahrhundert), ed. by A. Haverkamp et
al. (Hannover: Hahnsche Buchhandlung, forthcoming).
43 See the Diary of Gattaro, in Wackernagel (as in note 38).
44 This is the title of a book by the German historian Richard van Dülmen. Theater des
Schreckens (Frankfurt am Main: Fischer, 1985).
45 E. Cohen. The Crossroads of Justice (Leiden: Brill. 1993).
45 B. Schwenk, ‘Das Hundetragen. Ein Rechtsbrauch im Mittelalter’. Historisches
Jahrbuch. 110 (1990). 289-308.
Fig. JO. Death of Judas, c. twelfth century, from the so-called ‘Hildegard-
Codex’. © Bayerische Staatsbibliothek, Munich
intended to emphasize an overall didactic purpose of the enterprise: to demon
strate the advantages and the power of public penal law.
Another corresponding example of shifting semantics can be found in the
iconography of Judas’s suicide, especially in the feature of the ‘hanging tree’.
During the thirteenth century, Judas not only became the paradigmatic figure of
despair but also moved to the centre of the juridical discourse on suicide and the
problem of guilt. The offence of betrayal and Judas’ death at the rope—guilt and
repentance, so to say—became the primary aspects of the scene. Thus the motif
of suicide is transformed into an act of punishment.47
I would suggest that this development, which could be described as a
sharpening of terminology through the aid of juridical categories and concepts,
can be observed also in pictorial sources. In an illustration from a twelfth-century
manuscript, the so-called ‘Psalter of St Hildegard’ (fig. 10), we see Judas
hanging not from a tree but from the gallows. His hair is tousled, falling down
his neck, and ravens are flying around his corpse. What we actually are
confronted with is a scene of execution; nothing reminds us of suicide. The
illustrator presents the image of a traitor who is put to death and at the same time
figures as a spiritual idea, viz., the authority of iustitia, symbolized by the
executed body. This meaning is underlined by the corresponding picture above.
Usually the motif of Judas’s death is accompanied by the ‘betrayal in the garden’
or by a representation of the ‘crucifixion’. In this case it is a courthouse scene
depicting Jesus, who is examined by Pilate, identified by his headwear, the
47 N. Schnitzler, 'Judas' death—Some remarks concerning the iconography of suicide’,

in Law and its Image, ed. by N. Schnitzler (as in n. 1), pp. 103-18.
throne-like chair and his legs crossed like a judge.

Stating that pictures of medieval times are indispensable sources for the
historian may sound like a truism, in particular for the ears of those working in
the field of legal history. But asking for the shift of semantics, for the ways they
affect the relation between text and image, for the tensions and reciprocal
comments deriving from it. means to obtain methodologically important starting
points for more reliable interpretations. It was my concern to demonstrate the
mutual stimulations that in the late medieval period combined new juridical
concepts such as "deterrence’ with corresponding concepts of representation. At
least, the reader should keep in mind that the concept of public penal law in itself
consists of a specific technique of visualization in which the body of the
malefactor represents the projection-screen.
Fram ing history w ith salvation
S0REN KASPERSEN
discussion of pictures and history, including pictures as historical
A sources, might well induce a mild attack of methodological vertigo at a

time when perceptions of history and the question of the viability of
historical reconstruction and the problems surrounding the contextual study of
pictures are being debated exhaustively. According to Hayden White, ‘the best
grounds for choosing one perspective on history rather than another are
ultimately aesthetic or moral rather than epistemological'.1 I am basically in
agreement with him, whether we are talking of the Middle Ages, whose view of
history will here be considered from a history of art perspective, or of historical
research in modern Western culture. It would be good if the way were cleared
‘for the reconstitution of history as a form of intellectual activity which is at once
poetic, scientific, and philosophical in its concerns’? And is it not true to say that
the humanities are actually poetic and philosophical by implication, and that they
are at their most relevant when they are aware of these dimensions? The real
touchstone of historical writing, therefore, is probably how it substantiates its
claim to be a science. It is this that distinguishes it from the historical writing of
other ages. But it is precisely the scientific nature of the humanities that has
provoked a great deal of discussion and disagreement in these postmodern,
deconstructivist and pluralistic decades.
To my mind, historical consciousness, as a separate scientific way of thinking,
belongs to what might be described as ‘the scientific turn' that is found in Kant,
' H. White. Metahistory—The Historical Imagination in Nineteenth-Century Europe,

2nd edn (Baltimore: The Johns Hopkins University Press, 1975), p. xii.
; See ibid.
380 Soren K aspersen
the first philosopher to try to isolate the conditions for scientific cognition. This
‘turn’ marks the growth of a new culture, which now increasingly attempts to
answer the great existentialist questions in a scientific way. This places us in a
culture whose search for truth differs radically from that of the Middle Ages.
However, at the same time, the status of the humanities has been seriously
challenged. The natural sciences are making great advances on all fronts—in the
study of the universe, the Earth, human beings, etc. These sciences in their
examination and understanding of the origins of the universe and life, their
building blocks and energy, the structures and processes of this incompre
hensible cosmos, are so close to the deepest layers or secrets, as it were, that we
must expect that the knowledge they will be able to reveal in the future is what
earlier ages sought in myths and religion.
Although, as Hayden White says, the cognitive interests of the humanities are
based on morality and aesthetics, this simply means that like all other scientific
knowledge, they are related to the historically, socially, and psychologically
determined creation of consciousness. And any radical difference between them
and the natural sciences is probably due primarily to the fact that the main focus
is not so much on epistemological and empirical relation between the subject and
the world as on the education and experience of the subject in a particular
cultural context. For example, when Panofsky referred to art history as a
humanistic discipline, the subtext was that it is a discipline that by means of its
models of interpretation would provide knowledge respecting a particular view
of the individual and of the world, and which was thus concerned with specific
historical circumstances and put specific questions to the material.' Since this
ideal has vanished, and since the concept of educational formation has been
heavily relativized and the differences between the actual education and cogni
tive interests of the subject have come to the fore, it is both natural and necessary
for art history, and the humanities in general, to adopt a pluralistic approach in
terms of methodology and theory of science and to concent itself with a broad
spectrum of conditions that relate to the position of the subject and the formation
of the consciousness of the age. In my opinion, these fundamental conditions do
not prevent us from continuing to see the justification for the existence of the
humanities in the scientific nature of the disciplines. Hayden White would
probably have called the following discourse ‘romantic’ in its ‘mode of
employment’, ‘formist in its 'mode of argument’, and ‘conservative’ in respect
of its ‘ideological implications’. But as far as its scientific credentials are
concerned, you will have to judge for yourselves.
In Keldby Church on the island of Mpn, in the High Gothic period around the
' See E. Panofsky, "The History of Art as a Humanistic Discipline’, in The Meaning of
the Humanities, ed. by T. M. Greene (Princeton: Princeton University Press, 1940), pp.
89-118, reprinted in idem. Meaning in the Visual Arts, 2nd edn (Harmondsworth:
Penguin. 1970), pp. 23-50.
Framing history with salvation 381
Fig. 1. Kelclby Church. Decoration on the eastern part o f the north wall o f the
nave. c. 1325. © Lennart Larsen, Nationalmuseet, Copenhagen
year 1325. a large salvation history decoration was painted onto the walls of the
nave, with a very elaborate Last Judgement scene as the unifying main theme on
the Triumphal Arch.4 At the eastern end of the north wall of the nave, large parts
of a six-zoned decoration have been preserved, reduced by the vault which was a
later addition. The three top friezes depict scenes from the Old Testament, the
lower ones scenes from the story of Anne and Joachim as well as the childhood
of Jesus (fig. I ). There must have been about forty episodes altogether in this
section of the decoration, which made up about a quarter of the entire decoration
on the south and north walls, and it is possible that the west wall, too, had
narrative scenes in many friezes. Thus there might have been as many as two
hundred motifs, and fragments on the eastern part of the south wall with scenes
from the story of Adam and Eve indicate that the decoration started here. We do
4 See J. Kornerup. 'Om Kalkmalerierne i Keldby Kirke’, Kirkehistoriske Samlinger. 5

(1897-99), pp. 149-66; Danmarks Kirker—Freest</>Amt, ed. by Nationalmuseet (Copen
hagen: G. E. C. Gads Forlag. 1933-35). II, PP. 962-66 and fig. 5: S. Kaspcrsen. ’Kalk-
maleri og samfund 1241-1340/50’. in Kulturblomstring og samfundskrise i 1300-tallet.
ed. by Brian Patrick McGuire (Copenhagen: C. A. Reitzels Boghandel), pp. 130-33 and
figs. 16-18.
382 SORKN K a SPRRSBN
not know, however, to what extent the chronological order of the sequence was
followed. Nor can we say for certain whether the new Covenant was confined to
the last section of the north wall. If so, the Old Testament cycle would have been
much more extensive than the New Testament series, but there is already an
example of this kind of pattern in Pope Sixtus iv’s (432-40) mosaic decoration in
the church of Santa Maria Maggiore in Rome, and this weighting can also be
seen, for example, in Master Bertram’s high altar triptych in St Petri’s church in
Hamburg, to which we will be returning later.
Of particular interest in this context is the fact that a large rectangular panel
has been inserted into the salvation history decoration in the middle of the north
wall, at the same height as the three lower zones. The panel contains a circular
composition with a broad frame like a mirror (fig. 2). The central scene shows
two enthroned figures: a king with a lily-topped sceptre and a bird on his raised,
right hand:56and a smaller figure, a woman with a crown, hands raised in front of
her breast, the palms facing outwards to signify that some wonderful act is taking
place, that something exalted can be seen. She is enclosed by a scroll, only a few
letters of which, unfortunately, have survived. Similarly, only a small part of the
inscription that surrounded the whole scene remains. In the frame immediately
above the pair is another enthroned king en miniature, flanked by two courtiers,
one with one hand grasped around the wrist of the other hand, in the manner
prescribed in the Norwegian Speculum Regale." The other figure is gesticulating
and appears to be referring to a coronation scene to the right of the king: two
bishops are crowning a standing figure, while an abbot or ecclesiastical helper
carrying a crozier looks on. It is not possible to say for certain whether the
central figure is male or female, though it does have short hair.
There can be little doubt that the scenes refer to historical characters and
events. But is it a wedding, a coronation, or something quite different that is
being depicted? There are theories that the ‘royal couple' were Eric Menved and
Queen Ingeborg, who were married at Helsingborg in 1296, long after Eric, as a
5The bird. dove(?). may have a purely symbolic meaning—see below—but it may also
refer to a real insignia, most likely a sceptre with a very short staff or rather a long staff,
now missing. E.g.. two sceptres of silver are mentioned in connection with the crowning
1247 of the Norwegian king. Hâkon Hâkonsson. one crowned with a gold cross, the other
with an eagle of gold; see Martin Blindheim. 'Spor av hersker- og kroningsinsignicr i
norsk middelalder'. in Kongens mala og cere—Skandinaviske herskersymboler gjennom
1000 dr. ed. by M. Blindheim. P. Gjærder, and D. Sæverud (Oslo: Universitetets
Oldsaksamling, 1985), p. 65; and his Soga um Hâkon Hâkonsson (Oslo: Del norske
Samlaget, 1928). pp. 244-45. An eagle of gold also is mentioned among the regalia of the
Swedish king, Birger Magnusson. in 1311. but here as part of the coronation robes; see R.
Cederström. De svenska riksregaliema och kungliga vardighetstecknen (Stockholm:
Kungliga livrustkammaren, 1942). p. 10.
6See Kongsspegelen, trans, by A. Hellevik (Oslo: Det Norske Samlaget. 1965), p. 96.
Fig. 2. Keldby Church. Decoration on the central part o f the north wall o f the
nave, c. 1325. © Nationalmuseet, Copenhagen
child king, came to the throne in 1286,7*9or that it was Valdemar Atterdag, who
married Princess Helvig of Holstein at Whitsuntide 1340 at Spnderborg Castle
and who that same year was acclaimed at the Viborg assembly on St John's Eve.*
But the first event seems rather early and the second one too late, in view of the
general style of the paintings. The state of preservation, however, means that we
have to operate within a wide range when it comes to setting a date. The only
intermediate possibility is that the king is Eric Menved’s younger brother,
Christopher n, who married Euphemia of Pomerania before 1307 and succeeded
Erik to the throne in 1320. But if it refers to that wedding there is no significant
difference, as far as dating is concerned, from Eric Menved’s wedding in 1296.
Two facts are significant for further determination of the course of events.
First, the woman is clearly subordinate to the king. She is smaller than he and is
sitting in a slightly withdrawn position. Second, what relation does the corona
tion scene have with the two enthroned figures? Why is the attention focused not
on the coronation of the king, but on that of the queen or a third person? It is of
interest here that in 1324, that is, four years after he came to power, Christopher
n was anointed and crowned at Vordingborg, together with his son, Eric, by
Archbishop Esger Juul. And at Christmas that same year, another important
event in Christopher’s reign took place. Count Ludvig of Brandenburg, son of
King Ludwig of Bavaria, arrived at Vordingborg Castle to marry Christopher’s
young daughter, Margrethe. This alliance must have seemed a very promising
move in Christopher it’s attempt to strengthen his position and create a strong
royal power in what must, in many ways, have been a very difficult situation.'7
Thus, instead of representing a royal couple enthroned at their wedding and/or
coronation, the decoration at Keldby might well be showing Christopher sitting
in state with his daughter in connection with her wedding.10The bird in his right
hand could signify a connection with the German ruler by reference to the so-
called ‘bird sceptre’ in Aachen, a silver-gilt rod crowned by a dove from around
1220 or soon afterwards." At the same time, the woman’s gestures are rendered
7 See Niels M. Saxtorph. Danmarks kalkmalerier (Copenhagen: Politikens Forlag.

