Supports in Roman marble sculpture: workshop practice and modes of viewing 9781108418430, 9781108290036, 9781108407106, 1108407102, 1108418430

Table of contents :
Introduction --
Part I. Material and History: 1. Narrative and individuality
2. Supports and Greek marble sculpture
3. The world of struts
4. Surface and colour --
Part II. The Limits of Stone: 5. The quest for solidity
6. Statuesque statues
7. Value and ingenuity
8. Carving and tradition --
Conclusion.

Citation preview

i

Supports in Roman Marble Sculpture Workshop Practice and Modes of Viewing

Figural and non-​figural supports are a ubiquitous feature of Roman marble sculpture; they appear in sculptures ranging in size from miniature to colossal and of all levels of quality. At odds with modern ideas about beauty, completeness and visual congruence, these elements, especially non-​figural struts, have been dismissed by scholars as mere safeguards for production and transport. However, close examination of these features reveals the tastes and expectations of those who commissioned, bought and displayed marble sculptures throughout the Mediterranean in the Hellenistic and Roman periods. Drawing on a large body of examples, Greek and Latin literary sources, and modern theories of visual culture, this study constitutes the first comprehensive investigation of non-​figural supports in Roman sculpture. The book overturns previous conceptions of Roman visual values and traditions and challenges our understanding of the Roman reception of Greek art. Anna Anguissola is a Lecturer in Classical Archaeology at the University of Pisa and directs field projects at Pompeii and Hierapolis in Turkey. She is the author of Intimità a Pompei: Riservatezza, condivisione e prestigio negli ambienti ad alcova di Pompei (2010) and Difficillima Imitatio. Immagine e lessico delle copie tra Grecia e Roma (2012).

ii

iii

Supports in Roman Marble Sculpture Workshop Practice and Modes of Viewing Anna A n g u is s ol a Università di Pisa

iv

University Printing House, Cambridge CB2 8BS, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–​321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi –​110025, India 79 Anson Road, #06-​04/​06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108418430 DOI: 10.1017/​9781108290036 © Anna Anguissola 2018 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2018 Printed in the United Kingdom by TJ International Ltd., Padstow, Cornwall A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-​in-​Publication Data Names: Anguissola, Anna, author. Title: Supports in Roman marble sculpture : workshop practice and modes of viewing / Anna Anguissola, Università degli Studi, Pisa. Description: New York : Cambridge University Press, 2018. | Includes bibliographical references and index. Identifiers: LCCN 2017046248 | ISBN 9781108418430 (hardback) Subjects: LCSH: Marble sculpture, Roman. | Struts (Stone carving) Classification: LCC NB115 .A47 2018 | DDC 730.28/4–dc23 LC record available at https://lccn.loc.gov/2017046248 ISBN 978-​1-​108-​41843-​0 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-​party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

v

To Sarah

vi

vii

Contents

List of Figures  [page viii] Preface  [xv] Acknowledgements  [xx] List of Abbreviations  [xxiii]

Introduction  [1] Part I  Material and History  [25] 1 Narrative and Individuality  [27] 2 Supports and Greek Marble Sculpture  [44] 3 The World of Struts  [59] 4 Surface and Colour  [84] Part II  The Limits of Stone  [103] 5 The Quest for Solidity  [105] 6 Statuesque Statues  [128] 7 Value and Ingenuity  [157] 8 Carving and Tradition  [179] Conclusion [199] References  [222] Index  [253]

vii

viii

Figures

1

2

3

4

5

6

7

8 viii

Agostino Cornacchini, Equestrian monument of Charlemagne, 1720–​5. Marble. Vatican, south end of the atrium of the basilica of St Peter (photograph: © Getty Images)  [page 2] Statue of Ganymedes abducted by the eagle. Marble. H. 103 cm. Vatican, Museo Pio Clementino, Galleria dei Candelabri, inv. no. 2445 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​4862)  [6] Group of the Scylla from Sperlonga, detail of the boat and helmsman. Marble. H. max preserved of the boat 165 cm and L. max 290 cm; L. max preserved of the helmsman 160 cm. Sperlonga, Archaeological Museum (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​65.110, J. Felbermeyer) [17] Statue of Melpomene from the Baths of Faustina at Miletus, early Antonine period. Marble. Detail of the neck strut. Istanbul, Archaeological Museum, inv. 1993 (photograph: German Archaeological Institute, Istanbul, neg. D-​DAI-​IST-​78-​165)  [20] Statue of Melpomene from the Baths of Faustina at Miletus. H. 151 cm (photograph: German Archaeological Institute, Istanbul, neg. D-​DAI-​IST-​78-​168)  [21] Statue of a Dancing Satyr, Antonine period. Marble. H. 146 cm. Rome, Galleria Nazionale d’Arte Antica di Palazzo Corsini, inv. 710 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ ROM-​74.717, C. Rossa)  [30] Statue of a Dancing Satyr from the Kerameikos, late Hadrianic or Antonine period. Marble. Athens, Museum of the Kerameikos, inv. 8071 (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​Kerameikos-​10701. All rights reserved) [31] Statue of the Diadoumenus from Delos, ca. 100 BC. Marble. H. 195 cm with the plinth (186 cm without the plinth). Athens, National Archaeological Museum, inv. 1826 (photograph: German

ix

List of Figures

9

10

11

12

13

14

15

16

Archeological Institute, Athens, neg. D-​DAI-​ATH-​NM 5321, Eva-​Maria Czakó. All rights reserved)  [35] Statue of Dionysus with a personification of grape vines (Ampelus), second century AD. Marble. H. 158 cm. London, British Museum, reg. no. 1805,0703.1/​no. Sculpture 1636 (photograph: © The Trustees of the British Museum. All rights reserved) [37] Group of Silenus cradling Dionysus, first century AD. Marble. H. 199 cm with the plinth (191 cm without the plinth). Munich, Glyptothek, inv. Gl. 238 (photograph: © Staatliche Antikensammlungen und Glyptothek München, Renate Kühling) [40] Group of Silenus cradling Dionysus, Flavian period. Marble. H. 190 cm. Paris, Louvre, inv. MR 346/​Ma 922 (photograph: © Musée du Louvre, RMN-​Grand Palais, Thierry Ollivier)  [41] Aristodikos kouros, ca. 510–​500 BC. Parian marble. H. 198 cm. Athens, National Archaeological Museum, inv. no. 3938 (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​NM-​5049, Eva-​Maria Czakó. All rights reserved) [47] Statue of Hermes with infant Dionysus. Marble. H. 215 cm. The statue has either been considered an original carved by Praxiteles of the fourth century BC or a copy made in the late Hellenistic or Roman period. Olympia, Archaeological Museum (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​Olympia-​392. All rights reserved)  [50] Statue of Aknonios from the so-​called Daochos monument, seen from the left side. Marble. H. 180 cm. Delphi, Archaeological Museum (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​Delphi-​397, Gösta Hellner. All rights reserved) [53] Statue of a boy from the shipwreck of Antikythera, early first century BC. Parian marble. H. 111.5 cm with the plinth. Athens, National Archaeological Museum, inv. 2773 (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​1969-​1706, Gösta Hellner. All rights reserved)  [56] Statue of Silenus from a Roman villa at Ariccia, Trajanic period. Marble. H. 175 cm. Vatican Museums, Galleria Chiaramonti, inv. 1441 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​90Vat.384, K. Anger)  [60]

ix

x

x

List of Figures

17

18

19

20

21

22

23

24

25

26

Group of Three Graces from the frigidarium in the baths of Cyrene, Hadrianic period. Marble. H. 140 cm. Shahat (Cyrene), Museum, inv. 14.348 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​58.2307R, F. X. Bartl)  [64] Group of Three Graces from the frigidarium in the baths of Cyrene, Hadrianic period. Marble. H. 116 cm. Shahat (Cyrene), Museum, inv. 14.346 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​58.2197R, F. X. Bartl)  [65] Headless statue of Ulysses from the nymphaeum at Punta Epitaffio, first half of the first century AD. Marble. H. 175 cm. Baiae, Museo Archeologico dei Campi Flegrei, inv. 222736 (photograph: © Bildarchiv Foto Marburg /​Schumacher, Dieter; Gloc, Jan; Haag, Paul)  [67] Statue of a youth, interpreted as Eros, Thanatos, or Apollo, second century AD. Marble. H. 164 cm. Rome, Musei Capitolini, inv. 1092 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ ROM-​2001.1900R, K. Anger)  [71] Statue of a Pouring Satyr from Torre del Greco near Naples (Villa Sora), mid-​first century AD. Marble. H. 146 cm (163 cm with the raised right hand and oinochoe). Palermo, Archaeological Museum A. Salinas, inv. 1556 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​71.661, G. Singer)  [73] Statue of the Diadoumenus, last quarter of the first century AD. Restored with casts of the Delos Diadoumenus. Marble and plaster. H. 185 cm. New York, Metropolitan Museum of Art, Fletcher Fund 1925, acc. no. 25.78.56 (photograph: © The Metropolitan Museum of Art)  [76] Drawn reconstruction of the Diadoumenus at the Metropolitan Museum. Plaster integrations are in grey; the reconstructed right hand and wrist, fillet and struts are represented with a dotted line (drawing by the author and Donato Bruscella)  [77] Drawn reconstruction of the Discobolus from Castel Porziano (now in Rome, Museo Nazionale Romano). Reconstructed parts are in grey (drawing by the author and Donato Bruscella)  [79] Statue of Polyphemus from Sperlonga, right arm and hand. Marble. Sperlonga, National Archaeological Museum (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​69.1941, G. Singer) [81] Standing male portrait from Delos, best known as pseudo athlete, early first century BC. Marble. H. 225 cm. Athens, National