1986). p. 158.
s See Danmarks Kirker (as in note 4). pp. 965-66.
9 See S. Tägil. Valdemar Atterdag och Europa. Bibliotheca Historica Lundensis. 9
(Lund: C. W. K. Gleerup, 1962). pp. 1-14.
10The crown of the female figure is then a bridal crown.
" The sceptre is part of a treasure that Richard of Cornwall as German king (1257-72)
founded in the Aachener Münster. This kind of sceptre was otherwise unknown in Germany,
while a ‘Sceptre with the Dove- belonged to the insignia of English kings—as a sign of
peace-keeping. See Karl von Amira. Der Stab in der germanischen Rechtssymbolik.
Abhandlungen der Königlich Bayerischen Akademie der Wissenschaften [Philos.-philol. u.
hist. Kl.. 25.1] (Munich, 1909), pp. 119-21; Emst Percy Schramm. 'Über die Herrschafts
zeichen des Mittelalters'. Münchner Jahrbuch der Bildenden Kunst. 3.1 (1950), pp. 45, 51.
more understandable as an expression of the marvellous event that is taking

place, and it becomes clear why a scroll is needed around her. Thus, the episode
at the top of the ‘mirror’s’ frame then refers to the slightly earlier coronation of
the son, Eric, attended and blessed by Christopher, whose succession is hereby
formally assured. The seated, much-effaced figure with the head covering in the
middle of the frame seems to be pointing towards the young princess, judging by
the turn of the body and the uplifted left hand. Finally, the banquet at the top left
corner of the rectangular panel might illustrate the wedding feast in an abbre
viated form or another ceremonial meal connected with the function. Does the
dog in front of the table signify that the bride will be a faithful wife?
This possible identification of characters and events can be corroborated by a
number of accentuations and iconographie features in the remaining parts of the
salvation history decoration. First, we note that the Trinity Judge on the Triumphal
Arch, over and above a few individual figures, is flanked partly by two groups of
apostles and then by male and female saints, and that the female saints are placed on
the right hand side of the judge (fig. 3). The first is St Catherine, with a sword and a
spiked wheel. The next is on a completely different scale: she stands upon a small,
tortoise-like dragon, which extends across the border between the two zones. Above
the vault, the top of a cross-staff and perhaps a crown can be discerned. The saint is
undoubtedly St Margaret, the name-saint of Christopher's daughter. We can also see
that the foremost figure in the corresponding group of male saints is not John the
Baptist, as one might expect in a Last Judgement scene where Mary is praying so
intensely and persistently that her knees are bleeding, but St Lawrence, the patron
saint of the archdiocese and of the cathedral at Lund. Behind him stands John the
Baptist. The order may have been a sop to Esger Juul for the coronation of
Christopher and his son—despite the fact that relations with the bishop were strained.
Or it might simply have been a mark of the special relationship with the patron saint
of the archdiocese, which naturally came about through the coronation.
If we turn to the six-zoned composition on the north wall (fig. 1), it also
reveals a number of interesting features. There are several breaks in the chronol
ogy. In the stories from the childhood of Jesus, scenes of the shepherds receiving
the message of the angels out in the fields and their adoration of the Child have
been moved to the lower zone, so that in the upper zone we jump directly from
the Nativity, a scene that was greatly damaged by the later installation of the
great window, to that of the Flight into Egypt. And this scene not only contains
the apocryphal story of the arrival of the Holy Family at the town of Sotinen but
also presents a very unusual version of the event. Beneath two arcades stand two
figures, their right hands raised in greeting as the Holy Family arrives (fig. 4).
The foremost, nimbed figure wears a long cloak, fastened across the chest by a
At the end of his article. P. E. Schramm emphasizes that ‘Alles, was der König an Insignien
und Gewändern trägt, hat einen Sinn oder strebt darnach, einen Sinn zu erlangen. Die
Auslegung ist jedoch in ständiger Bewegung und wird nie eindeutig’ (p. 59].
Fig. 3. Keldby Church. The Last Judgement on the east wall o f the nave. c. 1325.
©Author
Fig. 4. Keldby Church. The Flight into Egypt/Sotinen, c. 1325. ©Author

Framing histon’ with salvation 387
brooch, and seems to be veiled. The other figure is obviously male. East of this
there is a gabled arcade with a trefoil arch crowned by turrets, that is, like the
canopy above Mary in the Annunciation. Beneath it a crowned virgin is seated,
turning west, with a smaller figure kneeling in adoration in front of her and
another behind her as well. Both these figures are rather fragmentary. She holds
a small book in her right, raised hand, and she seems to open her arms towards
the arriving family. The easternmost part of the canopy—and the last section of
the scene(?)—is concealed by the vault.
This depiction of the town of Sotinen does not have much to do with the
apocryphal text,12 but since the story records that both ruler and inhabitants
converted to Christianity when the Holy Family arrived, the nimbed figure
and/or the crowned virgin might well personify Ecclesia, receiving Christ as her
bridegroom.*11This interpretation might be underpinned by a ‘vertical' reading of
the program (cf. fig. 1): the frieze above depicts the childhood story of the Virgin
Mary, and the reception of the Holy Family to Sotinen corresponds with the
Meeting of Anna and Joachim at the Golden Gate, while the panel just above the
crowned virgin in Sotinen most likely depicted the Presentation of the Virgin to
the Temple. One level up, the Adoration of the Golden Calf is represented in an
Exodus-frieze just above the Virgin's Birth. The opposition of the newborn Mary
and the new-cast idol may call to mind the contrast between the Golden Calf and
Ecclesia in the Bible moralisée.'* The apparent intertextual relationship, then, of
the three pictorial sequences heightens attention to themes such as meeting and
reception, bride and bridegroom, husband and wife, Mary and Ecclesia, the true
Saviour versus false gods. At the same time, the unusual iconography of the
Sotinen sequence seems to point towards a metaphorical reading of the scene or,
perhaps, rather to indicate that it was dictated by the need to fulfil the require
ments of a transposed reading: at Christmas, in 1324, the bride, Margrethe,
welcomes her young bridegroom to Vordingborg.
Medieval thought was no stranger to this kind of blending of factual history
and the salvation story, the procession of any ruler into a city could be and was
12 For the gospel text of Pseudo Matthew, see Evangelia Apocrypha, ed. by
Constantinus Tischendorf (Leipzig: Avenarius & Mendelssohn, 1853), pp. 84-87;
Apocrypliical New Testament, trans, by M. R. James (Oxford: Clarendon Press. 1924). pp.
75-76; or Jakobs Forevangelium og Del ucegte Matthanisevangelium, trans, by Knud
Banning (Copenhagen: G. E. C. Gad. 1980). pp. 31-36.
11 On the triumphal arch of Santa Maria Maggiore in Rome, the Arrival in Sotina together
with the Adoration of the Magi signifies the foundation of the ‘church of the heathen-; see G.
A. Wellen. Theotokos: Eine ikonographische AbhandlunJ über das Gottesmutterbild in
frühchristlicher Zeit (Utrecht and Antwerp: Specuum. 1961), pp. 97-116.
IJ See. e.g., Bible moralisée—Faksimile-Ausgabe im Originalformat des Codex Vindo-
bonensis 2554 der Österreichischen Nationalbibliothek, ed. by Reiner Haussherr, Codices
Selecti. 40 (Graz: Akademischer Druck- und Verlagsanstalt. 1973), fol. 25v.
388 SO REN K a SPERSEN
often seen in the light of Christ’s entry into Jerusalem.15Therefore, when a son of
Ludwig, king of the Romans, arrived at Vordingborg at Christmas to marry
Christopher's daughter, it would have seemed obvious to seize on a parallel from
the childhood of Jesus.16 In fact, tableaux vivants representing the nativity of
Christ were displayed during King Henry vi’s first entrance to Paris on 2 Dec
ember 1431,17and still later, when Louis xi entered Lyon in 1476 the day before
the Annunciation, mystery plays about both the infancy of the Virgin and the
childhood of Christ were performed.1819We have no records of theatrical perfor
mances in Vordingborg 1326, and it is less than plausible that mysteries actually
took place,1’ even if it is tempting to imagine that the unusual representation of
Sotinen referred to some staged version which made it readable at that time.70
15See Ernst H. Kantorowicz. Laudes regiae—A Study in Liturgical Acclamations and

Medieval Ruler Worship, University of California Publications in History, 33 (Berkeley:
University of California Press. 1958); Idem. ‘The King’s Advent and the Enigmatic
Panels in the Doors of Santa Sabina', Art Bulletin 26 (1944), pp. 207-31 and illustrations;
Bernard Guenée and Françoise Lehoux. Les entrées royales françaises de 1328 à 1515,
Sources d'Histoire Médiévale. 5 (Paris: Centre National de la Recherche Scientifique.
1968).
16See Guenée and Lehoux (as in note 15), pp. 202-03.
17 See Guenée and Lehoux (as in note 15), p. 67: ‘Et, a l'endroit de la Trinité, avoit
escarfaulx moult richement aomez, es quel(e)z estoient. figurés de personnaiges vifz. les
ystoires desmonstrans la Nativité de Notre Jhesu Crist: lesqueles personnes aucunement
ne se mouvoient, et apparoient estre ymages, et estoient bien huit vins personnaiges'.
18See Guenée and Lehoux (as in note 15), pp. 202-03.
19The earliest French record of theatrical representations during a royal entry is from
1380. when Charles VI entered Paris—see Guenée and Lehoux (as in note 15). pp. 12 and
58.
It is difficult to find comparative material to elucidate the depiction of Sotina in
Keldby. When the Sotina-episode takes place as an adventus, it is normally, in Byzantine
art, the personified Egypt, Aegyptika. who welcomes the Holy Family, while different
models are used in Western art: Aphrodosius and his retinue, an anonymous group of
inhabitants, a personification like Aegyptika etc.; see Karl Vogler. Die Ikonographie der
'Flucht nach Aegypten' (diss. Heidelberg; Arnstadt: Otto Böttner, 1930). pp. 16-19;
Gertrud Schiller, Ikonographie der christlichen Kunst. I (Güthersloh: Gerd Mohn. 1966).
pp. 128, 129. figs 310, 315; Lexikon der christlichen Ikonographie, I—Vili, ed. by
Engelbrecht Kirschbaum, et al. (Rome et al.: Herder. 1968-76), II. cols 43-50. fig. 1 (C.
Schweicher and G. Jászai): Marceli Restle. Die byzantinische Wandmalerei in Kleinasien.
I—in (Recklinghausen: Aurel Bongers, 1967). il, figs 77. 153. 267. and in, figs 307. 386.
395: Kantorowicz (1944. as in note 15). pp. 220-21. (fig. 33a]. The triumphant character
of the Flight into Egypt in its early phases is stressed by Hildebrecht Hommel. pointing
out the ancient sources of the motif: see idem, ‘Profectio Mariae—Zur Ikonographie der
‘Flucht nach Ägypten” , Theologia Viatorum, 9 (1963), pp. 95-112 and ill. The nimbed,
femalef?) figure in Keldby Church is most likely a transformation of Aphrodosius or
Whether the unusual broadcasting of the glad tidings of the arrival of the
Saviour by one of the shepherds in the lower register on a large horn, an ‘Alpen-
horn‘. also springs from the festivity of a princely adventus is debatable, of course.
But it should be noted that the frieze contains a large banqueting scene as well,
where Stephen the stable boy tells Herod of the newborn Saviour. In any case, the
wedding was quite an event. The chronicle of Detmar tells us that many princes
and noblemen were present and that the margrave stayed on until Epiphany.!l
Linking the historical events and the salvation story in the context in which
they appear to have been placed, there is a picture under one of the original
windows whose connection with the two spheres is conspicuously ambiguous
(fig. 5). Under a canopy that resembles a ciborium, a young man sits writing,
reading, or leafing through a large manuscript. The cloak fastened in front of the
breast with a buckle suggests that he is an ecclesiastical figure, but not least the
round cap with a large turned-up brim argues against his being an evangelist. It
must be either an Old Testament prophet or a contemporary chronicler, and his
youthful appearance would seem to indicate the latter.
However, the somewhat unreal or, perhaps, allegorical nature of the scene
springs from the lectern, which as a sturdy stem emerges from a compound of a
triple corbel-stepped base, a rhomboid object—possibly a large precious stone—
and a king’s crown, culminating in a large laciniated ‘leaf blade’. Perhaps even
more important, in front of the domed baldachin with trilobed ornaments or
perforations five staves (originally seven?) rise from the arcade arch, and three
more (originally four?) appear behind the dome, together with a stem with
winding offshoots bearing bunches of grapes. By analogy with ‘barren’ rods, the
vine may be an allusion to Aaron’s sprouting rod (cf. Numbers 17.17-26), which
in turn is a préfiguration of the Virgin Birth, or to Joseph’s flowering rod, when
he was singled out for betrothal to Mary. The scene thus may be interpreted as
both a prologue to and a symbol o f the Incarnation, and as a wonderful statement
about betrothal, about the blessing which, like a fruitful vine, will spring out of
the union of the bride and bridegroom.
In the light of all this, if we turn back to the enthroned figures in the ‘earthly
Aegyptika according to the ecclesiological symbolism of the scene, and the following
male figure a representative of the inhabitants of Sotina. Yet. a single deputy is an unusual
feature and the crowned, seated Virgin is without parallel, making the whole Sotina-
depiction difficult to ‘read' if it does not. e.g.. refer to a specific, historical event. In fact,
the whole ‘adventus'-iconography is an early phenomenon and very surprising, at least in
Western art as late as the fourteenth century.
21 See Die Chroniken der niedersächsischen Städte: Lübeck, ed. by K. Koppmann. Die
Chroniken der deutschen Städte vom 14. bis ins 16. Jahrhundert. 26 (Leipzig: S. Hirzel.
1899).
390
Soren K aspersen
Fig. 5. Keldby Church. Prophet/chronicler? c. 1325. ©Author

Flaming history with salvation 391
mirror’,“ the king’s dove-like bird and lily sceptre might well recall the Annun
ciation: to signal the arrival of the bridegroom, Christopher raises the ‘dove
sceptre’ towards his daughter (fig. 2). Furthermore, the lily and bird motifs are
repeated as ornaments on the seat of his throne. Finally, the dimensions between
the king and the young woman, presumably his daughter, are grosso modo the
same as between the archangel and Mary at the Annunciation, and the Virgin is
crowned like the young bride.“
As the only iconic pictures among long narrative sequences, the two last-
mentioned panels are like concentrates of the whole decoration: they mirror its
character of annunciation, advent, and revelation. ‘Ecce mitto angelum meum’ (cf.
Malachi 3.1) was one of the liturgical utterances employed at the reception of
rulers in the Middle Ages/4 In the present decoration, the Lord sends his angel both
to Joachim and to Anna(?):5 and to the Virgin and the shepherds; he reveals himself
to Moses on Horeb, ‘the mountain of God’ (cf. Exodus 3.1) and originally also on
Mount Sinai;“ he allots Aaron to Moses as a spokesman and precedes him
“ On the complexity and centrality of the mirror metaphor in the Later Middle Ages, see
Herbert Grabes. Specillimi. Mirror und Looking-Glass—Kontinuität und Originalität der
Spiegelmetapher in den Buchtiteln des Mittelalters und der englischen Literatur des 13. bis
77. Jahrhunderts. Buchreihe der Anglia. 16 (Tübingen: Max Niemeyer, 1973).
'The crown is an unusual attribute for the Virgin at the Annunciation, but antecedents
can be mentioned, e.g.. a relief on Bamberg Cathedral from c. 1230-40—see Gertrud
Schiller (as in note 20), I. p. 50 and fig. 89.
:j See Kantorowicz (as in note 15). pp. 217-21 : Malachi’s prophecy continues: ‘. .. et
praeparabit viam ante faciem meam: et statim veniet ad templum suum Dominator quem
vos quaeretis, et angelus testamenti quem vos vultis' (3.1). The Lord, who suddenly will
come to his temple, here refers to the second coming of Christ, and in that respect the
Arrival in Sotina heralds the Last Judgement-scene on the east wall of the nave.
“ The original, late Romanesque window easternmost on the north wall was situated
on a level with the window above the prophct/chronicler and a little further towards east,
leaving space for at least one episode between the Annunciation to Joachim and the
Meeting at the Golden Gate. Following the chronology of pseudo-Matthew, it could have
been the Sacrifice of Joachim during the angel's first visit or Joachim’s Second Vision,
both episodes known from the Arena Chapel, see Evang. Apocr. (as in note 12), pp. 54-
60; Apocr. New Testament (as in note 12), p. 73 or Jakobs Forevang (as in note 12), pp.
31-36, and James H. Slubblebine. Giotto—The Arena Chapel Frescoes, Critical Studies
in Art History (London: Thames and Hudson. 1969), pp. 75-77 and figs. 7-12; but most
likely the Annunciation to Anna was represented, according to the chronology of The
Golden Legend by Jacobus de Voragine; see ibid., pp. 118-20.
“ Otherwise, the sequence with the Golden Calf seems unmotivated, and there was
space enough for at least one episode between "the Burning Bush' and the original
window, followed by 'the Casting of the Golden Calf’. On Mount Sinai. God also
promises Moses: 'Ecce ego mittam angelum meum, qui praecedat te, et custodiat in via, et
introducat in locum quem paravi' (Exodus 23.20).
392 S 0RBN K a SPHRSEN
Fig. 6. Keldby Church. Tombstone for Henning Moltke, his wife and his
daughter (?), 13557/1360-70. from J. B. Loffler, Danske Gravstene fra Middel-
alderen (Copenhagen, 1885)
as a pillar of cloud or fire to lead the Israelites on their wandering in the desert
(cf. Exodus 13.21-22). Finally, as the newborn Saviour he is acclaimed at the
arrival in Sotina and proclaimed at the banquet by the vivified cock. All this
gives the decoration a specific tone well suited to reverberate with the
significance of the historical event and to invest it with profound layers of
meaning.
In the Middle Ages, Keldby Church was approximately a day’s journey from
Vordingborg, just over thirty kilometres away. The size and content of the
decoration are evidence of the status of the donors and planners. The Crown had
long owned a number of properties on the island of Mpn. For example, in 1177—
78, three and a half ‘bof (c. 700 acres) in Keldby itself and one ‘bof in Tornby,
a settlement that has now disappeared but which probably lay within the parish,
were conveyed to Valdemar the Great by St Peter’s Monastery at Næstved.27
Furthermore, Keldby parish adjoins Stege, which during the thirteenth century
flourished as market town, sheltered by a royal castle, Stegehus or Stegeborg,
built by Valdemar the Great (1157-82). The castle and the landed properties on
Mpn were often mortgaged to North German princes.28 But in May 1322,
Christopher n repaid a mortgage on the island of two thousand marks of silver,29
and the island appears to have been in his possession until May 1326.50In Keldby
Church there is, furthermore, a tombstone showing the knight, Henning Moltke
(f 4.10.1326), his wife Elsebe (t 25.3.1355), and their daughter!?), Lady
Christina (t 1347), originally erected in St George's chapel at Spejslsby in the
parish by Lady Christina’s husband, Fikke Moltke, knight and member of the
royal council (‘Rigsrâd’) (fig. 6). Henning Moltke stemmed from an old Wendic
family, of whom, from the end of the thirteenth century, several members began
to assert themselves in Denmark as royal officials, and as a knight Henning
consequently also swore fealty to the king. Henning Moltke’s wife, Elsebe. was
most likely a daughter of Cecilie Jonsdatter Lille (t 1307), a member of the once
important Hvide family. Through her, Henning himself may have owned landed
property on M0n. not least in the parishes of Keldby and Stege, where the Hvide
family had had greater properties years previously, most recently known through
the will of Cecilie’s father, Jon Lille.'1 In any case, after 1360, Fikke Moltke
27See Diplomatarium Danicum (Copenhagen: Ejnar Munksgaards Forlag 1938-), 1.3,