xi

List of Figures

27

28

29

30

31

32

33

34

35 36

Archaeological Museum, inv. 1828 (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​1994-​32, Elmar Gehnen. All rights reserved)  [85] Portrait statue of Faustina the Elder with the body type known as Large Herculaneum Woman, mid-​second century AD. Marble. H. 210 cm. Detail of the veiled head with the neck strut. Malibu, The J. Paul Getty Museum, inv. 70.AA.113 (photograph: © The J. Paul Getty Museum, digital image courtesy of the Getty’s Open Content Program)  [90] Amazonomachy sarcophagus from Rome, mid-​second century AD. L. 247 cm. Rome, Capitoline Museums, Palazzo Nuovo, inv. S 726 (photograph: © Archivio Fotografico dei Musei Capitolini)  [93] Medea sarcophagus from Rome, mid-​second century AD. L. 227 cm. Detail of the left corner. Berlin, Antikensammlung, inv. SK 843 b (photograph: © Antikensammlung, Staatliche Museen zu Berlin –​Preussischer Kulturbesitz)  [94] Group of Aphrodite, Pan, and Eros (so-​called ‘Slipper-​Slapper group’ or Pantoffelgruppe) found in the House of the Poseidoniasts of Berytos at Delos, ca. 100 BC. Marble. H. 155 cm with the inscribed base. Athens, National Archaeological Museum, inv. 3335 (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​NM-​2989, Gabriel Welter. All rights reserved) [99] Gianlorenzo Bernini, Apollo and Daphne, 1622–​5. Marble. H. 243 cm. Rome, Galleria Borghese, inv. N. CV. Detail (photograph: © Getty Images)  [106] Statue of a boy, Augustan period. Bronze. H. 132.4 cm. New York, Metropolitan Museum of Art, acc. no. 14.130.1, Rogers Fund 1914 (photograph: © The Metropolitan Museum of Art/​Art Resource/​ Scala, Florence)  [110] Statue of Polyphemus from Sperlonga, left foot. Sperlonga, National Archaeological Museum (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​65.67R, J. Felbermeyer)  [113] Drawn reconstruction of the ship of Ulysses from the group of Scylla at Sperlonga (drawing by the author and Donato Bruscella) [123] Drawn reconstruction of the group of Polyphemus at Sperlonga (drawing by the author and Donato Bruscella)  [124] Statue of Aristogeiton, Hadrianic–​Antonine period. Marble. H. 183 cm. Naples, National Archaeological Museum, inv. 6010

xi

xii

xii

List of Figures



(photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​58.1796, F. X. Bartl)  [130] 37 Statue of Aristogeiton from the area of S. Omobono, first century BC. Marble. H. 180.5 cm without plinth. Rome, Musei Capitolini, inv. 2404/​Centrale Montemartini, no. II.58 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​99.1606, K. Anger) [131] 38 Statue of Pouring Satyr from Rome, second century AD. Marble. H. 144 with the plinth. Paris, Louvre Museum, inv. 2333 (photograph: © Musée du Louvre, RMN-​Grand Palais, Daniel Lebée/​Carine Déambrosis)  [135] 39 Statue of the so-​called Protesilaos, Antonine period (head). Marble. H. 220 cm. New York, Metropolitan Museum of Art, acc. no. 25.116, Hewitt Fund 1925 (photograph: © The Metropolitan Museum of Art)  [136] 40a–​b Drawn reconstruction of the so-​called Protesilaos. Front and rear views. The reconstructed struts are represented with a dotted line (drawing by the author and Donato Bruscella)  [137] 41a–​b So-​called Dresden Artemis, mid-​second century AD. Marble. H. 152 cm with the plinth. Front and rear views. Dresden, Albertinum, inv. Hm 117 (photograph: © Skulpturensammlung, Staatliche Kunstsammlungen Dresden, H. P. Klut/​Elke Estel) [140] 42 Statue of a Pouring Satyr, late Hadrianic–​early Antonine period. Marble. H. 156 cm. The head is a plaster cast of the Dresden replica. Berlin, Antikensammlung, inv. Sk 257 (photograph: © Antikensammlung, Staatliche Museen zu Berlin –​Preussischer Kulturbesitz) [151] 43a–​c Detail drawing of the strut under the left hand of the Pouring Satyr in Palermo. The strut is 39 cm long (drawing by the author and Donato Bruscella)  [153] 44 Statue of a man using the body-​scraper (Apoxyomenos), mid-​first century AD. Marble. H. 205 cm. Vatican Museums, Museo Pio-​Clementino, Gabinetto dell’Apoxyomenos, inv. 1185 (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​57.898, R. Sansaini)  [158] 45 Headless statue of Venus from the Baths of Faustina at Miletus, second century AD. Marble. H. 135 cm. Istanbul, Archaeological Museum, inv. 2004 (photograph: German Archaeological Institute, Istanbul, neg. D-​DAI-​IST-​78-​94)  [161]

xiii

List of Figures

46a–​b Statue of Dionysus with a panther from the nymphaeum at Punta Epitaffio, first half of the first century AD. Marble. H. 140 with the base. Front and left-hand side view. Baiae, Museo Archeologico dei Campi Flegrei, inv. 222739 (photograph: © Bildarchiv Foto Marburg/​Schumacher, Dieter; Gloc, Jan; Haag, Paul)  [162] 47a–​d Drawn reconstruction of the Vatican Apoxyomenos. The reconstructed strut is represented with a dotted line. Front, rear, left-​hand side, and right-​hand side view (drawing by the author and Donato Bruscella)  [166] 48 Statue of the so-​called ‘third companion’ of Ulysses, from the Polyphemus group of Sperlonga. Marble. Sperlonga, National Archaeological Museum (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​65.1932)  [173] 49 Group of the Scylla from Sperlonga, detail of a sailor seized by the monster. Marble. H. max preserved 186 cm. Sperlonga, Archaeological Museum (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​65.120, J. Felbermeyer)  [174] 50 Michelangelo, Bacchus, 1496–​7. Marble. H. 209 cm with the base. Florence, Museo Nazionale del Bargello, inv. S.10 (photograph: Kunsthistorisches Institut in Florenz –​ Max-​Planck-​Institut)  [180] 51 Statue of the wineskin-​bearing companion of Ulysses from the nymphaeum at Punta Epitaffio, first half of the first century AD. Marble. H. 169 cm. Left-​hand view. Baiae, Museo Archeologico dei Campi Flegrei, inv. 222737 (photograph: © Bildarchiv Foto Marburg/​Schumacher, Dieter; Gloc, Jan; Haag, Paul)  [184] 52 Torso of Zeus, from the Nymphaeum of Herodes Atticus at Olympia, mid-​second century AD. Marble. H. max preserved 167 cm. Olympia, Archaeological Museum, inv. Λ 170 (photograph: German Archaeological Institute, Athens, neg. D-​DAI-​ATH-​1979-​3529, Gösta Hellner. All rights reserved)  [186] 53 Statuette of Heracles, from Rome. Marble. H. 57 cm with the plinth. Boston, Museum of Fine Arts, inv. 14.733 (photograph: © Museum of Fine Arts, Boston)  [192] 54 Statuette of Diana, late fourth century AD. Marble. H. 71 cm. Dresden, Albertinum, inv. Hm 270 (photograph: © Skulpturensammlung, Staatliche Kunstsammlungen Dresden, H. P. Klut/​Elke Estel)  [197] 55 Utagawa Kunisada I, Scene from a Bunraku theatre performance, 1856. Woodblock print, ink and colour on paper. 35.5 × 24.3 cm.

xiii

xiv

xiv

List of Figures

56

57

58

59

60

Boston, Museum of Fine Arts, William Sturgis Bigelow Collection, acc. no. 11.30063 (photograph: © Museum of Fine Arts, Boston) [200] Statue of a sacrificing woman, Severan period. Marble. H. 189 cm. Florence, Galleria degli Uffizi, inv. 1914 no. 131 (photograph: © Gallerie degli Uffizi, Gabinetto Fotografico)  [202] Discus of a lamp, first half of the first century AD. L. 20.7 cm, W. 9.8 cm. London, British Museum, Reg. 1856,12–​26.552.a (photograph: © The Trustees of the British Museum. All rights reserved) [209] Sarcophagus chest with the myth of Prometheus, late third or early fourth century AD. L. 117.5 cm, H. 66 cm, D. of side 43.5 cm. Detail. Rome, Capitoline Museums, Sala delle Colombe, inv. 329 (photograph: Forschungsarchiv für Antike Plastik, Cologne, neg. FittCap71-​108-​05_​16404,10, http://arachne.uni-koeln.de/item/ marbilder/3563418) [211] Replica of the Pouring Satyr in Palermo. Left-​hand view (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​71.662, G. Singer)  [214] Replica of the Discobolus from Castel Porziano. Rear view (photograph: German Archaeological Institute, Rome, neg. D-​DAI-​ROM-​8396, C. Faraglia)  [220]

xv

Preface

‘Inartistic’, ‘offensive’, ‘disfiguring’, ‘unsightly’, ‘disturbing’  –​these are but some of the words that scholars have used to describe the subject of this book:  the structural supports of Roman marble sculpture. Functionally, a support is a mass of stone left in place to reinforce a point of potential weakness in a statue. We can distinguish two classes of supports: first, the figural supports shaped like tree stumps, vases, animals, weapons, etc., that secure a marble statue’s stance and contribute to the narrative; and, second, the non-​figural or structural supports which in most Roman marble statues seem only to sustain protruding extremities such as outstretched arms. This book focuses on the latter, commonly known as ‘structural supports’ or ‘struts’ (‘Stützen’, ‘Stege’, and ‘Streben’ in German, ‘tenons’ in French, and ‘puntelli’ or ‘tenoni’ in Italian). Scholars have seldom and only cursorily engaged in the task of accounting for the function of struts in Roman sculpture, wavering between two main explanations. The first envisions struts as tools that enabled Roman workshops to transform lighter bronze prototypes into heavier marble replicas. Precisely because they were made of comparatively light-​weight bronze, these prototypes did not need any additional reinforcements. The second interprets struts as devices to secure appendages for transport. In consequence, struts testify to a sculpture’s travel from workshop to the site of display. In contrast to past scholarship, this book revolves around aspects that have so far remained unexplored. Do the conventional readings of supports actually address the diverse phenomena that they physically represent? How did the carving, shape, and display of supports influence the readings of Roman marbles in different times and contexts? How can the assessment of supports contribute to the debates about the visual values of Roman culture? In order to answer these and further questions, figural and especially non-​figural supports need to be analysed within a comprehensive approach to Roman sculpture, Roman techniques of marble carving, and Roman debates about arts especially from the first to the third centuries AD. Certainly, many supports were primarily structural and not meant to play an important role in the appearance of the sculpture. Sometimes, however,