No. 67: and Scriptores rerum danicarum, ed. Langebek et al. (Copenhagen, 1772-1878).
2-8 The earliest known mortgaging occurred in 1290: see Frede Bojsen. Af Mans
Historie, I—IX (Copenhagen: Akademisk Boghandel. 1905-26), u (1906). p. 159.
29See Dipl. Dan. (as in note 27). 2.8, Nos 416 and 433: and F. Bojsen (as in note 28), II
(1906), pp. 171-72.
50It was mortgaged to Henry II of Mecklenburg and Johan ill of Werle 3 May 1326; see
Dipl. Dan. (as in note 27). 2.9. Nos 261 and 278.
" See Erik Ulsig, Danske adelsgodser i middelalderen (Copenhagen: Nordisk Forlag,
Fig. 7. Keldby Church. The Last Judgement: Purgatory, Resurrection o f the

Dead, and coat o f arms, c. 1325. ©Author
(f 1371) designated himself ‘of M 0n\ where the main part of his landed
property, consequently, must have been situated, properties which he quite well
may have inherited through his wife. Christina.* 32 In Keldby parish there are
remains of an old castle, called Gammelborg, which may have been a manor
owned by members of the Moltke family.3334
There are thus good grounds for assuming that Henning Molkte, as the king's
vassal and someone who had a special connection with the parish of Keldby.
planned and probably also donated the decoration in more or less close co
operation with the king himself. Unfortunately, the three great coats-of-arms in the
Last Judgement scene on the east wall of the nave no longer have any blazoning,
and among the risen, presumably the saved, we can see only a bishop with an ele
vated chalice and host, a kneeling and possibly tonsured figure facing the bishop,
as well as, higher up. another clerical figure wearing a pluvial and again with an
elevated chalice and host—perhaps, the parish priest, who was named in the
inscription surrounding the chancel arch (fig. 7).u The kneeling figure could, of
1968). pp. 25-26, 37-39 (Jakob Sunesen and Countess Ingerd), and 55-58 (Jon Lille).
32See F. Bojsen (as in note 28). ill (1917). pp. 13-14, 51. 184-88. Christina was a near
relative of Henning Molkte and Elsebc. probably their daughter.
33See ibid., p. 14.
34 Only a small fragment of an inscription in two lines with the word PLEBANI is
preserved.
Fig. 8. (a-b) Hamburg, high altar from St Petri’s Church, 24 panels by Master
Bertram, c. 1379-83. ©Kunsthalle. Hamburg
course, be the vassal and the bishop of Roskilde. but the relationship between them
and the portrayal of Purgatory just behind the bishop indicate, rather, that they
illustrate the importance of holding masses for the dead.
At any rate, the decoration at Keldby Church seems to show us an exciting
linking together of contemporary historical events and the history of salvation that
provided at the time a framework for understanding the former, at least on a higher
level. It is easy to pounce on the possible ideological, not to mention propagandist
exploitation of this situation. However, it is more important here to establish that
this linkage makes the Keldby Church decoration the kind of historical source of
which very few survive from the Danish Middle Ages, a source that throws a rich
spectrum of ideas and value statements over the events of 1324— if we have
identified them correctly. It was probably a clerical brain that was behind the
scheme, and there is not much evidence that Christopher n was a king whose
thoughts were dominated by religious matters. He emerges from the sources as
more of a practical politician, even a cynical power-seeker. But he clearly saw the
importance—at least in a raw political sense—of being an ecclesiastically cones-
crated and crowned king. The broad dimension of salvation history in which his
execution of government is placed in the decoration, possibly also springs more
from power-political than from religious considerations. Taken as a whole, it may
be seen as an impressive, rhetorical staging and invocation of the new alliance with
the king of the Romans, in spe the Holy Roman Emperor.-'5
Finally, in order to elucidate this further, we will consider Master Bertram’s
retable for the high altar at the St Petri Church in Hamburg, dating from 1379 to
1383, just over half a century later.** This altarpiece. with a double set of wings,
also depicts in the second position a salvation history cycle, here in two zones,
comprising twenty-four panels in all (figs 8a-b). The series goes from the
Creation up to and including the childhood of Jesus, although the Old Covenant,
as indicated above, has been allocated the most space, constituting three quarters
of all the scenes. It is a strongly structured narrative, permeated by inter-scene
references, and this being so, numerous layers of meaning are unfurled, which
intermingle and play together in a subtle way. It is primarily an Augustinian
account of the two societies, of the City of God versus the City of Satan, from
when they were predicted at the beginning of time through the split between the
good and the bad angels—in the first scene of the upper zone—via their actual
appearance on earth through Cain and Abel—in the first two scenes in the lower
zone (cf. fig. 9)—to the time of the New Covenant, where the final scene of the
Holy Family resting during the Flight into Egypt paints a picture of the City of
God on pilgrimage through the world of sin and evil (cf. fig. 10).'7One notes the
dramatic juxtaposition of the Murder of the Innocents in Bethlehem and the
fleeing family, and how Jesus’ own sacrificial death on the Cross is anticipated
by the malevolent donkey’s attack on the green sheaf, which explains Mary’s
35 Ludwig was crowned as Roman king in Aachen on 25 November 1314, and as

emperor in Rome on 17 January 1328.
* See, e.g., Christian Beutler, Meister Bertram—Der Hochaltar von Sankt Petri:
Christliche Allegorie als protestantische Ärgernis (Frankfurt am Main: Fischer Taschen
buch Verlag. 1984); and S. Kaspersen. ‘Grabower-alteret—En analyse af rnester Bertrams
frelseshistoriske cyklus', Synligt og usynligt—Studier tilegnede Otto Norn. ed. by Hugo
Johannsen (Herning: Poul Kristensens Forlag. 1990). 27—48, including ill.; and 'Der St
Peter-Altar zu Hamburg—Eine Analyse des heilsgeschichtlichen Zyklus Meister
Bertrams', in Der St peter-Altar zu Hamburg, Verein für katholische Kirchengeschichte in
Hamburg und Schleswig-Holstein (Husum: Matthiesen Verlag Ingwert Paulsen jr.. 1993).
pp. 8^44.
” See Kaspersen (1990; as in note 36). pp. 38—4-4. or (1993; as in note 36). pp. 23-36.
Fig. 9. Hamburg, six panels on the high altar from St Petri 's Church, uppermost
‘The Fall o f the Rebel Angels’, ‘The created Firmament/ palatium and ‘The
Creation o f Sun, Moon, and Stars’, last The Sacrifice o f Cain and Abel’, ’Cain
Slaying Abel', and ’The Building o f Noah's Ark', c. 1379-83. ©Elke Walford,
Kunsthalle. Hamburg
anxious look.
In this salvation history narrative. Mary is as much a bride as a mother and, as
a bride, she also represents Ecclesia. It is she whom the Child Jesus points to
every time—at the Nativity and the adoration of the three kings (cf. fig. 11 ). at
the presentation in the Temple, and during the flight into Egypt (cf. fig. 10). The
future union between Christ and his Church, the magnum sacramentum of
Augustine, is foreseen already in the creation of Eve through her carnal union
with the old Adam and her awe-struck spiritual union with the new Adam—
music-making angels strike up at the great wedding feast (cf. fig. 11 ). And the
bridal chamber or heavenly hall is ready from the dawn of time. In the second
398 S O RE N K a SPERSEN
'•Il ■ '■*. L 4- ' ■■ vA J-------- -4 ,

.............
Im i -4
. >4 . . w . . , . .
. iW 4 m-4 -1> -ï -—1 j< .— pi 4
TV.-......... • . - . V ................
ip j >- — .1 X
F/g. /0. Hamburg, six panels on the high altar from St Petri's Church,
uppermost ‘God’s Curse on Adam and Eve’, ‘The Expulsion o f Adam and Eve’,
and ‘The First Labour’, last 'The Presentation in the Temple’, 'The Massacre o f
the Innocents’, and 'The Rest on the Flight into Egypt’, c. 1379-83. © Elke
Walford, Kunsthalle, Hamburg
panel, the Creator stands on the firmament, according to Petrus Comestor called
coelum, quia celat, because it conceals heaven above the firmament, which is the
dwelling of the blessed, in Greek uranon, id est palatium (cf. fig. 9). By his use
of gesture, the Creator indicates that this palace is now clean and adorned ready
for the bride’s unveiled meeting with the bridegroom.’8
As a high altar. Master Bertram’s work also alludes, of course, to the
Eucharist, partly via sacrifices such as Cain and Abel’s offering of sheaf and
lamb and Abraham’s surrender of his own son, and partly through various
,s See Kaspersen ( 1990: as in note 36). pp. 30-32. or (1993: as in note 36). pp. 12-16.
Fig. II. Hamburg, six panels on the high altar from St Petri’s Church,
uppermost ‘The Creation o f Eve', ‘The Admonition o f Adam and Eve', and ‘The
Fall’, last ‘The Annunciation', 'The Nativity’, and ‘The Adoration o f the Three
Kings’, c. 1379-83. ©Elke Walford. Kunsthalle. Hamburg
emphases on food and drink as in the story of Jacob and Esau (cf. fig. 8a),
emphases that are also statements that one can eat one's way to both salvation
and damnation (cf. i Corinthians 11.29). This aspect is seen most clearly in the
parallel drawn between God admonishing Adam and Eve and the birth of Jesus
(cf. fig. 11 ). At the top, the Creator points toward the Tree of Knowledge, whose
fruit is not to be eaten by the inhabitants of the Garden of Eden. At the bottom.
Joseph holds the Child Jesus out to Mary, which is a very unusual feature in
Nativity scenes, but which reveals the Child as the redeeming sacramental food,
as the bread of life that has come down from Heaven. And the ox and the ass do
not stand at the manger; they lie down by and nourish themselves from a
Gcirtlein with all its attendant connotations of Mary as the Mother of God and
400 SOREN K aspersen
the Garden of Eden. It is hardly surprising, therefore, that food also has a central
role in the final scene depicting the Holy Family at rest (cf. fig. 10).39
Master Bertram’s altarpiece refers to a number of central aspects of the
Christian universe in a detailed, differentiated way, and in that sense it is a
hyper-iconographic or learned work, but it mediates its insights to a large extent
through the basic idiom of style and composition, not least through countless
inter-scene references of all kinds. The most important thing for us is to try to
understand how a history of salvation cycle of this type found its way at all onto
a high altar retable, for this is not at all usual. A slightly earlier silver frontal in
the cathedral at Pistoia is possibly the closest parallel, but that was executed in
several stages between 1357 and 71 .40
It is my hypothesis that the scheme of the St Petri altarpiece is bound up with
Hamburg’s historical situation and that the most pertinent circumstance in the
1370s is the relationship with the Holy Roman Emperor, primarily, Charles iv of
Bohemia. From the middle of the 1360s right up until his death in 1378, he
attempted to involve the Hanseatic towns, especially Lübeck and Hamburg, more
actively in the politics of the empire. As a trading town at the mouth of the Elbe,
Hamburg was a kind of entry port to Prague, the new capital of the empire, via
the river, and Charles wanted to turn the city into a centre for receiving and
transferring goods from the entire northern European area to Prague and vice-
versa.4' The emperor’s plans seem to have been received sympathetically in the
city and to have been promoted with particular enthusiasm by the mayor,
Heinrich Hoyer (t 1375). Nothing concrete may have ensued from the various
initiatives,4' but Hamburg's involvement in the practical politics of that period is*41
41See Kaspersen (1990; as in note 36), pp. 36-37, or ( 1993; as in note 36). pp. 20-22.
■“’See Kaspersen (1990; as in note 36), p. 41. or (1993: as in note 36). pp. 28-29: and
A. Venturi, Storia dell'arte italiana (Milan: Ulrico Hoepli. 1906). iv. pp. 940-52, incl. ill.
41 See Heinrich Reineke. Kaiser Karl IV. und die deutsche Hanse. Pfingstblätter des
Hansischen Geschichtsvereins. 22 (Lübeck: Verlag des Hansischen Geschichtsvereins.
1931). Charles had already granted Hamburg a privilege in 1359 to prosecute pirates and
highwaymen, especially on and at the Elbe, and thereby strengthened the town's
sovereignty over the river—see Horst Tschentscher. 'Die Entstehung der hamburgischen
Elbhoheit (1189-1482)'. Zeitschrift des Vereins für Hamhurgische Geschichte. 43 (1956).
pp. 1-48 (pp. 19-22); Urkundenregesten zur Tätigkeit des deutschen Königs- und
Hofgerichts bis 1451. ed. by Bernhard Diestelkamp, vn: Die Zeit Karls iv (1355 April-
1359). ed. by Friedrich Battenberg (Cologne: Böhlau. 1994). pp. 350-51. no. 497: or
Hansisches Urkundenbuch, ed. by Konstantin Höhlbaum et al. (Halle: Buchhandlung des
Waisenhauses. 1882-86). Ill, pp. 225-26. no. 460.
4' Yet, see Wolfgang von Stromer, ‘Der kaiserliche Kaufmann—Wirtschaftspolitik
unter Karl iv.’, in Kaiser Karl tv.—Staatsmann und Mäzen, ed. by Ferdinand Seibt
(Munich: Prestei. 1978). pp. 67-69. For a fine survey with further literature of Hamburg's
political and social circumstances in the late medieval period, see Jürgen Sarnowsky. ‘Die
evidenced, for example, by the fact that, at the end of the fourteenth century, the
city had at least five altars and an important confraternity dedicated to the
Bohemian Saint Wenceslas, the first martyr king of Bohemia and the name-saint
of Charles iv’s eldest son and successor/3
In the years when Lübeck and Hamburg were actively involved in practical
politics, new life was probably breathed into a world of ideas concerning their
importance in the empire, that is, their place in the im p e riu m C hristianorum and
its governance." As a result of Charles iv’s upgrading of their political impor
tance. the towns could now regard themselves as the advocates of the empire,
especially Lübeck, which was ‘empire immediate’ (‘Reichsunmittelbar’). In an
edict in German in 1374, the emperor elevates the people of Lübeck to deputies
of the empire and to guardians against all harmful elements such as murderers
and pirates, because he has learned about the particular trustworthiness, justice,
and common sense of the mayor of Lübeck and because he has seen how the
townspeople do their best to root out evil and protect the righteous, to the honour
of himself and the Holy Empire and to the general good of the country and
promotion of the peace. In all places, even in foreign parts, in all countries and
towns with all kinds of rulers and forms of government, the people of Lübeck, on
behalf of the emperor, are to pursue those who disturb the peace and try them in
their own courts.*435*Lübeck’s role in preserving the peace at land and sea was
hereby elevated to a function of empire, and a Hanseatic naval power became the
chief naval power of the empire. Even if the citizens of Hamburg were subordi
nate to the counts of Holstein,* they may well have entertained similar ideas
about themselves in relation to the Holy Roman Empire.47
politische Entwicklung und die sozialen Strukturen Hamburgs im Spätmittelalter', in Die

Kunst des Mittelalters in Hamburg—Aufsätze zur Kulturgeschichte, ed. by Volker
Plagcmann (Hamburg: Dölling und Galitz. 1999), pp. 97-108. ill.
43See H. Reineke (as in note 41 ). pp. 59-60.
44The Peace at Stralsund. 1370. had also considerably strengthened the position of the
Hanseatic League.
45See H. Reineke (as in note 41 ), pp. 29^-5 (40-41 ).
* And the emperor consequently supported the count against Hamburg's complaints
and its claim for freedom; ibid., pp. 11-12.
47 See above, n. 41. The relations between town and empire in the period generally
were marked by clashes of interests; see Peter Moraw, 'Monarchie und Bürgertum’, in
Kaiser Karl IV(as in note 42), pp. 43-63 and notes. For a thourough investigation of the
complex relationship between the Hanse and the empire in the fifteeenth century, see
Horst Wernicke. 'Hanse und Reich im 15. Jahrhundert—ihre Beziehungen im Vergleich’,
in Beiträge zur hansischen Kultur-, Verfassungs- und Schiffahrtsgeschichte, Hansische
Studien, x, ed. by Horst Wernicke and Nils Jörn (Weimar: Verlag Hermann Böhlaus
Nachfolger, 1998). pp. 215-37. 1 have not had access to E. Voigt. ‘Reichsgewalt und
hansisches Bürgertum'. Wissenschaftliche Zeitschrift der Ernst-Moritz-Amdt-Universität
Salvation history decorations, understood as events of the Old and New

Testaments related to each other as parallel or continuous chronological progres
sions, are generally classed with didactic Christian art. They have no cult or
devotional function but are considered to be 'painted likenesses, made for the
instruction of the ignorant, so that they might understand the stories and so learn
what occurred', as Pope Gregory the Great wrote.4* However, such more or less
extended Bible stories could evoke the conception of and be understood as a
reference to Augustine's two civitates,4" particularly when the emergence of these
two 'cities’ was emphasized so strongly, as is the case with Bertram’s altarpiece
in St Peter’s church in Hamburg in which the unusual ending to the tale of the
salvation story with ‘The Massacre of the Innocents’ and not least ‘The Rest on
the Flight’*48*50 also seems to conjure up a vision of the Church on a pilgrimage
through the world of sin (fig. 10).
However, in the Middle Ages, the civitas Dei could hardly be conceived without
intimating questions of how and by whom the Chosen People were to be led to
salvation. And the same type of questions were likewise present in any reference to
the condition and possible improvement of the imperium Christianorum:
Greifswald, Gesellschafts- und sprachwissenschaftliche Reihe, 12 (1963/65-66).

48See Caecilia Dawis-Weyer, Early Medieval Art 300-1150: Sources and Documents
(Englewood Cliffs: Prentice-Hall, 1971). p. 48: Lately there has been much discussion of
St Gregory’s statements; see Lawrence G. Duggan, ‘Was Art Really the 'Book of the
Illiterate’?', in Word and Image, 5 (1989). pp. 227-51; Celia M. Chazelle. 'Pictures.
Books and the Illiterate: Pope Gregory’s Letters to Serenus of Marseilles', in Word and
Image, 6 (1990), pp. 138-53; and L'image—Fonctions et usages des images dans
l'occident médiévale: actes du 6e 'International Workshop on Medieval Societies', ed. by
J. Baschet and J.-C. Schmitt. Cahiers du Léopard d’Or. 5 (Paris; Le Léopard d'Or, 1996),
pp. 7-26.
44 For a comprehensive Scandinavian material and its conceivable political implica
tions. see S. Kaspersen. 'Über spätmittelalterliche Genesis-Serien in der Kirchenmalerei
der nordischen Länder—Eine politisch-theologische Skizze’, in Bild och Brak. ed. by
Tove Riska. Taidehistoriallisia Tutkimuksia. Konsthistoriska Studier. 8 (Helsinki: The
Society for Art History in Finland. 1985), pp. 115-41, inch ill. Furthermore S. Kaspersen.
‘Cotton-Genesis, die Toursbibeln und die Bronzetüren—Vorlage und Aktualität’, in
Bemwardinische Kunst—Bericht über ein wissenschaftliches Symposium in Hildesheim
vom 10.10. bis 13.10.1984. ed. by Martin Gosebruch (Göttingen: Erich Goltze. 1988). pp.
79-103. inch ill. (84-96).
50It is worth emphasizing the originality of Bertram's panel and the fact that the motif
of the ‘Rest’ during the 'Flight’ is relatively young and does not gain ground until about
1500; see Hans Möhle, in Lucas Cranach d. Ä.—Die Ruhe auf der Flucht nach Ägypten,
Werkmonographien zur bildenden Kunst, 111 (Stuttgart: Philipp Reclam jun.. 1966). esp.
p. 17. According to G. Schiller (as in note 20), i. p. 132, the representation by Bertram is
the earliest known example.
Framing hisloiy with salvation 403
weil die kaiserliche Herrschaft immer auch als weltliches Pendant zur geistlichen
Herrschaft des Papstes in der ganzen Christenheit verstanden wurde. Die
Verbindung des deutschen Königtums mit dem universalen Reich hat es mit sich
gebracht, daß sich der deutsche Staatskörper von Anfang an nicht primär als
politische, sondern als religiöse Ordnung verstehen konnte und sein Zweck nicht
die Macht um der Macht willen, sondern die Vorbereitung des ewigen Heils für
die Menschen war. So erweist sich die Staatsideologie des mittelalterlichen
Reiches als religiös und individualistisch.51*
Thus, when Hamburg, as a city on the empire’s northern periphery, was unex
pectedly drawn into imperial politics, new parameters were set for the doings of
the city, which could now view its political, commercial, and religious activities
even more strongly in the light of Salvation history. It had become a part of the
City of God. which had from the dawn of time been on a pilgrimage toward the
great wedding feast at the end of time. The high altar retable in St Petri, the
townspeople’s main church, showed the biblical prelude to the current situation
and placed it in an overall framework. St Augustine divided world history into six
or seven ages, corresponding to the six days of Creation and the six stages of life.
The first age lasted from Adam until the Flood, and the sixth age began with the
Incarnation and stretched out to Judgement Day. The seventh age was the seventh
day, the day of rest, ‘our Sabbath, whose end (finis) will not be an evening, but the
Lord’s eternal eighth day. sanctified by the resurrection of Christ’.*
Bertram’s altarpiece begins with the days of Creation, which take up the first
two ‘sections’, that is, six panels (cf. fig. 8a), and ends with the creation of Eve
in the seventh (cf. fig. 11 ), a depiction that, as previously indicated, points
toward the magnum sacramentum and eternal rest. It might also be said that
among the first seven panels there are six panels dealing with the story of
Creation, while the second panel in the row primarily depicts, as indicated, a
state—that is, the established Kingdom of Heaven, the bedecked bridal chamber,
which is now prepared for the great wedding (cf. fig. 9). In this sense, the first
two panels encompass the entire history of the world from the dawn of time until
its end. At the same time, the overall narration reaches the last and most
important age, whose beginning—with the Incarnation of Christ—fills the final
51 See Heinz Angermeier. Die Reichsreform 1410-1555—Die Staatsproblematik in

Deutschland zwischen Mittelalter und Gegenwart (Munich: Beck. 1984). p. 16. For the
indissoluble connection between reform of the Church and of the empire in the late
medieval period, see. e.g.. Tilman Struve. ‘Utopie und gesellschaftliche Wirklichkeit—
Zur Bedeutung des Friedenskaisers im späten Mittelalter’. Historische Zeitschrift, 225
(1977), pp. 65-95: 'Die enge Verknüpfung von kirchlicher Reform und gesellschaftlicher
Veränderung darf deshalb geradezu als Charakteristikum der Kaiserprophetie in
Deutschland herausgestellt werden- (p. 88).
5‘ Quoted from M. Gerwing, ‘Weitende, Weltzeitalter-, in Lexikon des Mittelalters
(Munich: Artemis—Lexma. 1977-2000). vili. col. 2169.
404 S oren K aspersen
two ‘sections’.” The ‘clutch’ between the days of Creation, which form the
parameters of the chronology of salvation history, and the final, decisive age
takes place through the linkage of the ‘Creation of Eve' and ‘The Annunciation’,
a comparison rich in revelatory, visual and symbolic-allegorical resonances (cf.
fig. 11). At the same time, in an Augustinian sense, it can be said to be a
juxtaposition of the original pattern of human community and the full appearance
of ‘the City of God’ in its complete earthly manifestation through the incarnation
of Jesus.” The creation of Eve depicts humanity in its original condition, turned
towards its Maker and determined for eternal bliss. At the same time, her mutual
bodily connection with Adam serves as the symbol of all symbols of the
harmony and unity that should exist among people, and as a warning against the
human inclination towards waste and strife.55
Thus, the question is what sort of framework can be imagined for such an
Augustinian doctrine of two cities during the latter half of the fourteenth century.
It can generally be asserted that the reception of Aristotle in the thirteenth
century had a fundamental impact on the development of political and political-
theological thinking in the late Middle Ages: for here begins not only ‘die
Vorgeschichte des Begriffs und der Vorstellung der ‘bürgerlichen Gesellschaft’
(societas civilis), sondern aller damit zusammenhängenden Begriffe und deren
Verhältnisbestimmungen zu Glaube, Theologie und Kirche'.56 John B. Morali
calls the two centuries leading to Luther ‘the age of ambiguity’, and stresses how
difficult it is to come to grips with and understand the various positions and
developments during this period.57
There can scarcely be any doubt that the close ties between regnum and
sacerdotium were in the process of dissolution during the late Middle Ages. Yet,
the Hamburg altarpiece directs us to look at things in a historical perspective in
which all events are understood in relation to eternal salvation, and it thereby
sets a limit to Aristotelian-based thinking in which the worldly regime is entirely
55 The altarpiece does not show all six ages but. instead, focuses on the patriarchs
Abraham. Isaac, and Jacob in these three panels between the first and sixth ages.
54See Wolfgang Stiimer. Peccatum und Potestas—Der Sündenfall und die Entstehung
der herrscherlichen Gewalt im mittelalterlichen Staatsdenken (Sigmaringen: Thorbecke.
1987). pp. 67-85 (pp. 68-70. 82-83).
55See De civitate Dei. XII. 22: 'But he created man’s nature from the one original man".
Augustine, Concerning the City of God against the Pagans, trans, by Henry Bettenson.
ed. by David Knowles (Harmondsworth: Penguin, 1972). pp. 502-03. Fora Latin edition,
see Corpus Christianorum. Series Latina. XLVIH (Brepols: Turnhout. 1955). See also ibid..
XU, 28: 'For the human race the union of man and wife’, p. 508.
56 See Ulrich Duchrow, Christenheit und Weltverantwortung—Traditionsgeschichte und
systematische Struktur der Zweireichelehre (Stuttgart: Emst Klett. 1970), p. 398.
57 See John B. Morali. Political Thought in Medieval Times. 2nd edn (London:
Hutchinson, 1960). pp. 119-20.
liberated from such religious goals. The Christian empire is distinguished by its
special sacral dimension from the regnum in general. Within the bounds of hiero-
cratic-curial and theocratic-caesaropapist modes of thinking, the late Middle
Ages was characterized by hard-hitting and consistent theories of the papacy’s
superior position, and particularly in the so-called ‘Augustine School’, such
figures as Aegidius Romanus and Jacobus de Viterbo denied any place at all to
the state in civitas Dei, because of its lack of righteousness, which is why
plenitudo potestatis went to the pope as the representative of the Church.5* But
even within this strongly hierocratic camp, it could not be imagined that the
papacy could do without the emperor or that these two powers were to be
separated. And the idea of a Christian state in which imperium has an important
role in the interplay of these two powers lives on in such thinkers as Dante,
William of Ockham, and Nicolaus of Cusa.
At the same time, at the beginning of the fourteenth century, new parameters
were being set for the relationship between the regnum (imperium) and sacer
dotium after Pope Boniface vm’s defeat by King Philippe iv le Bel, resulting in
the transfer of the papal seat to Avignon in 1309. The final great struggle
between the papacy and the Holy Roman Empire was played out by John xxii
( 1316-34) and Louis iv of Bavaria. The relationship between the two powers
may now be said to have taken on a more pragmatic character, even though the
papacy continued to champion its hierocratic point of view. All of this means
that the doctrine of two states fits into a broad spectrum of ideological view
points, which were coloured by their factual circumstances and whose vitality
can only be understood in light of their interplay with political and social reality.
The sources do not tell us who was responsible for the theological programme
for the high altar of St Petri, but Christian Beutler has suggested Wilhelm
Horborch (c. 1330/32— c. 1384), a theologian and lawyer, who was dean of the
cathedral chapter in Hamburg from 1363.5859 In 1371, he was summoned to Prague
by Charles iv to establish the law faculty at the new university, and in 1375, Pope
Gregory xi made him the youngest of the twelve members of the highest papal
court, the Rota Romana. He won international fame for his collection of papal
rulings. Decisiones Rotae Romanae. He was in Charles iv’s party at a meeting in
Lübeck in 1375. and his brother Bertram, who was also linked to both pope and
emperor during his thirty years as mayor, led the Hamburg delegation at the same
meeting.® In such circles, one could easily surmise an Augustinian world view
58See. e.g.. ibid., pp. 86-89: and A. Zumkeller. "Die Augustinerschule des Mittelalters:
Vertreter und philosophisch-theologische Lehre', Analecta Augustiniana, 27 (1964), pp.
167-262.
59See Chr. Beutler (as in note 36). pp. 66-72.
See Wilhelm Mantels. "Kaiser Karls iv. Hoflager im Lübeck vom 20. bis 30. Oktober
1375'. Hansische Geschichtsblätter (1874). pp. 109-41, reprinted in idem. Beiträge zur
Lübisch-Hansischen Geschichte: Altsgewählte historische Arbeiten (Jena: Gustav Fischer.
activated in the form of a salvation history decoration.