xv

xvi

xvi

Preface

their share in the visual impact of a work outweighs their tectonic role. The careful finish of many struts belies the idea that they were mere technical devices, intended to be concealed from the public eye –​an assumption rooted in the modern aesthetic interpretations of the classical past. While stabilising an expressive body gesture, struts draw attention to the pose and encourage consideration of its meaning. These observations contradict the hypothesis that struts were always intended as practical precautions for transport. Besides not being removed once a statue was finally put in place, they were occasionally added without regard to actual static concerns, in prominent yet non-​functional positions. The exceptionally large dimensions of some struts put them (and the limbs they support) at greater risk of breaking than any other part of a statue. Expert craftsmanship was required to chisel out such struts without damaging or destroying them. The question of supports in statues that belong to well-​known replica series, too, prompts further considerations. The shape, size, and position of both figural and non-​figural supports tend to differ within a single replica series. Some of the versions of a given prototype avoid supports almost entirely, while elsewhere struts proliferate in both useful and unnecessary spots. The implications seem to be more complex and wide-​ranging than has been anticipated by traditional copy criticism. Supports and conspicuous struts might have stimulated the memory of similar sculptures, the composition of which followed different criteria, and might have tied multiple copies into a system of mutual references. Besides, supports may have been recognised as copyists’ additions and, in turn, as distinctive quality markers of Roman marbles. Since supports were the only features not defined by the prototype, they complicated the already difficult process of translating bronze into marble. Roman sculptors had to devise new methods and technologies to both reproduce their model and insert huge three-​dimensional struts. Struts may show the workshop’s ability to retain the basic forms of the bronze prototype while producing striking, creative modifications, from both the visual and narrative points of view, in its marble versions. The choice of appending large struts to a statue also advertised the conspicuous consumption of marble. Larger struts required a great deal of extra marble and implied that the figure was carved from a single block. Acquiring such a block was a much more expensive alternative to carving the extended limbs separately and then fixing them to the core. Perhaps most essentially, supports complemented the standardised poses and played an essential role in adapting a set of visual formulae derived from the Greek tradition to Roman visual semantics. In this sense, figural and non-​figural supports acted as a means of specification and guided the

xvi

Preface

viewer in the process of reading the image. Within an essentially conservative visual culture, the value of supports and struts as indicators of both Greek tradition and Roman innovation, in terms of either composition or content, may explain their widespread popularity. The Introduction sets the topic within the broader scholarly debate. How did scholars respond to the ubiquity of struts in the corpus of Roman statuary? Which explanations were offered to account for this material? Do technical needs necessarily rule out aesthetic choices? The problem of struts is laid out here in a broader perspective, including later periods of Western art. My argument unfolds over two sections and eight chapters. I address the topic by discussing individual case studies in detail and relating them to two main sets of broader issues: first, the ancient Roman technologies for marble sculpture and second, the contexts and modes of viewing sculpture in the Roman world. This book explores a pervasive –​yet largely overlooked –​phenomenon about the presentation of Roman marble statuary as a source of information about the mechanisms of production, trade and appreciation of art. The first section, ‘Material and History’, lays the groundwork to contextualise the discourse on struts in Roman marble sculpture. To do so, I first reconsider the question of figural supports, which, unlike the subject of this book, have traditionally been incorporated in the study of Roman art, as elements deemed integral to the composition. Chapter 1 questions the narrative potential of figural supports and examines their role in defining a figure’s structure, symmetry, and movement –​a perspective that has much broader implications in the case of non-​figural supports. Chapter  2 discusses the emergence of supports in the Greek art from the sixth to first centuries BC. The concept and practice of supports was deeply ingrained in the Greek tradition of marble sculpture. However, it was from the late first century BC onwards that supports and struts became one of the most familiar features of marble sculpture, in a variety of shapes and with different surface finish. In Chapter 3, I describe the types and shapes of struts in the Roman sculptural production. Struts occur in statues of exceptionally high quality as well as those of poorer workmanship. They are found in both colossal groups and statuettes. As is described in Chapter  4, the types of surface treatment are equally various and idiosyncratic. Within this framework, the issue of colour is, at the same time, exceptionally relevant and elusive. Painting has traditionally played an important role in the belief that struts would be hidden from sight. Unfortunately, traces of paint have been detected on the struts of very few sculptures. This remarkably scanty

xvii

xvi

xviii

Preface

record hardly warrants general conclusions. This empirical presentation of the material provides the coordinates to navigate the subject, as well as the basic categories to distinguish and describe non-​figural supports. Since these features have, at best, attracted only cursory remarks, the study of struts cannot ignore the need for a review of the evidence and a vocabulary to discuss it. While doing this, the presentation of the material helps to pinpoint certain geographical and chronological concentrations. In the second section, ‘The Limits of Stone’, I review the main functional arguments that account for the presence of struts, and then explore the potential of struts in conveying information about the quality of a statue and the ingenuity of its maker. Struts may be useful to copy a bronze prototype into marble, to brace protruding extremities during the work, to guarantee the safety of a statue during transport, or to ensure stability and balance once the statue was in place. Most struts certainly had an immediate functional utility of this sort. Yet, as I show in Chapter 5, these explanations are not all equally persuasive. Connecting bars were not exclusive to marble statuary. Although rare, they also occur on bronze statues, where they serve as stabilisers or attribute holders. Additionally, the evidence that struts could serve as safeguards against the percussive strokes of the sculptor’s tools does not apply to the whole sample. On the contrary, the shaping and smoothing of very long struts with deep undercuts poses a further threat to the safety of a statue. Lastly, struts do not seem to be the sole or the safest choice to ensure that a statue would survive transport intact. Alternative methods, such as shipping the statue unfinished or in pieces, would reduce the risk of damage and breaks during transport. Chapter 6 moves away from merely functional questions and addresses the case studies with a view to the compositional role of struts in highlighting expansive gestures and dramatic poses. Sometimes, struts function as abstract complements to the human body and framing devices. The use of struts counts among several strategies –​including typology, style, labelling, the representation of a statue base or architectural niche –​that the artist may exploit to establish the artificial nature of his figure. Especially in the case of sculptural copies, struts contribute to shifting the reference from the statue’s living counterpart to the art-​object in itself, the series to which it belongs and the prototype. As I argue in Chapter 7, the use of large and structurally unnecessary struts may also advertise the means for the conspicuous consumption of marble. These struts demonstrated the availability of marble and therefore the material value of a work of art. Furthermore, struts enable a variety of expressive and emphatic poses comparable to those of bronze statues and the figures in paintings or reliefs. The use of

xix

Preface

struts displayed the sculptor’s proficiency and ability to overcome the limits of his material. Chapter  8 examines regional and chronological concentrations. The number, size, placement, and shape of struts cannot provide, alone, any positive attribution to individual ateliers  –​let  alone to individual carvers. However, if the display context is known, these technical features may confirm attributions based on style and treatment. This chapter also discusses the function of struts in small-​size statues and miniatures. As in larger works, struts may serve to announce the sculptor’s skill in carving the details. Additionally, struts could be used as a more generalised allusion to full-​scale statuary. The presence of intricate struts in miniatures becomes increasingly significant in the sculptural production from the mid-​third century to the mid to late fifth century AD, reflecting the popularity of elaborate compositions. The Conclusion wraps up the main points of my argument. Struts had a general, undeniable structural utility. Nevertheless, they could also function as both allusions to a model and a testament to the carver’s ability. I  believe that, with all due precautions, non-​figural supports should be included among relevant stylistic features upon which to rely to examine the production, choice, exhibition, and viewing of sculpture in the ancient Roman world.

xix

xx

Acknowledgements

xx

The general idea and the main arguments of this book took shape while working on another essay on imitation in Roman art and thought (then published with the title ‘Difficillima imitatio’. Immagine e lessico delle copie tra Grecia e Roma, 2012). In that earlier work, my remarks on struts and supports in Roman marble statuary remained confined to a few lines of text and footnotes. Only as my thoughts developed into a self-​standing paper to be presented at the 22nd Theoretical Roman Archaeology Conference, held in Frankfurt in 2012, did it become apparent that none of these was sufficient and that the subject required a whole monograph. It would have to describe the types and forms of figural and especially non-​figural supports, illustrate their possible role in the construction and reading of images, and explain how this material could have been so widely undervalued throughout the history of classical scholarship. When one of the press’ anonymous referees commented that this ‘is really a German book written by an Italian in English’, I found their remark to be quite reflective of my experience during the research and the writing. This book is indeed the result of several years of work at a number of institutions, where I had the chance to discuss my ideas with many colleagues from different backgrounds, whose approach to the material was at times strikingly different from my own. At the time when I  first focused on supports and struts in Greek and Roman sculpture, I was affiliated with the School of Historical Studies at the Institute for Advanced Study in Princeton, as the Gladys Krieble Delmas Foundation Member (2011). There I had the rare privilege to commence my study in an equally challenging and secure environment. Further research for a catalogue of supports, struts, and props in Roman marble statuary has been made possible by a grant for early-​career scholars, generously awarded by my alma mater, the Scuola Normale Superiore of Pisa (Fondo Giovani Ricercatori). To both institutions go the warmest thanks; my thinking and learning have enormously benefited from their cosy and intellectually lively atmospheres. I wrote the majority of this book, however, in Munich where, from the winter of 2012 until summer 2015, I held a post-​doctoral teaching