If we look at Emperor Charles iv himself, he was clearly a great admirer of St
Augustine of Hippo, and promoted the Augustinian canons and the Augustinian
anchorites in Bohemia." The first convent for the Augustinian canons was
founded in 1333 by the bishop of Prague, John iv, in the city of Raudnitz on the
Elbe, where the bishop’s castle was located. According to reliable tradition, at its
dedication in 1340, on the day of the Ascension of the Virgin, the canons were
brought there from the canonry at San Pietro in Ciel d'Oro in Pavia, where the
relics of St Augustine were kept. Charles had stayed there for a while in 133!
during his first mission to Italy—for his father. King John vu—and re
established a convent that had been banned by Louis vu of Bavaria.“ The
summoning of the Italian canons must certainly be seen in the context of the
negotiations of King John, Charles, and the bishop with the curia in Avignon to
have the archbishopric established in Prague. St Pietro at Ciel d’Oro was a
northern Italian centre for the reform of the Augustinian order based on Pope
Benedict x ii ’ s constitution Ad decorem ecclesie from 1339. And with these
canons, Raudnitz became a centre for a reform movement, which, among other
things, promoted, a devotio moderna similar to that of the Windesheim
congregation.*61*3 The second major monastery involved in the Raudnitz reform
was Karlshof, which Charles iv established in Prague in 1349 in connection with
the great expansion of the city and which was now to be not only the capital of
Bohemia but also the centre of the Holy Roman Empire. And in 1358 Charles
called the Augustinian canon and reformist preacher Konrad von Waldhausen to
Prague from Vienna to stimulate the new movement of piety.
Thus, the Augustinians made up an important element of the spiritual milieu
that surrounded Charles iv. What impact this had on his understanding of himself
as a Christian ruler may be more difficult to pinpoint.61 However, it may be
1881), pp. 287-323. Concerning the ties between the chapter and the Town Council it is
worth mentioning that St Petri’s, like the three other parish churches in Hamburg, was
incorporated in the chapter and that the Council renewed itself by cooptation every year
on the feast of St Peter's Chair (St Petri ad cathedram), i.e.. 22 February. See J.
Sarnowsky (as in note 42). pp. 99, 103-04.
61 See Joseph Hemmerle, ‘Karl iv. und die Orden', in Kaiser Karl iv. (as in note 42),
pp. 302-04; and Franz Machilek, ‘Die Augustiner-Chorherren in Böhmen und Mähren".
Archiv für Kirchengeschichte von Böhmen-Mähren-Schlesien, 4 ( 1976). pp. 110-21.
“ See Vita Caroli Quarti—Die Autobiographie Karls tv., ed. and trans, by Eugen
Hillenbrand (Stuttgart: Fleischhauer & Spohn. 1979). p. 90. 91.
63 See I. Zibermayr, ‘Zur Geschichte der Raudnitzer Reform’, in Mitteilungen des
Instituts für österreichische Geschichtsforschung, supplementary vol. 11 (Innsbruck.
1929). pp. 323-53; and Johanna Schreiber, ‘Devotio moderna in Böhmen'. Bohemia—
Jahrbuch des Collegium Carolinum. 6 (1965). pp. 93-122.
61 Unfortunately, I have been unable to consult Zdenek Kalista’s account of Charles's
Framing, history’ with salvation 407
ascertained with relative certainty that, to a very high degree, Charles tried to
live up to the ideals of a rex justus. In his two ‘mirrors for princes,’ Majestas
Carolina *65*678and Vita Caroli Quarti,“ he paints a picture of a God-fearing ruler
whose goal is to create peace and security for the empire’s inhabitants through a
just regime, based upon good advisers, in which evil is punished and goodness
rewarded. And he lived up to those duties that the spiritual advisor of his youth,
the abbot Pierre de Fécamp, now Pope Clement vi, enumerated for a Christian
emperor in connection with Charles’ assumption of the Roman throne/as a
Roman king in 1346,"—for example, he attempted to propagate Christian
doctrines, that is, the Church of Rome, by an appeal to the heathen prince of
Lithuania® and by the establishment of a Slavic monastery in Prague.69 He also
fought against heresy7" and provided something of a defence of the Church by
supporting—to the extent possible—papal authority in its difficult circumstances
in Italy.71* Ritually, he displayed his willingness to defend the Gospel by
participating actively in Christmas services in all his trappings of office and
swinging the sword of state three times during the chanting of the Christmas
mass.77
relationship to Augustine in idem. Karel iv.—jelio duchovni tvár (1971), p. 51 ff.

65 See E. Werunsky.'Die Maiestas Karolina’, in Zeitschrift der Savigny-Stifiung für
Rechtsgeschichte. 23. Ger. Abt., 9 (1888), pp. 64-103: R. Schneider. ‘Karls IV.
Auffassung vom Herrscheramt'. Beiträge zur Geschichte des mittelalterlichen deutschen
Königtums. Historische Zeitschrift. Beiheft 2. ed. by Theodor Schieder (Munich: R.
Oldenbourg. 1973), pp. 122-50; and Bernhard Töpfer, ‘Zur Staatsideologie in den
Prooemia der Konstitutionen von Melfi und der Majestas Carolina", Karl IV.—Politik und
Ideologie im 14. Jahrhundert, ed. by Evamaria Engel (Weimar: Hermann Böhlaus
Nachfolger, 1982). pp. 150-57. I have not had access to Majestas Carolina, ed. by H.
Jirecck. Codex juris Bohemici. 2.2 (1870).
“ See Vita Caroli Quarti (as in note 62).
67See Schneider (as in note 65). p. 147. Here. too. he is to reward and honour the good
and drive out the bad through the exercise of justice and. with his power, he is to aid the
oppressed and annihilate tyranni.
68 See K. Conrad, ‘Litauen, der Deutsche Orden und Karl tv. 1352-60’, Zeitschrift fiir
Ostforschung, 21 (1972), 20-42; and H. Grundmann, ‘Das Schreiben Kaiser Karls iv. an
die heidnischen Litauer-Fürsten 1358'. Folia diplomatica, 1 (1971). 89-103.
69 See M. Paulová, 'L'Idée Cyrillo-Méthodienne dans la politique de Charles IV et la
fondation du monastère Slave de Prague’, Byzantinoslavica, 11 (1950), pp. 174-86.
70 See A. Patschovsky, 'Straßburger Beginenverfolgungen im 14. Jahrhundert’,
Deutsches Archiv für Erforschung des Mittelalters, 30 ( 1974), 56-198.
71 See Ludwig Schmugge. ‘Kurie u n d Kirche in der Politik Karls IV.. in Karl tv.—
Staatsmann und Mäzen, ed. by Ferdinand Seibt (Munich: Bayerische Nationalmuseum—
Adalbert-Stifter-Verein. 1978). pp. 73-87. et passim.
77 See Franz Machilek, ‘Privatfrömmigkeit und Staatsfrömmigkeit’, in Karl iv (as in
408 S oren Kaspersen
It is significant in our context that Charles' vision of the Christian ruler

harkens back to the great imperial tradition in his promotion of the cult of
Charlemagne, which was performed primarily by the Augustinian canons at
Karlshof in Prague.71 Moreover, he interlinked his understanding of the Christian
ruler in the introduction to his autobiography with themes and texts on the
Ottonian crown of state,71 designating, for example, the ruler as Christ’s vicar on
earth (‘cuius typum et vices geritis in terris’).**75 In this way, Charles iv anchors
and takes his starting point in the classical period of political Augustinianism—
in the period in which the Holy Roman Emperors understood themselves to be
the leaders of the City of God in its totality.
Certainly, there is no direct linking of the autobiography’s ‘mirror for princes’
to De civitate Dei, but the entire introductory text deals with the two societies,
the God-fearing and the Godless.76—indeed, it could be said to be an elaboration
of or exegesis on Psalms 1,77*79Through a reference to Luke 10.27,7SCharles also
indirectly characterizes the two societies, ‘the heavenly city’ and ‘the earthly
city’, in an Augustinian way: composed respectively of those who love God and
those who love themselves.”
note 42), p. 100; and Hermann Heimpel. ‘Königliche Weihnachtsdienst im späteren

Mittelalter’, Deutsches Archivfür Erforschung des Mittelalters, 39 (1983). pp. 131-206.
75See. e.g., J. Hemmerle (as in note 61 ), p. 302-03.
4 For the following, see Vita Caroli Quarti (as in note 62), pp. 27-30 (Intro.), 72, 73-
76. 77 and ili.
75Corresponds to the plate of Christ enthroned and with the inscription: ‘Per me reges
regnant' (Proverbs 8.15) on the Ottonian crown.
76 ‘Wenn ihr nun in der Furcht Gottes wandelt, wird dies der Anfang eurer Weisheit
sein. So werdet ihr nicht vom Wege abirren, weil der Weg des Herrn gerade ist Die
Gottlosen werden vor euch niedersinken, die Gerechten aber werden frohlocken'.
77Charles specifically refers to this psalm when he admonishingly writes: 'Meidet also
von Kind an jede Sünde. Wandelt vielmehr ohne Makel im Gesetz des Herrn, So seid ihr
wie ein Baum, geplanzt an strömendem wasser. der seine Frucht trägt zur rechten Zeit und
dessen Laub nicht welkt’ [see Psalm 1.3]; see Vita Caroli Quarti (as in note 62). pp. 79
and 81.
w Von besonderer Bedeutung aber ist jenes Gebot, Gott den Herrn zu lieben aus
ganzem Herzen und aus ganzer Seele und den Nächsten wie sich selbst.
79 See De civitate Dei, XV, 1 (as in note 55). In light of all this, it is quite under
standable that there are statements in the autobiography that are reminiscent of the
narration in Master Bertram's altarpiece—most clearly, the quotation below, which
corresponds to the introduction to the lower and upper zone, that is respectively, the story
of Cain and Abel and that of the Fall of Satan and the Rebel Angels followed by the
celestial aula: "Neidet einander nichts, sondem habt Liebe zueinander, denn Neid gebiert
Haß, Wer haßt, wird nicht beliebt und richtet sich in seiner Raserei selbst zugrunde. . .
Will euer Herz sich erheben, so erniedrigt euch. . . . Der Stolz ist dem Schöpfer verhaßt
The whole of Charles iv’s religiosity—his name policy, his promotion of the
cults of saints and relics, coupled with his spectacular staging—has most often
been viewed as a particularly politically motivated piety—Rudolf Chadraba calls
it a devotio antiqua, which has much more to do with ‘mit seiner welthistorischen,
von ihm geglaubten Sendung als mit privat-religiösen und ästhetischen Interessen
im modernen Sinne’.® It would also be a mistake to underestimate this view,
since it is such an obvious and important part of Charles’ talent as a Christian
ruler.8081 However, Heinrich Neureither seems correct in his emphasis on Charles
iv’s personal piety, viewing his particular brand of religiosity as an intersection
between his inward piety and his political understanding of the significance of
the cult of the Christian ruler.8'2*For example, even as a child in the court of
Charles the Fair in Paris, he read ’die marianischen Antiphonen des Stundenge
bets und las sie, als ich ihren Sinn einigermaßen verstand, in meiner Kindheit
von Tag zu Tag lieber’,8586and throughout the rest of his life, he regularly con
tinued to observe canonical hours as a priest.84As a six-year-old, he experienced
a sort of religious awakening as a result of a sermon by Pierre de Fécamp,85and,
later in his life he also had ‘visions’.811 Moreover, his theological training and
knowledge of the Bible was far above the norm for a secular ruler.87
In this light, it is telling that Charles iv exhibited an intense, inner piety in
taking the sacrament of holy communion.88 For in this sacrament, there is an
und allen, die guten Sinnes sind. . . . Gott aber wird am Ende den Stolz brechen, er wird
die Mächtigen vom Throne stürzen und die Niedrigen aus dem Staub erheben . . .’. Vita
Caroli Quarti (as in note 62). p. 79 (cf. Luke 1.52, and I Samuel 2.80).
80 See Rudolf Chadraba. ‘Kaiser Karls IV. devotio antiqua’, Mediaevalia Bohémica, 1
(1969). pp. 51-68.
81 See Fr. Machilek (as in note 72). pp. 87-101,441-43; R. Schneider (as in note 78),
esp. pp. 126-33: and Jirí Spevácek. ‘Frömmigkeit und Kirchentreue als Instrumente der
politischen Ideologie Karls iv’, in Karl /V.—Politik (as in note 65), pp. 158-70.
82See H. Neureither. Das Bild Kaiser Karls IV. in der Zeitgenössischen Französischen
Geschichtsschreibung (diss. Heidelberg University. 1964). pp. 18-203.
85See Vita Caroli Quarti (as in note 62), p. 83.
84 See Johan von Jenstein's funeral sermon for Charles iv—Fontes Rerum
Bohemicarum, ill (Prag: Nákladem Nadáni Frantiska Palackého, 1882), p. 429: 'nam horas
eciam suas canonicas, sicut unus sacerdos dicebat'.
85See Vita Caroli Quarti (as in note 62), pp. 43, 86.
86See Fr. Machilek (as in note 72), pp. 88-89.
87See ibid., pp. 87-88.
88 In his funeral sermon for Charles iv, Johan von Jenstein enumerated seven points
that demonstrate the emperor to be beatus: ‘Quinto fortissime plorabat, cum corpus
dominicum sumebat et alios ex devotione adducens': see Fontes Rerum Bohemicarum (as
in note 84). p. 429.
410 Soren K aspersi-n
important intersection between personal piety and institutionalizing liturgy. By

receiving the body of Christ, the believer becomes incorporated into the Church,
understood as the mystical body of Christ, the inward remembrance of the
Redeemer and His acts of redemption fused with the sacramentally established
institution of salvation.
Perhaps this eucharistie piety was also Charles’ way of accommodating the
most radical of the papal apologists, who, for example, re-interpreted Romans 13.1
as reading that such authority ‘as comes from God is ordained’ (‘que autem sunt a
deo, ordinata sunt’), that is, more explicit a hierarchical interpretation in which
secular power does not derive immediately from God but is only conveyed through
the spiritual powers. In De ecclesiastica potestate, Aegidius Romanus contributed
the view that only through the sacraments of the Church could a secular ruler be
free from original and actual sin, that is, only by subordination to the highest power
[i.e., the Church] could one become worthy of the office, which de facto derived
from God.89 But through his intense participation in the sacrament of Holy
Communion, Charles might well have believed that a good Christian ruler could
stand in direct relation to the King of Kings and, in this way. make the more subtle
arguments of superiority and subordination uninteresting.
Undoubtedly, the most interesting thing for us is that this aspect of the
Eucharist appears directly in Charles tv’s understanding of the Christian ruler.
His autobiography opens with a discourse on the two forms of temporal life
related to the sacramental nourishment: bodily food is transitory and leads to
death. But we learn from the Saviour: ‘Wer mein Fleisch ißt und mein blut trinkt,
der bleibt in mir und ich in ihm.’ Further:
Die von solcher geistigen Speise leben, bleiben in Ewigkeit Wer aber jene geistige
Speise ißt. von der die Seele lebt, muß dieser Mensch nicht in seiner Seele ein
Verlangen empfinden, diese Speise begierig aufzunehmen und in hingebender Liebe
zu umfangen, Doch viele essen die geistige Speise ohne Verlangen und Begier und
speien sie aus ihrem Herzen aus. Weh ihnen!
‘They eat their way to punishment and damnation. Instead, one should seek
spiritual nourishment, the bread of heaven, which is both ‘flesh’ and ‘word’, the
Saviour continues, and:
Durch jenes Brot hat uns Christus selbst das neue und ewige Testament bekräftigt.
Jenes Brot stärke eure Herzen und Seelen in heiligher Zuneigung und Liebe zu
ihm. damit ihr stark genug seid, das Zeitliche zu durchwandern, ohne das ewige
Reich zu verlieren. Amen.90
This discussion of Jesus’ body as sacramental bread and of eating one's way
to salvation or damnation brings Bertram’s altarpiece to mind. The panels can be
89See U. Duchrow (as in note 56). pp. 405-06.

90Four quotations all from Vila Caroli Quarti (as in note 62). pp. 67. 69. 71.73.
said to make the same connection between the story of the two societies and the
sacrament of holy communion that Charles iv makes in his autobiography
between heavenly and worldly nourishment on one hand and of the god-fearing
and the godless on the other. This is far from saying that the panels must be
understood as a simple illustration of Charles iv’s conception of the Christian
ruler. Rather, it implies that they were created in an environment in which
Charles’ idea of the regimen christianum was one among a wide range of
conceptions, more or less related to one another, that is, there was a spectrum of
possible frames within which the altarpiece may delineate a certain configuration
of ideas. What is essential is not that the statement of the altar-piece can be
compared with a specific, written source but the kind of conceptions the painted
program in its interaction with the general ideas and prevailing mentalities is
supposed to reflect and evoke. The hypothesis of the politico-theological
meaning of the high altar may be supported by other sources in future
investigations, but, for the moment, Master Bertram’s panels, like the wall
paintings in Keldby Church, stand out as a potential testimony about certain
opinions and expectations, as a possible source for insights which may be
otherwise unaccessible.
We have now had a closer look at two examples of a painted salvation history
and have tried to substantiate that they were closely related to specific historical
events. In both cases, the use of pictures as historical sources is placed in the
context of intellectual history. But to understand history in a frame of religious
salvation was at the same time quite customary, one might say a fundamental
aspect of the medieval outlook, the medieval mentality. There are no clear and
distinct boundaries between intellectual history and the history of mentalities.
For example, the politico-theological meaning of the high altar in St Petri may
include many specific ideas and concepts. However, it also is rooted in views
that were common in the Middle Ages, in conditions of life that the individual
internalized more or less unconsciously or to which, at least, the individual was
forced to adapt in one way or another, if not to revolt against them. The
Hamburg citizen, the merchant or the shopkeeper, the artisan or the day labourer
may not speak of an Augustinian world view or be acquainted with the different
positions and arguments of the curial and imperial camps, but they would most
likely have been familiar with the idea of the two cities, of the opposition
between those who love God and those who love only themselves and worldly
things, and, just as certainly, they would have considered both the Church and
the empire to be important communities or institutions for attaining salvation.
I think it is in the context of the history of mentalities and intellectual history
or ideology that pictures have their greatest importance as historical sources. It is
true that this study, unlike the Annales school, is concerned with specific,
historical episodes, with the narrative of events, but it also concerns important
modes of thought and belief systems. The course of events is interpreted within
these greater structures and these corroborate the readings. In Keldby Church.
412 S0REN K aspersen
the exegesis, of course, is very involved with and dependent on the identification
of the official royal act. but it is still grounded in general structures. In St Petri’s,
Hamburg, there are no immediate links between the altar-piece and the political
history of the decade. The hypothesis can be supported and made probable only
by references to certain modes of thought. In relation to the actual events, one
may even point out that Charles tv might have been dead before the altar-piece
was begun or, at least, that it was not finished until five years after his death, in
1383. Therefore, a closer look at the acts of King Véncelas may seem natural.
But according to the basic premise of the present interpretation, it was Charles iv
who formed the framework. One may ask whether Véncelas met the expectations
of Hamburg,1'1 or whether the somewhat sombre issue of the salvation story on
the altar-piece reflects anxiety about the future.9 192 But such minor, socio-
psychological fluctuations are difficult to determine with any certainty.
The important thing is that pictures constitute their own kind of historical
source. They make statements according to their own linguistics—for example.
91 The question is not followed up here. In passing one may notice that an important
part of the imperial politics, the King’s Peace (der Landfriede), which Charles iv had
extended to the northern region of Germany, soon crumbled under Vencelas’s
government; see Heinz Angermeier, Königtum und Landfriede im deutschen
Spätmittelalter (Munich: Beck, 1966), pp. 266-317. In general. King Véncelas was
severely critized as a regent by all contemporary and later chroniclers; see Klaus
Schreiner, ‘Correctio principis—Gedankliche Begründung und geschichtliche Praxis
.spätmittelalterlicher Herrscherkritik', in Mentalitäten im Mittelalter—Methodische und
inhaltliche Probleme, ed. by Frantisek Graus (Sigmaringen: Thorbecke. 1987), pp. 224-
31. For one thing. Véncelas was accused of inebrity (see ibid., p. 224) and there may be a
connection between this and the unusual setting off of Ham as a drunkard in the panel
with The Building of the Ark. But it is a somewhat daring assumption since the surprising
iconography merely may arise from an allegorical understanding of the salvation story on
a religious level. Finally, it can be mentioned that St Véncelas does not appear, at least
today, among the many sculptured saints of the shrine; see Goldgrund und Himmels
licht—Die Kunst des Mittelalters in Hamburg (exhib. cat.), ed. by Uwe M. Schneede
(Hamburg: Dölling und Galitz, 1999). pp. 101-11. inch ill.
92 The Statement of Bertram’s cycle on the position of the secular ruler in salvation
history is ambiguous. Certainly, we see the three holy kings worshipping the Christ Child
just below the depiction of the Fall, which made secular authority necessary: see W.
Stumer (as in note 54), but in the following and final ‘section’, secular power is presented
in the role of tyrant—in the form of Herod, and the cruelty of his action intrudes itself on
to the Holy Family, with the mourning morthers continuing ‘behind’ the framework of the
scene. Furthermore, it may be ascertained generally that the altarpicce's program of
salvation history is being used not to tell the story of the Chosen People’s leaders or to
depict rulers under the Old Covenant—Moses. Joshua. David, Solomon, etc.—but. rather,
to tell the story of the patriarchs, who. according to Augustine, designate the City of God
before the coming of Christ, together with the prophets; see ibid., pp. 81-82, and De
civitate Dei, xvii, 16 (as in note 55).
programmatic and iconographie rules—and they have their own syntactic struc
tures determined by space and surface, colour and line, light and shadow, etc.—
their own rhetoric of invention and composition. These latter aspects of figura
tive language are easily utilized in an investigation of mentalities and may even
be relevant in the context of a more specific intellectual history. For example,
conceptions of the regimen christianum, of the regnum and its place in salvation
history, have an organologie aspect, which is interesting in relation to corporeal
ity in Master Bertram’s panels.
It was a slock in trade in the Middle Ages to compare the relationship
between regnum and sacerdotium and the structure and the ideal function of
society with the human body. In the High and Late Middle Ages, a proper
organologie conception of the society or ‘State' developed.'' The different parts
of the organic society in the early period were distinctly subordinated to one
head; regnum and sacerdotium co-existed within Ecclesia; the ‘State' was
conceived as a function of the Church as the mystical body of Christ; and the
relationship between the secular and the spiritual was typically compared with
the mutual dependency of body and soul. However, the emphasis shifted in the
later period: the earthly institutions of Church and ‘State' broke away from each
other and became self-contained bodies within the corpus Christi. Salvation was
no longer, as a matter of course, the immediate or determinative objective of the
regnum. The superiority of the head was now relativized into a large number of
highly differentiated members, and their perfection and harmonious interaction
became the central theme of the ‘state’.
Within the figurative arts, this process may be compared to the general
development of the depiction of the human body from a general model, stressing
the head and the importance of the soul through prominent eyes and eloquent
hands, to another model, stressing corporeality and giving other parts of the body
greater importance and equality. One may argue, then, that the dichotomy
between a distinct corporeality and the incorporeal gold ground in Master
Bertram’s panels reflects and comments on the tensions in the late Middle Ages
between organologie and secular-oriented conceptions of a realm of estates, on
one hand, and. on the other, an understanding of the regnum, in casu imperium
christianorum, as an integral part and function of the corpus Christi, as a
community or corporation determined by its eschatological goals.
I think this kind of analogy can be worked out in greater detail with advantage
and become thereby a more subtle tool for the analysis of the specific language
of pictures. I also think that such analogies can be corroborated by a theory that
makes the structure of human society a determining factor not only for the
organologie conception of the ‘state' but also for the scientific understanding of
*' See Tilman Struve. Die Entwicklung der organologischen Staatsauffassung im

Mittelalter. Monographien zur Geschichte des Mittelalters, 16 (Stuttgart: Anton
Hiersemann, 1978).
414 S0REN K a SPERSEN
the human body and its functioning and for the mind’s apprehension of space, of
its continuum and its relation to solid bodies. Here these final remarks merely
provide a hint of the potentials of figurative language, to suggest a final aspect of
the use of pictures as historical sources in their own right.
O n the epistemology o f images
LENA LIEPE
Introduction
n the frontispiece of a 1452 French translation of Aegidius Romanus’s
O De regimine principian (Government of Princes),' we see a detailed and

colourful depiction of the presentation of the volume that contains the
illumination to the patron who commissioned it (fig. 1). The translator himself,
Jean Wauquelin, kneels in front of the duke of Burgundy, Philippe le Bon, and
the court. He holds out the open book in both hands. This miniature will serve as
my point of departure in this investigation of images as source material. The
main question will be: how and on what—or on whose—premises, should
images be interpreted as historical sources? 1 will promote an art-historical claim
that images can and should be used as sources in their own right, according to
the distinctive nature of their visual characteristics: that the main potential of
images as historical sources lies precisely in their visual formation.
Everything that has survived from past times can be defined as historical
source material if it is used functionally, that is, to answer questions about what
once happened and how things once were. When historians treat images as
sources, they naturally handle them similarly to other kinds of sources. The
images are viewed instrumentally, as something that is used to reveal informa
tion about something else. This means that the main interest of the historian does
1Brussels, Bibliothèque Royale Albert I", ms. 9043. For a detailed description of the
manuscript, see Frédéric Lyna, Les principaux manuscrits ci peintures de la Bibliothèque
Royale de Belgique tome m. ed. by Christiane Pantens (Brussels: Bibliothèque Royale
Albert F, 1989). pp. 279-83.
4 1 6 L ena L iepf.
feo» neftfpcaaf ¡ih»fai! teuorftMcttr twwrtwom

IfatmiuntfiY mattana» tr ft»! m a fim a tu ttt'
faiwtietrtnf fv fo p x fr te(irme.-. Oían eiefd m
¡faillît Alette feufnatr fafin ta u t fo p tfm q im
!(|j¡;i)ttO!i/c!tytatc h r put rourreimitats ueßnepnr
lyftfiyrÿîittîr k fíta C c h m eefraiuxe)U((ßc6'
Yiirtatrt tf f»tr ftoiw ma n c h tm trm trta n rC m u «mimt
famtauiifftyriMi' fattnurkfco; fattfnt tffj i ttfaictynttr fo mtt
rcj'tytwtcrjWcftföfctamffr f ;mte¿mit tfM fhtnt/cuflm nt neu
fonte te fiats ftnimte/amrauû feyor\>tt<rwcifmc«ntm mais
F/g. 1. Unknown artist: frontispiece. Aegidius Romanus, De regimine principimi.

translation to French by Jean Wauquelin, copy from 1452 by Jacques Pilavaine.
Brussels, Bibliothèque Royale Albert F', ms. 9043 fol. 2r. © Bibliothèque Royale
Albert f . Brussels
On lhe epistemology o f images 417
focuses not on the image as such but on the image as a means of understanding
the past. Art historians, on the other hand, often work the other way round; our
aim is to reconstruct an historical context in which to situate the image, to
understand it better. This approach appears to me as problematic, because of its
tendency to relegate the image to a secondary position while the historical
context becomes primary. The interpretation of an image becomes synonymous
with the mapping of the historical situation; the economic, political, social,
cultural, or intellectual circumstances that are chosen as relevant for the
understanding of the image. When this historical context is completed, the image
is brought in as a separate entity, as something that passively reflects the cultural
processes of its epoch: ‘the expectation is that context will control the text (the
work of art); once their opposition is staked out the terms taken together are
made to form an economy in which context is active and the text or work of art
passive’.2
This will be one of my main points. Interpretations that aim at situating an
image in its historical context are too seldom centred on the image as such,
regarded as having a cognitive capacity of its own based on its distinctive char
acter as an item of visual communication. An image is not just something that is
understood by translating it verbally as completely as possible. An image also
has its own visual language, its own rhetoric. Admittedly, that aspect of the
meaning of the image is accessible to art historical discourse only verbally,
through the use of words, but this does not mean that the image should be seen
as nothing more than an illustration of an historical situation that has been
established a priori.
In the first part of this chapter. I will discuss the conditions under which an
image can be interpreted as an historical testimony: the extent to which images
have to be subject to the same critical examination as, for instance, a text, with
regard to the where, when, by whom, and, above all, why they have been pro
duced. I then will concentrate on my main issue, the inherent epistemological
value of images. Presupposing that images are powerful carriers of meaning
which participate in the shaping of a wider social and cultural context, and that
the main potential of images as historical sources lies precisely in their visual
formation, I will suggest possible readings of images as visual configurations.
The H istorian a n d the Im age as Source M aterial
If we look at the miniature from De regimine principian as an historical

document, what kind of information does it yield? If I were to interpret it as a
snapshot of an actual historical event, as a peephole through the parchment to
2 Norman Bryson, ‘Art in Context’, in Studies in Historical Change, ed. by Ralph

Cohen (Charlottesville: University Press of Virginia, 1992), pp. 18-42 (p. 19).
4 18 L ena L i epe
the scene that is depicted, it would tell me that at a certain moment a clerical
person presented a volume to a secular person in the presence of several other
people, both clerical and secular, but all of them male. This took place in a
stately room in a building situated at some distance from a town and, considering
the stars in the sky, it happened in the evening or early morning. There was also
a dog present, albeit sleeping.
Now, this is a reading that is rather meaningless. From the perspective of the
historian, the information that the miniature itself can give us is minimal. To be
able to situate it in time (middle of the fifteenth century), to locate it geographi
cally (to Burgundy or its territories), and to define its historical context, it is
necessary to turn to external sources. Even the identification of the persons as
clerical or secular requires the use of a general cultural knowledge that
transcends the information content of the painting. In the classical interpretative
scheme of Erwin Panofsky, this means going from the first, pre-iconographical
level to the second level of interpretation, the iconographical level.3*5
More important, however, is the fact that the miniature does not give us any
information at all about past events, since it is not a snapshot or a peephole back
into history. It cannot be used as evidence of how, where and in the presence of
whom the volume was presented. The only thing that this painting tells us is that
somebody once had reason to depict, or arrange to be depicted, a presentation of
a volume in this way: but how it actually happened, if at all, is a totally different
thing. Images, like texts, are historical products manufactured for certain
purposes. This means that the instrumental use of images as source material
necessitates the analysis of the image itself as an historically defined product.
This is an ordinary, old-fashioned critical approach to sources: when, where, by
whom, and why was this miniature produced? The answers to these questions
determine the instrumental value of the picture.
A brief summary of the historical setting of the illumination would look as
follows.'1The text. De regimine principian, was translated in 1450 by the author,
translator, and scribe Jean Wauquelin (f 1452) for Philippe le Bon (1396-1467.
3 Erwin Panofsky, ‘Introductory’ in Studies in Iconology. Humanistic Themes in the

Art of the Renaissance (Oxford: Oxford University Press, 1939), pp. 3-31.
J My account is based on Alain Arnould and Jean Michel Massing. Splendours of
Flanders. Late Medieval Art in Cambridge Collections (Cambridge: Cambridge Univer
sity Press, 1993); Otto Cartellieri. The Court of Burgundy. Studies in the History of
Civilization (New York: Haskell House. 1925); Georges Dogaer. Flemish Miniature
Painting in the 15th and 16th Centuries (Amsterdam: B. M. Israël. 1987): Georges
Doutrepont, La littérature française à la cour des Ducs de Bourgogne (Paris 1909:
Geneva: Slatkine reprints, 1970); La librairie de Philippe le Bon. Exposition organisée à
l’occasion du 500e anniversaire de la mort du Duc, ed. by George Dogaer and
Marguerite Debae (Brussels: Bibliothèque Royale Albert T. 1967): Richard Vaughan.
Philip the Good. The Apogee of Burgundy (London: Longmans, 1970).
On the epistemology o f images 419
duke of Burgundy 1419-67). Philippe le Bon is among the most prominent of

the Burgundian rulers of the house of Valois. Unlike his predecessors, he
maintained peace with the French king and refrained from threatening French
territory, expanding his duchy in other directions through purchases, alliances
and conquests: Namur (1421), Hainaut (1427). Brabant (1430), Holland and
Zeeland (1424-33) and Luxembourg (1433). Philippe le Bon manifested his rule
in an extravagant court life, part of which was his patronage of art and culture on
the highest level: artists such as Jan van Eyck and Rogier van der Weyden were
among those engaged by the duke. The rule of Philippe le Bon has been
described as ‘l’âge d'or de la littérature bourguignonne',5 and his collection of
manuscripts was one of the largest and finest of the Middle Ages; it had been
founded by his grandfather, Philippe le Hardi who ruled 1363-1404, but was
much enlarged by the grandson. When Philippe le Bon died in 1467, it
comprised close to nine hundred volumes. The duke spent considerable amounts
on the acquisition of books through purchases and commissions; he regularly
gave commissions to translators, copyists, illuminators and bookbinders and
sometimes supervised their work personally. As a result of the patronage of
Philippe le Bon, Burgundy in the later half of the fifteenth century became one
of the most important art centres of Europe, second only to Paris.
There is no reason to doubt the genuine and dedicated enthusiasm of Philippe
le Bon for art and literature, but through his patronage he also established and
confirmed his political power and status, not least with the help of illuminated
manuscripts. The collection of art and literature was a political and social means
of influence among the highest levels of the Burgundian society: ‘art functioned
as a passport to the ruling elite’.6 The example of the great-uncle of Philippe le
Bon. the French king, Charles V, who reigned 1364-80, clearly shows the
potential of the art of illumination as ideological propaganda. The king made
strategic use of words and images to strengthen the position of the French
monarchy, by commissioning translations of carefully chosen texts on royal
power, furnished with illuminations showing himself in different contexts, but
primarily as ‘le roi sage’.7
The text in question here. De regimine principian, was written in 1277-79 in
Paris by Aegidius Romanus (1243/47-1316) as a political and moral guide for
Philip the Fair, future king of France (reigned 1285-1314). The text was transla
ted into many languages (in French as early as 1282) and was read by rulers all
over Europe well into the sixteenth century. It became one of the most wide
spread mirrors of princes of the Middle Ages and one of the most frequently
5Doutrepont (as in note 4), p. 480.