xxi

Acknowledgements

position linked to the Centre for Ancient Worlds (MZAW) of the Ludwig Maximilian University. Leading the junior research group ‘Constructions of the Beautiful’, within the framework of ‘Distant Worlds: Munich Graduate School for Ancient Studies’, was an invaluable experience in terms of scientific growth and increased theoretical awareness. My gratitude goes to the faculty, fellows, graduate students, and staff of Distant Worlds, who have been a source of support and intellectual challenge throughout my research. Finally, the work of preparing the manuscript for the press was conducted at the Friedrich Alexander University of Erlangen–​Nuremberg, where I served as assistant for the academic year 2015–​16, and at the Getty Research Institute in Los Angeles. In the writing of this book, over the years, I have incurred a great many personal debts, and it is a pleasure to acknowledge some of them. Many colleagues and friends, along with the press’ reviewers, read earlier drafts of the manuscript at various stages with care and generosity, each improving it in distinctive ways. I am indebted to everyone who has read and commented on this text, especially to Lucia Faedo, Martin Hose, Sarah W. Lynch, Rolf Michael Schneider, Salvatore Settis, and Paul Zanker. For advice, criticism, and suggestions on various individual points of the research, impulses to and comments on my ideas, I am thankful to Ruth Bielfeldt, Clarissa Blume, Vinzenz Brinkmann, Hans-​Ulrich Cain, Petra Cain, Gabriella Cianciolo Cosentino, Amanda Claridge, Emily Cook, Patrick Crowley, Whitney Davis, Lorenzo Fatticcioni, Andreas Grüner, Felix Henke, Tonio Hölscher, Michael Koortbojian, Irving Lavin, Paolo Liverani, Jan Stubbe Østergaard, Annamaria Peri, Arne Reinhardt, Corinna Reinhardt, Thoralf Schröder, Verena Schulz, Lucia Simonato, Giovanna Targia, Cornelius Vollmer, William Wootton. Besides, I  am immensely grateful to Silvano Bertolin, Astrid Fendt, Michael Pfanner, and Peter Rockwell, who kindly shared with me some of their first-​hand knowledge about stone carving and workshop’s practices. Needless to say, any errors remain my own responsibility. Sections of this work were presented at conferences and workshops in Frankfurt (22nd Theoretical Roman Archaeology Conference) and Oxford (Classical Art Research Centre), and as lectures at the Universities of Basilicata (Matera, Scuola di Specializzazione in Beni Archeologici), London (Institute of Classical Studies), Leipzig, Heidelberg, Munich, Erlangen, Berlin (Humboldt University), as well at the Kunsthalle in Kiel and at the Getty Villa in Los Angeles. I am thankful to the audiences of all of these places for their questions and remarks that enabled me to focus on aspects that were previously underdeveloped and crystallise my thinking on various elements of this project.

xxi

xxi

xxii

Acknowledgements

A number of institutions and foundations met the substantial costs of visiting many of the museums and sites mentioned in the following pages and preparing the manuscript for publication. I  gratefully acknowledge the help of ‘Distant Worlds: Munich Graduate School for Ancient Studies’, the American Philosophical Society (Franklin Research Grant, 2015), and the Max Weber Foundation (Gerald D. Feldman Travel Grant, 2015–​17). A  generous grant from the Henry Moore Foundation (Small Research Grant, 2013–​14) covered the illustration costs and some of the travel associated with this project. Thanks are also due to numerous curators, who assisted during my long work of inspection and close examination of sculptural works in many collections of antiquities. Acquiring the illustrations included in this book would have been impossible without the collaboration of numerous colleagues, curators, and assistants. It is a pleasure to acknowledge the particular aid of Daria Lanzuolo (German Archaeological Institute, Rome), the staff of the German Archaeological Institutes at Athens, Istanbul, and Madrid, as well as of the Cologne Digital Archaeology Laboratory (University of Cologne). For their support in searching and obtaining pictures, my warmest thanks go also to my research assistants at the Scuola Normale Superiore, Elisabetta Stinco, and at the Getty Research Institute, Roselyn Campbell. My translations of the Greek and Latin authors owe much to those of the Loeb Classical Library, and at times –​as acknowledged in the relevant passages –​borrow directly from them. The bibliography makes no attempt to be comprehensive:  to have cited it all would have more than doubled the size of an already long list. Rather, I have tried to acknowledge those works to which I have been directly indebted for specific information or which arguments I have depended on explicitly, as well as those contributions that have played a relevant if only implicit role in the formation of my own views. This book was submitted for peer review in May 2016; citations of works published afterwards are selective.

xxi

newgenprepdf

Abbreviations

With few exceptions, the names of ancient authors and artists, as well as the titles of texts and artworks, are cited according to S. Hornblower and A. Spawforth (with E. Eidinow), The Oxford Classical Dictionary. 4th edn. Oxford: Oxford University Press, 2012. An author-​date scheme is used for all secondary references to works cited in the text. ABV

ARV2 ÄGB

Clarac CVA DNO

EAA Helbig4

LIMC LSJ

Beazley, J.  D. 1956. Attic Black-​Figure Vase-​Painters. Oxford:  Clarendon Press (reprint:  New  York, Hacker Art Books, 1978). Beazley, J. D. 1963. Attic Red-​Figure Vase-​Painters. 2nd edn. Oxford: Clarendon Press. Barck, K., Fontius, M., Schlenstedt, D., Steinwachs, B.  and Wolfzettel, F. (eds) 2000–​5. Ästhetische Grundbegriffe. 7 vols. Stuttgart: Metzler. Clarac, F. (comte de), 1839–​41. Musée de sculpture antique et moderne. Paris: Texier. Corpus Vasorum Antiquorum Kansteiner, S., Hallof, K., Lehmann, L., Seidensticker, B. and Stemmer, K.  (eds) 2014. Der Neue Overbeck. Die antiken Schriftquellen zu den bildenden Künsten der Griechen. 5 vols. Berlin: De Gruyter. Enciclopedia dell’arte antica classica e orientale. Rome: Istituto della Enciclopedia Italiana. Helbig, W.  1963–​72. Führer durch die öffentlichen Sammlungen klassischer Altertümer in Rom. 4th edn. 4 vols. Tübingen, Wasmuth. Lexicon Iconographicum Mythologiae classicae. 1981–​97. 8 vols. Zurich: Artemis. Liddell, H.  G.  and Scott, R.  1996. A Greek-​English Lexicon. 9th edn (revised and augmented by H.  S. Jones). Oxford: Clarendon Press. xxiii

xxvi

1

Introduction

In late November 1724 in Rome the sculptor Agostino Cornacchini was putting the finishing touches to his major endeavour, the equestrian monument of Charlemagne. The colossal group, commissioned by Pope Clement XI and completed under Benedict XIII, was to be placed in the atrium of St Peter’s basilica as a pendant to Gianlorenzo Bernini’s famous Constantine, created almost sixty years earlier (Fig. 1). An anecdote about the statue illustrates Cornacchini’s temperamental nature. In the story, as told by the satirist Ludovico Sergardi, Cornacchini was bitterly annoyed by the comments that many viewers made about his nearly completed Charlemagne, especially those referring to the short column that he had left under the horse’s belly. Would the group stand without it? Some believed this support was a testament to Cornacchini’s unskilful design and his inability to create daring yet stable compositions in stone. Offended by these remarks, which were triggered by the striking difference between his statue and Bernini’s rearing horse, Cornacchini secretly removed the support during the night at the risk of destroying his own work. To his critics’ surprise, the statue did not crumble and has since stood safely on the horse’s two slender legs.1 Regardless of this technical feat, Cornacchini’s Charlemagne has suffered criticism since the moment of its unveiling.2 In the shadow of Bernini’s dramatic monument to the first Christian emperor, Cornacchini’s sculpture, which intentionally lacks narrative and drama, has paled in comparison with what is considered a masterpiece of equestrian sculpture. Notwithstanding the apparent facility of Bernini’s audacious invention, he too had to resort to an expedient in order to stabilise his work, a sturdy bar-​ like strut between the rearing horse’s front legs. Bernini seems not to have made any effort to disguise the conspicuous marble brace, although this 1 Letter of Ludovico Sergardi to Giulio del Taja, Siena, Biblioteca Comunale degli Intronati, Autografi Porri, Busta XIV. The relevant passages are transcribed by Simonato (2005: 51). On the monument see Poeschel (2002–​3). 2 E.g. F. Valesio, Diario di Roma (Milan: Longanesi, 1978: vol. IV, 486). Rudolf Wittkower called Cornacchini’s Charlemagne ‘nothing but a weak and theatrical travesty of its counterpart’ (1999: vol. III, 56). Cf. Wittkower (1961).

1

2

2

Introduction

Figure 1  Agostino Cornacchini, Equestrian monument of Charlemagne, 1720–​5. Marble. Vatican, south end of the atrium of the basilica of St Peter

actually remains invisible from the main point of observation, the right-​ hand side of the figure.3 Cole (2007: 64). Images of the strut between the horse’s front legs are published in Bacchi and Tumidei (1998: 159, 161, 167) and Lavin (2012: 193 Fig. 167). Bernini’s portrait of Louis XIV on horseback, completed in 1684, demonstrates that stability represented a crucial challenge for monumental equestrian statuary. In Bernini’s project the rearing horse would rest on a mass of stone symbolising the Mountain of Virtue that only the worthiest few can ascend. See Lavin (1998: 206–​8 and Fig. 25) and Körner (2010: 259). 3

3

Looking at Supports

The challenge faced by the two sculptors is nothing new. At least since Graeco-​Roman antiquity, the problem of guaranteeing stability in marble statuary has often been addressed by the addition of supports. Especially in marble sculpture produced in imperial Rome, attributes and ancillary objects such as tree stumps, weapons, clubs, pieces of clothing, vases, dolphins and other animals often play a dual role as both supporting and compositional features; they reinforce the figure while providing information on its identity and qualities. In addition to these figural supports, in most Roman statues one or more non-​representational props in the form of bars, rods, or cylinders, with no apparent narrative function, contribute to stability by linking volumes and projections. For the sake of clarity, in the following chapters the latter will be called ‘struts’ or ‘non-​figural supports’ so as to distinguish them from the equally widespread –​yet radically different in their visual effect –​category of supporting attributes. Most scholarship has tackled the issue of supports and struts in Greek and, above all, Roman sculpture by focusing on the very same set of questions that seem to have been central in the judgement of Cornacchini’s Charlemagne by his contemporaries: are supports always necessary for a statue to stand safely? What do eminently technical measures such as supports betray or reveal about the nature of the material and the skills of a sculptor?