“Arnould and Massing (as in note 4), p. 143.
7Jonathan J. G. Alexander. Medieval Illuminators and Their Methods of Work (New
Haven: Yale University Press, 1992). p. 143.
4 2 0 L ena L iepe
read texts of the period as a whole.8*10Today it is preserved in about 250 copies in

eight different languages. The translator of this particular edition, Jean Wauque-
lin, was Philippe le Bon’s principal scribe. In 1441, he set up his atelier for the
production of manuscripts in Mons. Wauquelin’s Le Gouvernement de Princes
was by no means the only didactic text available to Philippe le Bon: George
Doutrepoint. who has studied the literature of four generations of Burgundian
rulers, enumerates a considerable number of mirrors of princes that belonged to
the ducal library, among them several copies of De regimine principum.'1
This manuscript from 1452 is the only extant copy of the 1450 translation
made by Jean Wauquelin. The scribe responsible for this copy is Jacques
Pilavaine, who took over Wauquelin’s atelier on the death of the latter.11’
Accounts testify that Pilavaine did the calligraphy and decorative ornamentation,
but the name of the artisan who produced the presentation picture—the only
pictorial scene in the manuscript—is not known." This kind of division of labour
in the making of a manuscript was not unusual: in the middle of the fifteenth
century the production of illuminated texts was a highly specialized craft, and
the best illuminators were responsible for the most important pages and the
pictorial scenes, while assistants took care of the ornamentation and decorative
frames and copyists the calligraphy. The different artisans were seen as similar
to other craftsmen and sometimes formed part of the ducal household together
with other servants, as ‘valets de chambre’. Philippe le Bon, however, did not
keep a regular scriptorium of his own for the manufacture of illuminated
manuscripts. He gave commissions to ateliers as far apart as Paris, Dijon, Mons,
Lille, Brussels, and Bruges, and the volumes were sent among scribe, illumina
tor, and copyist to be edited, illuminated, and engrossed.12
All this is circumstantial information not directly related to the image in ques
tion. In an instrumental interpretation of the kind an historian would find
8Wilhelm Berges. Die Fürstenspiegel des hohen und späten Mittelalters. Schriften des
Reichinstituts für ältere deutsche Geschichtskunde. Monumenta Germaniae historica 2
(Stuttgart: Hiersemann, 1938), p. 211.
’ Doutrepont (as in note 4), pp. 290-319. 488: Dogaer and Debae (as in note 4), pp.
72-73.
10Dictionnaire des lettres françaises. Le Moyen Age. Encyclopédies d'aujourd’hui
(Paris: Fayard. 1992), pp. 543, 713.
" Dogaer (as in note 4), p. 61 : Dogaer and Debae (as in note 4). p. 73; Lyna (as in note
1), p. 280. He has been given the anonymous name Master of the Gent Privileges, and
several more paintings in illuminated manuscripts have been attributed to him; see
Friedrich Winkler. Die flämische Buchmalerei des xv. und xvi. Jahrhunderts. Künstler
und Werke von den Brüdern van Eyck bis zu Simon Bening (Amsterdam: B. M. Israël.
1978). p. 31.
12Cartellieri (as in note 4). p. 167: Doutrepont (as in note 4), pp. 457-58.
^ « ^ L ^ ^ g jL p ttU u n : ocfaipnA <£^riftjuf a « f h m fh tn tx m ap non

tMfltU·X.·ßIÍnmC
éítfnc.poutféc e ..e tm a ir e a ft»t,' »
^ ■ f t appetì ímioic «néfmfnr.éfSîKfeff fõififs it
||S - t cm*tfiofSe nouiKffce ltotn; 1«Vi (teaftesH Ktxnwtaxaiffint.ê^sj^^j^
^ a w n o n ir fjïiitotf és p n ro uor,rtíícjfm<ûfftonifínèiíõíuc
y^rtuR . *offr <]ucfits on «soté ése ififFétnar »or¡tvffícéot.firc x t v t f
g^djoffe d inínate .xndìtOTOfmoir /fr->, nutffonf<iinnftino>t^y 4 t f : ^ ^ Cm^^:
tr.cifr. on és6 (yiuftce «f stoSfí»,•jT; par lòttrtltcéféttalW éc
/;5j);of|<ie <t siipnfó: éce noßfire *A ttoiitnf.éf (òtnnnfif*éf 6.-*ufatff
* fiomtnfo poasce írcwtfnrre Jxe éf (cinSour.ibntréc
I lìrtuike « m>8('<&poavnnone ft
funuf6.ttu6.po«oommfe apone
y f i fottcnécmcnrfjTcucriíi fiotmfur
toio.éf Soaitwunicjfíjfcfjníf fio?o t Ok ¿
nau.éf fioflt»téf*éf «.Ctttòf fféf
•namar.àvmiuteéapmrfntyííf.'k'Çj^^Ç'í
~^ÈMsÎ (ûflur"î tstfe p ie «cnrYscpsécrp; 2â>futn«m écfvtsf^s p.i ?-.•■■ rOvinrÄ1^ ^ ^ V
^ ® ? ÿ :vfiit;-fnf ï dmfœisjtrfjt tsfaoitiïf
Fig. 2. Unknown artist: frontispiece, Jacques de Guise, Annales Hannoniae,

translation into French by Jean Wauquelin, copy front 1448 by Wauquelin and
Jacotin du Bois. Brussels, Bibliothèque Royale Albert F. ms. 9242 fol. / ’.
© Bibliothèque Royale Albert F, Brussels
4 2 2 L ena L iepe
interesting, the evidence of the miniature itself has to be supplemented with

external information to be of any use; after all, it is not the image itself that
interests the historian. The miniature has a confirming function: it affirms what
we already know about Philippe le Bon as a commissioner of manuscripts, or at
least does not contradict it. The illumination informs us that Philippe wished to
demonstrate his role as collector and patron visually. Actually, the painting takes
its design from a slightly older presentation picture: a well-known miniature in a
translation of Annales Hannoniae, an historical chronicle written in the
fourteenth century by the Franciscan Jacques de Guise (t 1399) and translated
by Wauquelin for Philippe le Bon in 1447—48 (fig. 2).1314This painting has been
used as the model for several presentation pictures in manuscripts ordered by the
duke, and the miniature in De regimine principian follows it closely, with certain
variations: here, the group of people to the right of Philippe le Bon consists of
seven, not eight, persons, as in the Annales Hannoniae; the boy immediately to
the right of the duke (his son and future duke, Charles, who reigned 1467-77) is
in the later painting shown as being a few years older; the colouring is partly
different; the volume in the hands of Wauquelin is open and not closed; and the
illuminator of Gouvernement de princes has added a view through the
windows.w
Although the more recent miniature is thus made according to a model from
another text, it brings with it some information to its new surroundings. There is,
for instance, no reason to doubt that the translator, according to the painting, was
a cleric (in the preface he calls himself the ‘clerq et serviteur’ of Philippe le
Bon), although there is a record of a widow making demands for unpaid
remunerations after the death of Wauquelin.15 The obvious effort to give indi
viduality to the features of Philippe le Bon corresponds approximately with other
contemporary depictions and descriptions of him as tall and lean with a bony
face, protruding nose, and wide mouth.16 One can also assume that the painting
shows contemporary dress fashion more or less accurately. However, this last
statement has to be made with a certain reservation: depending on the use of
models, or on pictorial conventions, or for other reasons, images can show
people in old-fashioned or archaic dress. Thus, the man immediately to the left
'■'Brussels, Bibliothèque Royale Albert I", ms. 9242 fol. 1'.

14On the Annales Hannoniae manuscript, see Dogaer (as in note 4). p. 63; Dogaer and
Debae (as in note 4), pp. 125-26; Lyna (as in note 1), p. 281 ; Rogier van der Weyden—
Rogier de le Pasture. Official Painter to the City of Brussels. Portrait Painter of the
Burgundian Court (Brussels: Maison du Roi, 1979). p. 172.
15Doutrepont (as in note 4), p. 22-26.
16 Vaughan (as in note 4), pp. 127-28; Francis Haskell, History and Its Images. Art
and the Interpretation of the Past (New Haven: Yale University Press. 1993). pp. 491 —
92.
of Philippe le Bon, chancellor Nicholas Rolin, following identifications made of

the individuals in the original miniature, is conservatively dressed according to
1430s fashion, rather than to that of the late 1440s.'7
Furthermore, the miniature informs us that the presentation of a manuscript
was an occasion of a certain formality that could take place in the presence of a
larger company—or at least that this was the way it could be visually repre
sented. Also, one can note not only what the picture shows but also what it does
not show: obviously dogs were not considered out of place when manuscripts
were presented, but there are no women present. There certainly were women at
the Burgundian court, and records of female commissioners and owners of
manuscripts during the Middle Ages tell us that women were not per se excluded
from the world of books and reading.1718A possible explanation would be that the
text in question is a mirror of princes, a political and didactic text on the art of
ruling, that belonged to a sphere in which women had no place. To confirm this
assumption one would have to compare this miniature with other presentation
pictures. A comparison to some other manuscripts from the Burgundian court
shows that the design of the presentation pictures follow a general scheme, but
for all we know this may be due rather to the fact that the pictures adhere to a
conventionalized formula than to the fact that this was how presentations were
actually performed at the court.19
Thus, in using images as historical sources, one needs to take into account the
degree of representativity of the individual picture. Does this miniature represent
a unique instance, or does it have a more general validity? The answer is simple:
without comparisons with contemporary images from the same historical setting
it is impossible to decide if a specific picture is unique or in accordance with a
prevailing visual tradition. It is vital to be careful in attributing authority to an
17On Burgundian fashion and the dress habits of Philippe le Bon. see Margaret Scott.
Late Gothic Europe, 1400-1500. The History of Dress Series, ed. by Aileen Ribeiro
(London: Mills & Boon. 1980).
18 John Bartier, ‘Art Patronage in Van der Weyden's Time’, in Rogier van der
Weyden—Rogier de le Pasture (as in note 14), pp. 24-35 (p. 27); Susan G. Bell,
‘Medieval Women Book-Owners: Arbiters of Lay Piety and the Ambassadors of
Culture’, Signs. Journal of Women in Culture and Society 1 (1982), 742-68.
19 See Dogaer and Debae (as in note 4), pi. 1. 42, 43. 47, 49. 61. In a presentation
picture from 1403, showing Christine de Pisan presenting her Le livre de chemin de
longue étude to Charles vi, the author herself is the only woman present (ibid., pi. 23).
Cf. Dogaer and Debae on the standard design of manuscripts ordered by Philippe le Bon:
‘Au début du manuscrit se trouve un prologue dans lequel l’auteur ou le traducteur dédie
son œuvre au Duc. Ce prologue est généralement accompagné d’une belle miniature où
l’on voit Philippe le Bon en costume d’apparat, entouré de dignitaires de la Cour,
recevoir un manuscrit des mains de l’auteur ou du traducteur. Les armoiries, la devise et
l’emblème du Duc sont parfois peints dans cette miniature frontispice’ (ibid., p. 5).
4 2 4 L ena L iepe
individual image that lacks comparable counterparts. The issue of repre-

sentativity is central, but far too seldom critically evaluated, when dealing with
fragmentary source material such as that of medieval imagery. In this case, when
the relevant comparable material consists of illuminations from the same milieu
and derived partly from the same model, it is necessary to take into account the
possibility that, rather than reproducing actual circumstances, the visual
formation depends on a conventionalized formula.
To summarize this part of my argument: what I have tried to do so far is to
indicate some of the conditions for the use of images as historical sources. An
image, as are all other artefacts, is a product of what Michael Baxandall terms a
‘situational intention': at a certain moment, somebody found reason to make it.
or have it made.2” The image is the result of certain choices made in a certain
historical situation, and this insight is fundamental for the use of it as a source.
Actually, an image cannot tell us very much about a specific historical event
other than that somebody has seen occasion to depict this event in a certain way.
In addition, I have pointed to the necessary awareness of the fact that pictures
follow formalized conventions in their representation. This assertion is valid for
the imagery of every historical period, not solely for the late Middle Ages and
for all types of pictures: portraits and ‘realistic' genre paintings, as well as
glorifying historical paintings or allegorical motifs. An image can never be
trusted as a testimony of anything other than the intention of its instigator.
The A rt H istorian a n d the Im age as Source M aterial
So far, I have focused on an instrumental use of images. As an art historian,

however, I am more concerned with the question of the inherent epistemological
value of images. In a post-structuralist and materialist theory of visuality, images
are seen as agents in history, as articulations of social processes that in
themselves constitute and influence those processes. Images are formed by the
cultural system they represent, but they also partake in the formation of this
system, a formation that works as an ideological authorization. This means that
images should be interpreted not only as mere illustrations reflecting historical
processes; they are in themselves historical processes that contribute to the
culturally defined self-comprehension of an epoch. According to Griselda
Pollock: ‘Culture can be defined as those social practices whose prime aim is
signification, that is, the production of sense or making orders of 'sense' for the
world we live in. Culture is the social level in which are produced those images
of the world and definitions of reality which can be ideologically mobilized to20
20 Michael Baxandall. Patterns of Intention. On the Historical Explanation of Pictures

(New Haven: Yale University Press. 1985), chaps 1,2.
legitimize the existing order of relations of domination and subordination’.21

All images, not only those with an overtly ideological propaganda objective,
can be taken to interact with the cultural system of their setting in one way or
another. Even a picture that seemingly makes no impression whatsoever in its
time works as a focal point where different ideologically defined discourses are
confronted and visualized. The production and reception of an image as a
concrete manifestation in visual form of the social processes of a certain time
and at a certain place means that those processes are being worked upon.
At several points already 1 have defined images as products. This implies an
approach to images as being manufactured in specific historical situations,
subject to social and economic relations among the different agents and institu
tions, for instance the patron and the artist or, in this case, the artisan and his
atelier. Speaking of images as products also makes way for the concept of inten-
tionality: the images are systems of meaning, made to be read and internalized as
part of the ideological process that constitutes a culture. The images can function
in accordance with the dominant ideology or against it, but they always are
related to it in one way or another, a feature that they share with all other
contemporary discourses. The distinctive characteristic of images, of course, is
that they are non-verbal, and in the following, I will concentrate on the visual
communication of meaning, again with the Burgundian miniature as my object
of investigation.
To structure my analysis I will take up different aspects separately. I will
begin with the production aspect: to what extent does the miniature reveal the
condition of its manufacture? This question is connected with the temporal
aspect: in what ways is the making of the painting during a certain span of time
visible in the finished result? Next comes the spatial or visual aspect, by which 1
mean the co-ordination of picture and picture surface and of picture and
ornamentation. This leads me to the textual aspect, the relationship between the
pictorial elements and the text and finally to the functional aspect, the function
of the illumination in the manuscript as a whole. I do not claim to present an
analysis that is either unique for this miniature, or of general validity for late
medieval illuminations in general. My purpose is to verbalize and exemplify
conditions for a reading of images that is grounded in the epistemological value
of their visuality as such.
First there is the production aspect. This miniature, as most illuminations in
general, manifests itself as something strikingly costly and precious with refer
ence to the time and labour spent in manufacturing it, as well as to the value of
the materials used. Even without any more detailed knowledge concerning the
process of making a fifteenth-century illumination (the design is planned and the
21 Griselda Pollock. "Vision. Voice and Power: Feminist Art Histories and Marxism’,
in Vision and Difference. Femininity. Feminism and the Histories of Art. ed. by Griselda
Pollock (London: Routledge, 1988), pp. 18—49 (p. 20).
426 L en a L i epe
page ruled, the pictorial elements are sketched with a hard pointed stylus or in
graphite and then drawn more elaborately in coloured ink. the gilding is applied,
and then the colours and finally the shadings and other finer details are executed;
also, somewhere in the process, the text is engrossed) it is obvious for the viewer
that a single page like this one requires a considerable amount of time to make,
with its degree of detail and its intricate ornamentation in miniature scale. This
means that the illumination presents itself as the visual result of great effort—
and Philippe le Bon as the person who can afford to pay for this effort. The
materials used send the same signals, especially the gilding and the lavish use of
blue pigment that defines the manuscript as a markedly valuable item.“
There is a contradiction, however, between the time and effort that have been
invested in the making of the illumination and the temporality of the finished
result. The presentation scene and the ornamentation manifest themselves as
completed works: none of the previous stages in the production of the miniature
show in the finished result, apart from the thin red underlining of the dark letters
of the text. The presentation scene, as well as the miniature as a whole, is fixed
and frozen in a present that lacks a past, as a closed and finished fact that bears
no testimony to its history of production.*23 The effort required to make the
illumination is manifested not as the work of the different artisans but as
something for which Philippe le Bon has had the capacity to pay for. The
production process is not visible in itself, other than as the finished result.
Next comes what 1 have called the spatial aspect: the organization of the
visual elements on the page as a whole. The presentation scene naturally draws
most attention, but on closer examination there is a striking degree of integration
of the different parts of the composition—the pictorial scene, the flowery
ornamentation, the frames, the coats of arms, and not the least text—into a
tightly knit unity. In the overall composition, the presentation scene functions as
the main centre of gravity, while the text area and the more sparsely decorated
left margin provide some breathing space on the otherwise very dense page. The
vegetative ornamentation (consisting of acanthus leaves, vines, strawberry
tendrils, small thorny fruits, and various flowers; roses, columbines, corn
flowers, and carnations) along the margins functions as a compositional
anchoring of the totality: asymmetrical and graceful but at the same time rather
firmly structured, a feature that is strengthened by the slightly irregularly placed
coats of arms of Philippe le Bon himself, the duchy of Burgundy, Brabant
(twice), Limburg, Flanders, and Hainaut. The text scrolls repeat the motto of
Philippe le Bon: 'aultre n’aray tant que viveray’ [as long as 1 live 1 will not have
anything/anyone else].
“ Alexander (as in note 7). p. 40.

23 See Norman Bryson. Vision and Painting. The Logic of the Gaze (New Haven:
Macmillan, 1983), chap. 5.
The different visual elements are clearly demarcated and separated by the
frames. A hundred years later, in the middle of the sixteenth century, the epoch
of illuminated manuscripts is definitely over, replaced by panel painting of the
kind that is heralded by the presentation scene with its illusion of space in three
dimensions. The illumination of De regimine principum represents a middle
stage, an altitude of ambiguity towards the relation between picture and surface,
with text and ornamentation obviously belonging to the surface of the page,
while the presentation scene invalidates this two-dimensionality, as if the parch
ment has suddenly given way and we are seeing right through it onto a scene that
is taking place beyond its boundaries. In the words of Otto Pacht, this change
means that the pictorial field has become pictorial space: the illuminator, the
scribe and the decorator ‘each had an entirely disparate relationship to the page:
looking and reading had become two essentially distinct activities [. . .] now the
image was sometimes on "optical’ ground and at other times on a ‘real’ one’.2"1
As I see it, the combination of the illusionary space of the presentation scene
and the framing that stresses the illumination as surface-bound underlines the
representational character of the miniature. The intention is not to present an
eye-witness account of an actual event but to provide a visual representation not
only of the handing over of the volume but also of the validity of the text—
Aegidius Romanus’s De regimine principum—in relation to the rule of Philippe
le Bon. The coats of arms in the margins are symbolical representations of social
position with a significance that equals the figure of Philippe le Bon himself in
the presentation picture. The coats of arms function as visual claims of territorial
sovereignty and, thus, as a link between the presentation picture and the text.
The coats of arms in the surface-defined part of the illumination are counterparts
of the figure of the duke in the three-dimensional picture space .and similarly the
text of the manuscript has a real relation to the government of the duke; or
rather, the coats of arms can be seen as a verification that the regime of Philippe
le Bon should be interpreted in terms of De regimine principum.
This brings me to the textual aspect: the text as part of the visual as well as
the verbalized meaning of the illumination. Visually it is obvious that the text is
not the most important element of the page: the presentation picture dominates
powerfully, the gilded frames only just shield the text area from the dense
ornamentation and the text occupies no more than one fifth of the totality. Again,
it is the representational character of the illumination that is brought to the fore:
21Otto Pacht, Book Illumination in the Middle Ages. An Introduction (London: Harvey
Miller. 1986: orig. published as Buchmalerei des Mittelalters. Eine Einführung, Munich:
Prestel-Verlag. 1984), p. 200. Pacht does not specify which is which. As I understand
him. the text and the ornamentation are located on the ‘real’ ground, i.e., the two-
dimensional surface that is the actual surface of the illumination, while the three-
dimensional space of a miniature such as the one under discussion here belongs to the
"optical" ground, a picture plane that exists as an optical illusion but not in reality.
4 2 8 L ena L iepe
it is not the content of the text that is of importance, but rather the fact that it has
been presented to Philippe le Bon.
The text of the treatise De regimine principum deals with the ruler as rex
quasi semideus, meaning that he governs his subjects as God rules the world,
through a combination of Christian doctrines and Aristotelian ethics according to
which monarchy is the best form of government. The authority of the sovereign
is absolute, the subjects’ only duty is to obey. The text is divided into three
books: on controlling oneself according to the four cardinal virtues; on the ruling
of a household and a court; and on governing the state, with instructions as to
how those duties are to be performed. However, the text of the frontispiece is not
from De regimine principum itself; rather, it contains the dedication to Philippe
le Bon:
A son tresespecial seigneur, né de la royale et tressainte lignïe monseigneur
Phlippe premier, engenré et hoir de Trescler, homme monseigneur Phlippe par la
Grace de Dieu roy de France. Frere Gilles de Rome [Egidius Romanus] de l’ordre
des freres hermites saint Augustin son devot orateur toute recommendation et luy
meismes a toutes choses prest de servir. Mon treschier seigneur, la sentence des
politiques dit que toutes princiés ne sont point de egale durée c'est a dire que elles
ne durent point tant l'une comme l'aultre. et si ne sont point les regimes d'une
chascune seignourie rieulés par ung meisme cours, mais.. .
After an introductory reverence to the duke, the translator continues:
My dearest Lord, the judgement of the political historians says that not all periods
of rule are of the same duration, that is they do not all last as long as one another;
and nor do the forms of government in all countries follow the same rules.
Thus, the text of the page opens as a verbal counterpart to the presentation
picture and then continues as a transition to the theme of the treatise itself, with
an evaluation of different kinds of governments.
The last aspect is the function of the illumination related to the text and the
manuscript as a whole. In her article Reading Medieval Images, Vera F. Vines
sums up the functions of images in medieval manuscripts:
They sharpened the visual interest of the page by interrupting the flow of the
written word: cartoon-like, at times they provided the viewer with near-substitutes
for the text; alternatively, they could communicate meaning by furnishing stories
with descriptive explanations or moralizing commentaries, to heighten the impact
of the text. Picture and word were in conversation with one another, collaborating
in intricate ways’.“
3 Translated by associate professor Trond Kruke Salberg. Department of Classical and

Romance Studies. University of Oslo.
“ Vera F. Vine, ‘Reading Medieval Images: Two Miniatures in a Fifteenth-Century
Missal', in Medieval Texts and Images. Studies of Manuscripts from the Middle Ages. ed.
by Margaret M. Manion and Bernard J. Muir (Chur: Harwood Academic Publishers.
In this case it is hardly a question of the picture as a sharpening of the visual

interest by interrupting the flow of text, since the presentation picture is the
opening of the whole manuscript and furthermore is the only pictorial scene
included. It establishes with a certain emphasis the terms of reading: that the
relevance of Jean Wauquelin’s translation of De regimine principimi is to be
related to the government of Philippe le Bon. The presentation picture constructs
a relation between the duke and the text. The duke is the master of the situation:
he is standing in the centre of the room in an attitude that signals sovereignty,
and he is enclosed by empty space that creates a distance between him and the
other figures. Wauquelin is the only person on his knees: the way Wauquelin is
rendered, in a subordinate position in the composition as a whole, puts the
emphasis on the duke on the one hand and the open volume, rather than its
translator, on the other.
The presentation picture contextualizes the text. This brings me back to the
definition of an image as an historical process, something that actively generates
cultural self-comprehension. A picture such as this one visualizes and works
upon mental structures connected with the conditions of power politics, social
hierarchies, cultural education and political authority, and, more generally, the
relation between reality and principles, between the actual rule of Philippe le
Bon (manifested in the illusionary space of the presentation picture as well as in
the coats of arms in the two-dimensionality of the margins) and the precepts for
ruling as formulated in the thirteenth century by Aegidius Romanus and
inscribed here as part of the totality of the page. The illumination has
communicated its meaning to a larger public than the duke personally: I have
already described the collecting of splendid manuscripts as a manifestation not
only of cultural capital but also of political power and social status at the
Burgundian court in the middle of the fifteenth century. To this can be added
that the miniature has by no means ceased to create an historically defined
understanding of Philippe le Bon and his rule: in fact, the potential of the image
to set the terms for this understanding has been the core of my whole argument.
The E pistem ology o f Im ages
In this article, I have tried to formulate a reading of images that is not restricted
to either pure form or pure content, that is, a reading that is more than merely a
specification of the diagonals of the composition or a naming of the historical
persons who can be identified. The interpretations that I have suggested have
throughout been anchored in the visual properties of the illumination, according
to the fundamental assumption that the image as such has a cognitive potential. I
1991), pp. 127-48 (p. 127).

4 3 0 L e n a L ie p e
have searched for meanings that can be detected in the illumination as a visually
complex structure. Knowledge of the historical context has been fundamental,
but at the same time the illumination has been recognized as a formative power
in relation to this same context.
The question with which I began was: on what, and on whose, conditions
should an image be interpreted? My answer is that, if an image is defined as an
historical source, at the same time it is necessary to respect the epistemological
value that is grounded in its visuality. To reduce an image to a mere illustration
of historical conditions and at the same time to refuse to acknowledge it as a
visual representation on its own terms, is to disregard its distinctive character as
image. Ultimately I want to define this as a matter of ethics; it is about making it
possible for an image to communicate as what it actually is: something that, at
least sometimes, can have more to say than a whole lot of words.*
This paper has also been read in Swedish, in a report from the conference ‘De
historiska vetenskaperna och bilden'. Department of History, Lund University. February
1999.

BOLVIG - LINDLEY - History and Images - Towards A New Iconology

Uploaded by

Copyright:

Available Formats

BOLVIG - LINDLEY - History and Images - Towards A New Iconology

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BOLVIG - LINDLEY - History and Images - Towards A New Iconology

Uploaded by

Copyright:

Available Formats

What is the topic of the document?

What is the topic of the document?

What is discussed on pages 3-4?

What is discussed on pages 3-4?

History and Images

MEDIEVAL TEXTS AND CULTURES OF NORTHERN EUROPE

EDITORIAL BOARD UNDER THE AUSPICES OF THE

Wendy Scase. Chair (University of Birmingham)

Andrew Ayton (University of Hull)

Axel Bolvig and Phillip Lindley

© 2003 Brepols Publishers n.v., Tumhout, Belgium

All rights reserved. No part of this publication may be reproduced,

Printed in the EU on acid-free paper.

Art and history: the legacy of Johan Huizinga.......................................................3

Images and the historian.......................................................................................19

Pourquoi élaborer des bases de données d’image? Propositions pour une

Six St Jeromes: notes on the technology and uses of computer lighting

Innovative hybride graphische Systeme zur Denkmalüberwachung und

‘Serra ex ferro’— ’Serra ex vitro’: medieval history—computers—image

Quantitative image analysis: the Painter of wooden shoes.............................. 229

Les marges à drôleries des manuscrits gothiques : problèmes de méthode..... 277

‘Primitive’ paintings: the visual world of populus rusticus............................. 301

Representations of Jews in Danish medieval art—can images be used as

Anti-semitism, image desecration, and the problem of ‘Jewish execution’..... 357

Framing history with salvation..........................................................................379

On the epistemology of images..........................................................................415

Jörgen van den Berg, Hans Brandhorst, Peter van Huisstede

Rolf-Jürgen Grote and Annette Hornschuch

2. Haus Altenkamp, räumliche Darstellung des Entrées überlagert mit

Frank Colson, Jean Colson, Ross Parry, Andrew Sawyer

proper’ peasant), ink drawing, South German, 1475. Vienna, Östereich.

5. Chartres Cathedral, north transept, right porch, right outer archivolts,

3. Church of Nousiainen, Southern Finland, ‘Primitive painting’. ©

12. Tensta church. ‘Revelations of St Birgitta'. © Author

B Levy, Reader in French at the University of Hull, for his exemplary

Bayerische Staatsbibliothek, Munich

Church of St Gumbert, Ansbach

Dean & Chapter. Lincoln

Gabinetto di disegni Uffizii, Florence

Herzog August Bibliothek, Wolfenbüttel

Keble College Library, Oxford

Landesamt für Denkmalpflege, Hannover

Musée Carnavalet, Paris

National Board of Antiquities, Finland

Rijksprentenkabinet, Rijksmuseum. Amsterdam

Trinity College, Cambridge

Walford, Elke, Kunsthalle. Hamburg

artistic’ combination of a mimetic configuration and a linguistic statement is one of

5 Brendan Cassidy, ‘Introduction. Texts, and Audiences’, in Iconography at the Cross­

the mimetic representational power of the most realistic visual medium,

s Irving Lavin, ‘Iconography as a Humanistic Discipline (“Iconography at the Cross­

possibilities for comparative analysis of historical image information.14 Several

14 Gerhard Jaritz, ‘Comparative Analysis of Historical Image Information’, in

image—we create a new image with a resemblance to the original.16 Each

16 Spreti Kjprup. ‘Billcdkommunikation’, in Visuel Kommunikation, vol. 1. ed. by Bent

he present reputation of Johan Huizinga is a curious one. His masterpiece,

by the distinguished connoisseur of Northern Art, Horst Gerson, in an otherwise

2 H. Gerson, ‘Huizinga und Kunstgeschichte', in Johan Huizinga 1872-1972, papers

Fig. 3. ‘Annunciation and Visitation Melchior Broederlam. Tempera on wood,

5 Brendan Cassidy, ‘Introduction. Texts, and Audiences’, in Iconography at the Cross

s Irving Lavin, ‘Iconography as a Humanistic Discipline (“Iconography at the Cross

Plate 21: Haus Altenkamp, Entrée, Wandmalerei-Ausschnitt der Südwand, Über

Plate 24: Die Überlagerung korreliert die Schadenskartierung mit bauphysikal