Looking at Supports The difficulties intrinsic to any attempt to answer these questions are exacerbated by our written sources’ apparent lack of interest in the technical challenges of marble carving. Whereas much is known, for instance, about Renaissance artists’ attitudes towards ingenuity, virtuosity, and their struggle against the limits of stone, the Greek and Latin literary sources only allow sporadic glimpses of the demands, material or technological constraints, and artistic choices that were daily practice in a Greek or Roman workshop of marble sculptors. In particular, we know almost nothing about the ancient view of supports or any other stability devices for statues. Figural supports are occasionally mentioned as integral parts of a composition, without any references to their structural function. One example is Apuleius’ fictional account of the house of Byrrhaena at Hypata, in Thessaly. This passage from the Golden Ass satirically describes a luxury dwelling in a city of Apuleius’ own times, the second

3

4

4

Introduction

century AD.4 Lucius, the protagonist of the novel, is astonished at the artworks displayed in the lofty atrium, in the centre of which stands a group of Diana and Actaeon in white Parian marble. As impressive as the image of the goddess alone may have been, striding forward with her garments blown up by the wind, Lucius’ attention seems to be caught by other details: a pack of fearsome dogs in the act of running and barking that flanked Diana on each side, ‘with their front feet ready to run, and their hind feet set firmly on the ground’. This description calls to mind the many second-​century AD statues of Diana in the same attitude, derived from classical or Hellenistic prototypes.5 The goddess, upset and in haste, storms forward in fluttering drapery, ready to shoot an arrow at her enemies; her faithful companions, deer or hunting dogs, rush forward at the same pace. These attributes hint not only at the goddess’ life in a natural environment and at a number of myths where her animal companions played a crucial role, but also offer structural support to the human figure in its lower section. Whereas the narrative role of attributes is central to the presentation of the group in Byrrhaena’s mansion, their static function remains obscure. In fact, no known text mentions statues being supported or in need of support, and it is left to our imagination what vocabulary could have been used to describe such features. In ancient Greek a ‘support’ could be variously called hypereisma, sterinx, or anteris. These words refer to an array of concepts –​from animal bones (as supports for the body) to reinforcements of buildings and the stay-​beams that strengthen a ship’s bow.6 A Latin writer from the late republican and imperial age of Rome would also have had a number of options available to refer to supports. Participle forms of verbs such as coniungere (‘to join’), fulcire (‘to hold up, brace’), supponere (‘to place beneath’) would highlight the static function of a support. Besides, other and more specific words existed that stressed one specific use, regardless of form: walking sticks (baculum or bacillum), poles for plants and branches (pertica, adminiculum, statumen), and masonry structures (fultura).7 None Apuleius, The Golden Ass or Metamorphoses 2.4 (tr. J. A. Hanson, Loeb Classical Library). 5 See E. Simon, s.v. Artemis/​Diana, in LIMC (vol. II, 805–​9 nos. 27–​36) for the ‘Versailles–​Tripoli Artemis’ and related types deriving from classical prototypes of the fourth century BC or later Hellenistic creations. 6 Bones as supports for the bodies of animals are called hypereismata by Aristotle (Parts of Animals 655a.10) and sterigges by Xenophon (On Horsemanship 1.5). Sterinx, in Diodorus of Sicily’s history, is a supporting masonry structure for towers (18.70.5). With the word anteris, Thucydides describes the stay-​beams fixed to a ship’s bow that support and strengthen the projecting catheads (7.36.2). 7 Fultura, in particular, was used in a variety of contexts with regard to built structures and machines (sometimes together with substructio, ‘foundation, supporting structure’): e.g. 4

5

Looking at Supports

of these, however, seems ever to have been used for the supports of statuary, whether figural or not. The ancient understanding of such structural or narrative additions remains a dimension to be almost entirely grasped through other methods, in primis the detailed contextual analysis of individual artworks. This oblivion of supports by the ancient authors may also explain the singular reticence of modern scholarship. The first scholar who explicitly –​yet cursorily  –​addressed the issue of supports was Ennio Quirino Visconti, papal Prefect of Antiquities in the late eighteenth century and a leading expert of his time in the field of ancient sculpture. This earliest assessment of supports appears in the short section of his Museo Pio-​Clementino in which he discussed the famous marble group of Ganymedes abducted by Jupiter’s Eagle (Fig. 2). Visconti examined the possibility that this small-​scale group derives from a fourth-​century BC bronze masterpiece, made by the Greek artist Leochares and mentioned by Pliny the Elder in his Natural History (34.79). According to Visconti the sculpture should, in fact, be considered the copy of some lost (Greek) bronze original, as indicated by the striking prominence of its support, a large tree trunk behind the Trojan prince and the Eagle. Supports, Visconti argues, are peculiar to Roman marble copies of Greek bronze originals that reproduce in heavy stone the loose and expansive postures of lighter prototypes made of metal.8 Ganymedes’ pose presents in itself major challenges for a work in marble. This is demonstrated by the measures taken by Vincenzo Pacetti, the neoclassical sculptor and restorer who recreated this piece around an ancient nucleus consisting of a human torso, the claws and neck of an eagle, and most of the tree trunk. To secure the statue’s balance, Pacetti had to include a slightly curved strut beneath Ganymedes’ right foot. However, the unusual size of the tree trunk is due to the fact that the figures were carved as decoration on a table leg. Visconti’s conjecture that supports point to the translation from bronze largely depended on recent achievements in the study of ancient sculpture. Vitruvius, On Architecture 6.8.3, 10.1.2, 10.16.11; Columella, On Agriculture 1.5.9; Pliny the Younger, Letters 10.39.2. In his account of the Civil Wars (1.54), Caesar mentions as statumina the ribs of a ship, made of light timber. 8 Visconti (1782–​96: vol. III, 65–​6 no. 49). On the statue (Galleria dei Candelabri, inv. no. 2445) see Lippold (1956: 216–​19 and Pls. 103–​4) and Spinola (2004, 173–​4 no. 83). Giuseppe Antonio Guattani made similar remarks about the struts (puntelli) of the Lancellotti discus thrower: ‘Egli [Pliny] lo [the Discobolus] annovera fra le statue di bronzo. Dunque dovrà dirsi una copia di quello. Che sia così, un tal soggetto eseguirsi non potea, che col porre un grandissimo puntello sotto il braccio destro … il che siccome dovea produrre uno svistamento notabile, non è da supporsi, che avrebbero tenuta simile idea, facendola di prima intenzione in marmo’ (1784: XII–​XIII = Guattani in Cancellieri 1806: 31–​2).

5

6

6

Introduction

Figure 2  Statue of Ganymedes abducted by the eagle. Marble. H. 103 cm. Vatican, Museo Pio Clementino, Galleria dei Candelabri, inv. no. 2445

During the eighteenth century it had become increasingly clear that the list of Greek masterpieces handed down by ancient literary sources, above all by Pliny the Elder, and the surviving statues from Rome were not one and the same. The simple observation that many images existed that repeated the same subject grew into the understanding that they may be copies after famous, lost masterpieces.9 Supports, Visconti believed, provide the most blatant indication that a given statue not only is a Roman copy of a lost Greek original, but also that the prototype was in bronze. Visconti’s explanation, however, did not immediately inspire broader accounts and comprehensive appraisals of supports. Over a century had to pass before figural supports would be examined in more detail, in a lengthy article published by Ada Maviglia in the Römische Mitteilungen of the German Archaeological Institute.10 Maviglia’s study mainly addressed 9 See Anguissola (2012: 25–​31) for the ‘discovery’ of Roman copies in the eighteenth century and Gallo (1992–​3) on Ennio Quirino Visconti and his father Giovanni Battista. 10 Maviglia (1913). The necessity of a comprehensive work on (figural) supports had already been explicitly advocated a few years earlier by E. Löwy (1905: 271).

7

Looking at Supports

questions of iconography and narrative coherence. Supporting attributes, she argues, generally conform to the character of the human fi ­ gure –​e.g. training tools are the prerogative of athletes, weapons of warriors and emperors, writing instruments and scrolls of intellectuals, whereas individual gods and characters of Greek and Roman mythology would be paired with their usual companions and attributes. Supports that had no apparent effect in characterising the subject, such as drapery, generic tree trunks, vases, pilasters, or columns, fall outside the scope of the study. The assumption is that figural supports are created explicitly to identify the subject. The ‘hermeneutic exercise’ attempted by Maviglia left a number of key questions unaddressed, especially about technology and chronology.11 In his 1926 doctoral dissertation and a larger and more ambitious book published twenty-​five years later, Fritz Muthmann, a student of Ludwig Curtius, set off from radically different premises and expanded the reach of Visconti’s assumptions.12 Muthmann argued that figural supports are not only reliable indicators of derivation but, because of their being extrinsic, they also play a crucial role in dating Roman copies of Greek statues. Copyists’ additions, he argued, reflect the taste of the period when they were carved. In contrast to Maviglia, Muthmann concentrates on shape and carving techniques rather than the narrative role of supports. He identified distinctive features in the treatment of figural supports, which should allow artworks created in the earlier imperial period of Rome to be distinguished from those produced in the Hadrianic and Antonine age. Although revolutionary for its time, Muthmann’s research is limited. It does not account for the traditions of workshops in the shaping of supports or for the possibility that support-​types also followed patterns of circulation similar to body types and iconographies. The reach of Muthmann’s landmark essay was such that over the following fifty years the issue of supports was largely neglected. Most research since has focused on two questions: how do figural supports match, explain, or specify the subject of a composition, and to what degree are certain attributes and their treatment indicative of chronology? 11 See Albizzati (1916: 386 n. 31) and Lippold (1923: 254 n. 1). In particular, C. Albizzati recommends for the first time joint consideration of supports (i.e. figural supports) and ‘puntelli’ (i.e. non-​figural supports). 12 In his dissertation, defended in 1926 and published one year later, Muthmann focused on a narrower choice of figural supports (tree trunks with or without additional attributes or drapery and palm-​tree trunks) from the Hadrianic to the Antonine period. The larger 1951 monograph includes earlier examples that reach back to classical Greek art, as well as a larger selection of shapes (pilasters, herms, tripods, weapons, vases, animals, and Cupids).

7

8

8

Introduction

Only in recent years has consideration expanded to include Greek sculpture of the classical and Hellenistic periods. Interest, understandably, has shifted from the relationship between Greek prototypes and their Roman copyists to questions about the development of the human body and the origins of leaning postures in Greek classical art.13 Such a change in perspective matches a parallel evolution in both Greek and Roman art history during the late twentieth century, which led, on one hand, to de-​ emphasising the role of copies for the history of Greek art and, on the other, to their ‘rediscovery’ in the sphere of Roman civilisation. Since the 1970s, historians of Roman art have underlined the importance of viewing images created according to Greek styles and iconographies as genuine expressions of Roman culture and values.14 Whereas the two lines of inquiry tackle a set of crucial (and complementary) problems  –​the dating and coherence of supports and the origins of leaning bodies in Greek classical art –​they both ultimately fail in addressing the visual and communicative dimensions of supports.

Functional Implements The questions raised by supports  –​related to necessity, technology, and ingenuity –​prove all the more poignant in the case of non-​figural supports, those that are generally called ‘struts’  –​‘Stützen’, ‘Stegen’, and ‘Streben’ in German, ‘tenons’ in French, and ‘puntelli’ or ‘tenoni’ in Italian.15 It seems that struts add little or nothing to the narrative. They respond to the practical need of securing projections to the main, generally vertical, axis of a marble figure. That this was a major concern to sculptors in stone in all periods is made clear in the words of Benvenuto Cellini, perhaps the most informative early modern writer on the craft of bronze and marble statuary. According to Cellini, ‘Extravagant attitudes’ and the ‘undercuts’ of waving 13 Schoch (2009), Weinstock (2012), Koçak (2013), are all derived from doctoral dissertations defended in German universities over a short span of time. Whereas K. Schoch and M. Koçak concentrate on well-​known classical types of Aphrodite, H. Weinstock aims to provide a broader account of both ‘dekorative Attribute’ (i.e. figural supports) and ‘die einfachen, ungeschmückten Stützen’ (i.e. struts). His analysis, however, is limited to the most widespread types of tree-​trunk supports, integrated by a limited selection of struts near figures’ legs. See also the review by S. Kansteiner (2013). 14 Anguissola (2012: 25–​66 and esp. 44–​57 for the recent debate). 15 In the scholarly literature, the vocabulary remains largely discretionary. ‘Strut’, ‘support’, and ‘prop’ are generally used as synonyms, like their German counterparts. The Italian ‘puntello’ may also refer to a different device, the stone projection that copyists used as a fixed point to take measurements.

9

Functional Implements

garments, with sections of marble projecting outward from the main surface or core, are the most challenging points to carve.16 The sixteenth-​ century painter Pontormo also noted that such passages were anything but peripheral. Rather, they constituted the achievements that most impressed viewers, even those who were not practising marble-​cutters and therefore unaware of most technical struggles. In his words, overcoming the difficultà d’un braccio in aria without damaging the stone is a paramount example of artistic ingenuity.17 In the study of Greek and Roman marble sculpture, these concerns for the difficulties of a composition and the dexterity of its carving have received little consideration. Instead, struts have been investigated mainly –​ if not exclusively –​as reliable cues for derivation. The point was explicitly raised by Rhys Carpenter concerning the statue of Hermes and the infant Dionysus at Olympia, the origins of which have been disputed since its discovery in the late nineteenth century. The horizontal bar strut connecting the polished body of the god to the massive tree trunk at his left side, ‘running brutally into the naked flesh’, excludes that the statue may be a Greek original carved by Praxiteles. On the contrary, it is ‘eloquent of the needs of the copyist afraid of his more brittle medium’, that is, marble.18 Rhys Carpenter sees the sophisticated system of supports as an intruder in an otherwise harmonious ensemble. The argument of derivation, with its corollary of assumptions about bronze and marble, is only one aspect of modern scholarship on ancient struts. Where it cannot be posited that a marble statue with struts is a copy of an earlier bronze work, then an alternative explanation has been put forward to frame struts within the dynamics of Roman marble workshops. Struts, it has been argued, were security measures for transport. According to this line of thought, struts indicate that a given statue was produced in a place distant from its context of display.19 Some have 16 Due trattati, uno intorno alle otto principali arti dell’oreficeria, l’altro in materia dell’arte della scultura. Florence: Valente Panizzi & Marco Peri, 1568: 57v (anastatic reprint Modena: Edizioni Aldine, 1983 = The Treatises of Benvenuto Cellini on Goldsmithing and Sculpture, tr. C. R. Ashbee, New York, 1976: 136). An unfinished statuette from Aphrodisias of the type called Versailles Diana offers an example of the type of sottosquadri which Cellini describes as points of potential fragility: see Rockwell (1991: 139 Figs. 21–​3). N. Penny (1993: 76) uses the same definition of ‘extravagant’ to describe the expansive poses of certain Roman statues, such as the Belvedere Apollo. 17 Letter to Benedetto Varchi, 1548, on which see Barocchi (1960–​2: vol. I, 67). 18 Carpenter (1931: 254–​5, 257). On this statue and the controversy about its large supports see Chapter 2. 19 So Lippold (1923: 43, 72–​3, 133–​4), Studniczka (1926: 142), Richter (1954: 31), Stewart (1977b: 89), R. Bol (1984: 21), Linfert (1979: 781). The two latter refer in particular to the

9

10

10

Introduction

proposed that struts were (expected to be) removed once the statue was in place.20 That struts would be eliminated once a statue was set up is an assumption suggested by early modern practices of marble carving. The corpus of marble sculpture from the sixteenth century onwards leaves little doubt that most stone connectors, bridges, and brackets used for the carving were later carefully amputated. This final touch was occasionally omitted, due to economic and practical constraints. One example is the statue of St Sebastian made by Nicolas Cordier in 1604 for the Aldobrandini chapel of Sta Maria sopra Minerva, in Rome. The hair, beard, and right foot were left unfinished and struts remain between the splayed fingers of the raised left hand. Cordier’s apparent indifference to struts is unlikely to be an aesthetic choice, but may rather depend on his disappointment with unsatisfying financial conditions. After Pope Clement VIII died in 1605, the work on his family chapel was interrupted and the artists were compelled to accept lower fees than agreed, something that may have prevented Cordier from putting the finishing touches on this statue.21 Even more conspicuous are the struts left on the statue of St Paul carved by Francesco Mochi for the Benedectine monks of Montecassino (1638–​ 52), who had commissioned twin images of St Peter and St Paul for the Basilica of San Paolo fuori le Mura. Rejected by the patrons, who refused to pay for them, the marbles were acquired by Pope Alexander VII after the sculptor’s death and found a place for display only years later, on the outer façade of the Porta del Popolo.22 A bulky, irregular strut runs from St Paul’s raised right hand to the drapery around his shoulder, about as large as the arm it is supposed to sustain. Four minor bridges connect other points of so-​called ‘neck struts’ or ‘nape struts’, on which see Chapter 4. Doubts about struts being evidence for transport are expressed by B. Andreae and B. Conticello (1987: 14 n. 38) who believe that, in the case of the sculptures at Sperlonga, struts were specifically required by the challenges of the compositions, with projecting sections and outstretched limbs. 20 This view had already been expressed by Ennio Quirino Visconti (1782–​96: vol. III, 36). Commenting on the sculptural group of Artemis and Iphigenia at Copenhagen (see Chapter 2), Bieber points out that ‘struts should be eliminated, as they were probably made for the transportation of the group and through neglect were not removed after the work had been set up’ (1961: 77). Studniczka believes that sculptors removed struts according to their customers’ preferences (1926: 142). Lippold (1923: 73) and Blinkenberg (1933: 23–​4) are sceptical about struts being removed once the statue was in place. The latter suggests that they were rather painted so as to pass unnoticed. 21 Montagu (1989: 45 and Fig. 50) and Cole (2007: 58–​60 and Fig. 3.2). Unfinished areas can also be detected on Cordier’s Charity for the same chapel. See also Pressouyre (1984: 377–​80 nos. 4–​5). 22 De Luca Savelli (1981: 80–​2 no. 23), Montagu (1989: 44–​5 and Fig. 48), Favero (2008, 93–​6 no. 27).

11

Functional Implements

the figure. Uncertainty about where the statue would go probably caused the reinforcements not to be cut away as customary, and they were left in place out of negligence. Various explanations have been put forward for several apparently unfinished works of Gianlorenzo Bernini’s mature years, that likewise include prominent struts, including the Veritas at the Galleria Borghese (1646–​52) or the portrait bust of Clement X (ca. 1676). It has been argued that, after a certain point, Bernini lost interest in devising virtuoso solutions for complicated passages, as his reputation no longer depended on his skills as marble carver. A seemingly non-​finito work may therefore be preferred to the risk of compromising stability or cracking the marble. Another ­interpretation is that major changes had occurred in Bernini’s workshop, which in this later period lacked a trusted expert to handle the final stages of production.23 More likely, the now elderly sculptor was not able to pay full attention to each of his numerous projects and progress on the individual works may have been erratic.24 Only occasionally in modern marble sculpture –​and significantly in a later period, when sculptors’ relationship to antiquity followed quite a different agenda than in the Renaissance and Baroque –​were struts left in place as a deliberate aesthetic choice.25 This is the case of the classicising statue of Diana carved between 1775 and 1778 by the English sculptor Joseph Nollekens for the gallery of Wentworth Woodhouse, the country seat of the Marquis of Rockingham, which included a collection of ancient Roman statuary.26 Nollekens depicted Diana running and shooting an arrow, 23 See Cole (2007: 60, 62, 64 with bibliography at 65 n. 4) on the possibility that it was Bernini’s gifted assistant, Giuliano Finelli, who finished the group of Apollo and Daphne and devised the ingenious solution for the struts in the form of laurel leaves between Daphne’s fingers. If this was indeed the case, Finelli’s departure would have imposed limits on the work that the workshop could deliver. 24 The documents seem to indicate that both statues still had not found a final place of display by the time Bernini died in 1680. On the history of the Veritas (to which Bernini planned to add an allegory of ‘Time as revealer of Truth’) and the portrait of Clement X (unfinished when the pope died) see Wittkower (1997: 276 no. 49, 298–​9 no. 78a), Beaven (2004: 96 Fig. 2, 110). An analysis of the carving techniques of the Veritas is in Coliva (2002: 235–​45). 25 One may think of the many, conspicuous struts included in Canova’s marble works such as his Perseus Triumphant in the Vatican Museums (Museo Pio-​Clementino, inv. No. 969) and the portrait of Napoleon in the guise of Mars at Apsley House in London. 26 London, Victoria & Albert Museum, no. A.5-​1986. See Penny (1993: 77 Fig. 70) and Baker (2000: 15–​17 and Figs. 7–​8). Nollekens executed four statues of goddesses –​Diana, Juno, Minerva and Venus –​for the second Marquis of Rockingham, who probably envisaged the group as part of a sculpture gallery for Wentworth Woodhouse. However, the inventory of the family’s house in London shows that the Diana was displayed alone. At any rate, all views that allow full appreciation of the goddess’ rush forward and twist prevent a clear view of the strut.

11

12

12

Introduction

in a twisted pose that he derived from an early seventeenth-​century bronze statuette of Cupid, with a long marble bar to join her outstretched arms. The sculptor, here, must have regarded the strut as legitimate, licensed as it was by the ancient practice. Placed beside antique statues, his Diana would fit into the context not only because of its subject and iconography but also by virtue of its technique. The availability of written sources such as technical treatises, commercial contracts, registers, and personal letters provides a wealth of information on the mechanics, customs, and expectations in post-​Renaissance marble sculpture. The paramount nature of certain basic concerns about the tensile strength of stone and the stability of a composition, however, should not lead us to apply this information a priori to the art of the Graeco-​Roman world. Presumably, Roman carvers were equally anxious about the solidity of their creations as Cordier, Bernini, Mochi, and Cornacchini would be centuries later. Complex poses in stone have always demanded structural assistance. In case of marble copies after prototypes in bronze, this limit must have been felt acutely.27 Nevertheless, in ancient Rome the ubiquity of struts on statues of all sizes and subjects, including miniatures that needed no additional support, excludes that their only purpose was to ensure stability. Even in large-​scale statuary, as the next chapters will argue, struts occasionally defy assumptions about their practical function. Further, struts could sometimes be carefully carved and polished. Especially in the second century AD, struts occasionally display decorative motifs. In fact, there is no evidence to suggest that Roman struts always and only served pragmatic purposes, that they were expected to be removed, or that their occurrence is the sign of a work left unfinished.

Invisible Accessories That issues of function and pragmatism dominated the debate is in fact largely due to the widespread denial that struts may sometimes be the result of aesthetic choices rather than of concerns about strength, r­ esilience, and balance. What has puzzled scholars most is indeed the v­isibility of struts, which defies modern assumptions about naturalism. Unlike ­figural supports, which may be perceived as integral to the composition, 27 Awareness of the different static requirements is clear in the sets of otherwise identical bronze and marble statues created by Jean-​Antoine Houdon in the second half of the eighteenth century. Whereas the bronze versions require no addition, marbles rely on large figural supports. See Körner (2010: 268–​9 Figs. 157–​8, 275 Figs. 164–​5).

13

Invisible Accessories

non-​representational struts stand outside all narrative purposes, apparently unrelated to the human figure and its attributes. The post-​Renaissance practice of marble sculpture teaches that external supports could be avoided by means of skilful cutting. That struts had not been avoided by most Roman carvers could therefore only be explained by postulating either that they copied lighter bronze originals or that they retained these meaningless devices for transport. The conventional notion that struts are indicative of copies and served only practical needs has resulted in their dismissal from the pool of aesthetically motivated devices. Because of their unnatural connection to the limbs, garments, and attributes of human figures and animals, struts have been considered ‘offensive to the aesthetic sense’ and evidence, in themselves, of inadequacy.28 Even scholars otherwise sensitive to the multi-​faceted phenomena of imitation in Roman art have responded negatively to the visual effect of struts. Struts have been variously dismissed in English as ‘inartistic’, ‘obtrusive’, ‘disfiguring’, ‘unseemly’, ‘unsightly’, ‘visually disturbing’, and ‘distracting’.29 German scholarship offers a similarly grim overview, with struts discarded as ‘unorganisch’, ‘störend’, ‘überflüssig’, ‘hässlich’, and ‘leblos’, thereby matching Italian hostility towards the ‘orribili puntelli’.30 In the light of this attitude, it comes as no surprise that scholars contrived a number of expedients to mend the shortcomings of the archaeological material. In the late nineteenth century, the archaeologist Georg Treu offered four methods to remove a strut from sight.31 Ideally, plaster casts of the statue could be made that expunge its strut. As an alternative, the strut could be removed from published pictures, thereby restoring a clean view of the artworks at least on paper. Where both measures are impossible, he recommends covering the offending element in existing photos 28 See Antonsson (1937: 26 and 85) on the group of the Hermes at Olympia. He denies that such a ‘demonstrably … obtrusive’ strut could be attributed to one of the foremost Greek masters, Praxiteles. 29 Casson (1931: 266) and Antonsson (1937: 85) for the strut on the group of Hermes at Olympia, Ridgway (1966: 35) with reference to the Palermo replica of the Pouring Satyr. See also Strong and Claridge (1976: 203), Waywell (1978: 21), Claridge (1988: 140), and Penny (1993: 76). 30 Connecting struts are ‘störend’ copyists’ additions for Lippold (1923: 73) and ‘unorganische Zutaten’ for Studniczka (1926: 142). It is again the strut between body and tree stump in the group of Hermes at Olympia that has caused the most severe judgements, e.g. in Treu (1897: 201) and Blümel (1927: 41). For similarly dismissive comments on other statues see W. Fuchs (in Helbig4, vol. I, 197), Yfantidis (1984: 77), Andreae (2001: 159). The ‘orribili puntelli’ attached to many replicas of the Three Graces are, in Becatti’s words, a ‘misero espediente del marmoraio per sostenere … il gruppo, turbando la semplice armonia della composizione’ (1937: 52). Martinez (in Pasquier and Martinez 2007: 272) calls the strut between the left hand and thigh of the Pouring Satyr in Palermo a ‘disgracieux pont’. 31 Treu (1897: 202) with reference to the group of Hermes and Dionysus at Olympia.

13

14

14

Introduction

with a piece of black paper. The last and more imaginative solution entails the restitution of the ancient lost colours, as the Dresden court painter Ludwig Otto had done a few years earlier in a water-​colour of the Hermes and Dionysus at Olympia. In Otto’s sketch, the strut between the body and trunk is painted brown like the tree stump and treated as an element of the natural landscape.32 All of Treu’s suggestions were put into practice. Most modern collections of plaster casts include statues stripped of their struts, much to the alleged benefit of viewers; these purport to offer an undisturbed glimpse of the original. The most illuminating example is probably the statue of the so-​called Apoxyomenos athlete (The Man using a Body-​Scraper), which was originally found in Rome’s Trastevere district and is now in the Vatican Museums.33 The Vatican Apoxyomenos, which is considered to be a mid-​ first-​century AD replica of a late classical bronze by Lysippus, shows the broken-​off stubs of a huge rectilinear strut from the right thigh to the waist. The strut even partly obstructed the view of the instrument that gave the statue its title: the strigil to scrape away sweat and dust after exercise. A multitude of plaster casts after the Apoxyomenos exist in museums and university collections throughout Europe and the United States. Often, the remains of the huge strut were eliminated from the mould employed to create these plaster statues.34 Restorers of ancient sculpture seem to have shared this paramount concern for the visibility of struts by choosing, if possible, not to replace broken connectors. When Pietro Tenerani, in the mid-​nineteenth century, restored the newly found Apoxyomenos following the directions of the archaeologist Luigi Canina, he chose not to replace the long strut between wrist and thigh, of which only the smoothed stubs still remain.35 Only a minor and almost invisible strut between the left wrist and chest, instead, may have been difficult to omit for reasons of stability and 32 Knoll (1994: 174 no. 161a). 33 Museo Pio-​Clementino, Gabinetto dell’Apoxyomenos, inv. 1185. 34 The strut has been entirely or in part erased from the plaster copies now in Munich (Museum für Abgüsse Klassischer Bildwerke, inv. 136), Berlin (Freie Universität, inv. 90/​33), Moscow (Pushkin Museum, inv. II.1.И.556), Dresden (Albertinum, see Bildwerke des Altertums 1953: 111 no. 189), Austin (Blanton Museum of Art, no. 2004.12, old inv. DL–​90). Similar procedures have been tried on a number of other ancient statues, copied in casts that expunge their ‘obtrusive’ struts. E.g. the nineteenth-​century casts of the Boy Strangling a Goose in the university collections at Bonn (Akademisches Kunstmuseum, inv. 67) and Leipzig (Antikenmuseum, inv. G 608). See also Michaelis (1901: 35 Fig. 39) for a plaster cast in Strasbourg (cat. no. 743) of the Cnidian Aphrodite at the Vatican (Museo Pio Clementino, inv. 812) without the strut between body and vase. 35 E. Braun (1850: 238) criticised the decision to remove this ‘accessorio … di cui gli antichi non si vergognavano’.

15

Invisible Accessories

was accordingly replaced.36 Nonetheless, the restored strut was significantly thinner than the stubs of its ancient counterpart –​a concession to the modern distaste for supports.37 The second solution recommended by Treu is definitely less demanding, yet alarmingly deceptive. Struts, he advises, should be deleted from the pictures included in scholarly publications. This suggestion relies on the established practice, from the centuries that preceded the invention of photography, of omitting struts from engravings after the antique.38 Whereas earlier reproductions without struts seem not to have been based on careful examination of the pros and cons of visualised supports, Treu’s advice is part of a detailed programme for the scholarly presentation of ancient sculpture. In particular, he had in mind the ‘disturbing’ strut found in the group of Hermes and Dionysus at Olympia. For this work, he recommended the reproduction that his colleague Heinrich Brunn had included in an article in Deutsche Rundschau five years earlier. This image, which depicted the statue alone, without the stone bar between body and attributes, was presented to the educated readers of the German periodical.39 Scholarly literature of the last century provides many examples of statues being pictured to hide their struts. Occasionally, double plates offer alternative views with and without struts in the hope of clarifying crucial stylistic aspects by releasing the statue from any extraneous additions.40 Treu’s third suggestion regarding hiding struts, that is, attaching a small piece of black paper to photographs with dark backgrounds, has also found modern advocates. The effect that Treu had in mind was probably similar to 36 P. Liverani (in Moreno 1995: 201 no. 4.29.4, 205 no. 4.29.9). 37 For a similar example of ‘understated’ restoration, see the statue of Poseidon from Caesarea Mauretaniae –​Cherchell (Algiers, National Museum of Antiquities, inv. 4) pictured in Landwehr (2000: Pl. 49c). 38 See the plates from Clarac’s Musée de sculpture, such as the double view of the Apoxyomenos (vol. V, Pl. 848B no. 2168a). Cf. vol. IV, Pl. 634 no. 1428 for the Capitoline group of Marcus Aurelius and Faustina without its spiral struts. In other cases, however, the image includes the struts: e.g. vol. III, Pl. 273 no. 1572; vol. IV, Pl. 569 no. 1214a; vol. V, Pl. 875 nos. 2227 and 2232a. See also Langl (1887: Pl. 22). 39 Brunn (1905: 380 Fig. 46 = ‘Der Hermes des Praxiteles’, in Deutsche Rundschau 31, 1882: 188–​205). 40 Rizzo (1932: Pls. 21–​2, 71–​2). A more recent example is in Boardman (1995: 100 Fig. 84), one of the most widespread handbooks of Greek sculpture. In order to illustrate the famous Dresden Artemis, an image is chosen where the figure lacks the curved strut between hip and elbow. The credits do not clarify whether this is an amended picture of the statue or the photo of a restored torso of the same type, e.g. the one presented in Gercke and Zimmermann-​Elseify (2007: 87–​8 no. 17, esp. Fig. 17.5). Occasionally, the same attitude applies to post-​classical sculpture: e.g. the image of Bernini’s David (1623–​4) in Coliva (2002: 164).

15

16

16

Introduction

the strategy that the Danish archaeologist Neils Hannestad recommends for the display and illustration of the Sperlonga groups of Polyphemus and Scylla (Fig. 3). The whole concept behind these dramatic sculptures, he argues, ‘is blurred in the diffuse daylight’ of the museum that fails to obscure the many struts. In order to enhance both the viewing experience and the scholarly usefulness of photos, the struts should be ‘sealed off by wrapping them with black velvet’.41 It is no surprise that the modern bias on the visual role of supports has affected not only how ancient sculpture is presented, but also the way its features, contexts, reception, and historical circumstances are accounted for. Scholars have generally assumed that the ancient Romans themselves shared their dislike for struts. It is a commonly held opinion that painting helped to remove struts from view. In line with Treu’s fourth piece of advice, modern attempts at the restitution of ancient colour conform to this assumption.42 This presumption of invisibility has had far-​reaching consequences in the study of individual settings and architectural contexts, whose interaction with artworks is thought to have assisted in concealing their struts. For instance, it has been supposed that the many struts of the Scylla and Polyphemus groups at Sperlonga remained effectively disguised in the dark of the grotto. The Blinding of Polyphemus, in particular, could hide its more numerous and bulkier struts because, sunk as it was in gloom at the rear of the cavern, they became virtually unnoticeable for those gathered in the triclinium.43 Two considerations, however, may challenge this conjecture. First, the suggestion that the Scylla, placed in the middle of the pool, had noticeably fewer struts than the other group is questionable. Secondly, it is hardly conceivable that the dim atmosphere would be sufficient to conceal struts often larger than the limbs they sustain, while simultaneously allowing the human bodies and their minutely carved details to remain visible –​ unless, of course, we assume that struts were either painted a dark colour or wrapped in black cloth. 41 Hannestad (1989: 657). 42 For the Kassel Apollo, which was the object of one of the first academic exercises in the study and recovery of ancient polychromy in the early 1990s, see Gercke (1991: 183–​4 no. 48). Based on the traces of colour detected on the Kassel statue, as well as on parallel evidence from other replicas, the experiment aimed at recovering both the image of the putative fifth-​century BC prototype and that of the Roman copy as it appeared in its own times. In the example of a plaster cast painted in the manner of a Roman marble statue, only the struts and the support in form of a tree stump are left white, as neutral appendages that disappear against the white background of the published photographs. 43 Conticello (1974: 22, 47) and Stewart (1977b: 89).

newgenrtpdf

17

Figure 3  Group of the Scylla from Sperlonga, detail of the boat and helmsman. Marble. L. max preserved of the helmsman 160 cm. Sperlonga, Archaeological Museum

18

18

Introduction

Occasionally, the presence of bulky struts has affected the discussion on the display context and the modes of viewing of a given composition.44 A good example is the three identical replicas of the Hellenistic group of the Boy Strangling a Goose, found together in the Villa of the Quintilii near the Appian Way. Unlike other copies of the same composition, the statues from the Villa of the Quintilii are supported by a sturdy pillar-​like strut under the bird.45 This group, it has been suggested, should be viewed from a particular angle at the right side of the young boy that would ensure the full concealment of the strut behind his right leg.46 This perspective enhances the effect of the pyramidal composition. At the same time, however, it prevents complete appreciation of several other narrative and stylistic features: the dynamic twist of the child’s upper body and his dishevelled coiffure, the tension of the animal, the struggle between the squeezing arm and upraised wing, the detailed treatment of the goose’s quills and feathers. In the case of the statues from the Villa of the Quintilii, all sharing the same structure, it seems unlikely that they were arranged following the same standard. Rather, their display as decoration of one or more fountains or nymphaea followed the subtle and more compelling principle of variability, with different views exhibited and concealed in each individual piece.47 One further point may contribute to the discussion of struts in the Villa of the Quintilii statuary. All of the statues found in this luxury residence of the Roman suburbium present a distinctive technical feature; the strut, throat, and beak of the goose concealed a pipe so that water could flow from the goose’s mouth.48 Therefore, the strut was essential for these statues in their capacity as fountains. They were either equipped with a strut so that a channel could run through it, or chosen from a body of other replicas because 44 A discussion of this point is provided also in Hollinshead (2002a: 118). Examples of this approach are Conticello (1974: 22, 35, 47) for the groups at Sperlonga and Schröder (2004: 233) for the Hypnos statue in Madrid (Prado, inv. 89–​E). For comments about the subject and angle of display of the figure known as the Conservatori Charioteer (Rome, Centrale Montemartini, inv. 988) see La Rocca (1987: esp. 31–​6 and Pl. XIV) and Germini (2008: 105–​15 and 161 cat. 1, esp. 109–​14). For the Sandalenbinder from the South Baths of Perge (Antalya Museum, inv. 3.25.77), J. Inan (1993: 114) has postulated an arrangement that would keep the strut behind the figure’s neck invisible, assuming that it is identical with the angle of display of the putative bronze original. This idea entails a double contradiction. First, the assumption that struts were a structural necessity cannot be reconciled with the belief that their position could be determined by display. Secondly, it implies that those who produced the copy were aware of the original’s exact setting. 45 Munich, Glyptothek, inv. 268; Paris, Louvre, inv. MR 168 (Ma 40); Musei Vaticani, Galleria dei Candelabri, inv. 66. A list of replicas is in Kunze (2002: 142–​3 n. 791 nos. 1–​3); on the location in which these works were found, see Neudecker (1988: 194 no. 39.19). Andreae (2001: 159) adds to these a replica in Geneva (Musée d’Art et d’Historie, inv. 8944) on which see Chamay and Maier (1990: 37–​8 no. 37 and Pls. 45–​6) and Kunze (2002: 142–​3 n. 791 no. 4). 46 Andreae (2001: 159). 47 Neudecker (1988: 194 no. 39.19) and Schädler (1992: 211 and 215 Fig. 17). 48 Kunze (2002: 149 n. 823). On the water channel in the Vatican replica also Lippold (1956: 325–​7 no. 66). For a comparable solution –​yet far more meaningful on a narrative

19

Seeing Supports

of the strut that allowed for a channel to be pierced through it. As ‘disturbing’ as they may be, struts were evidently more acceptable to the villa’s owners than visible water pipes behind or beneath marble statues.

Seeing Supports The hermeneutical problem at the root of all these approaches is that of accounting for features that are not integral to the artwork’s narrative. The perspectives and explanations that I have described are deeply indebted to the distinction between the intrinsic and the extrinsic employed in Kantian aesthetics. The section on the so-​called ornaments (Zierarten), or parerga, is today among the best known and most contentious of his Third Critique. Under the label of parerga, Kant understands a feature ‘which is not internal to the entire representation of the object as a constituent (Bestandstück), but only belongs to it externally as an addendum’ –​for instance, ‘the borders of paintings, draperies on statues, or colonnades around magnificent buildings’.49 Parerga border or complete the work but are not an intrinsic part of it –​they belong to the work while being subsidiary to it. In his classic response to this passage, Jacques Derrida takes issue with Kant’s distinction between essential elements and external complements, that is between the ‘inside’ (ergon) and the ‘outside’ or ‘subsidiary’ (parergon) of the thing perceived.50 Because of their structural reciprocity, he explains in The Truth in Painting, the essence of the former can only be communicated by means of the latter. Since parerga are essential to the presentation and perception of the erga, Derrida finds that it is ultimately impossible to decide where a parergon begins and where it ends. All the examples of parerga provided by Kant are features defined less by their exteriority to the work in itself than by their fundamental structural connection to it. This holds particularly true in the case of the columns, which cannot be considered as inessential surplus but are ‘most difficult to detach from the work’ (the ergon) and have (like the supports of sculptures) an ‘internal structural link’ to it. Neither Kant nor Derrida are speaking exclusively about the interpretation of artworks. Derrida’s reading of Kant, in particular, is concerned with the framing of philosophy and the emergence of the level –​see the statue of Ulysses at Baiae (Chapter 3). Cf. the Dancing Faun from the Kerameikos at Athens and its support (Chapter 1). 49 Kant’s Critique of Judgement (§ 14) is cited according to the text edited by P. Guyer for Cambridge University Press (2000: esp. 110–​11). For an introduction to Kant’s aesthetics see McCloskey (1987) and more recently Wenzel (2005) and Hughes (2010). 50 See Derrida (1987: 53–​64) for the comments on Kant’s distinction between erga and parerga. For a discussion of Kant’s passage and Derrida’s response see Carroll (1987: 131–​54), Harvey (1989), and Rodowick (1994).

19

20

20

Introduction

Figure 4  Statue of Melpomene from the Baths of Faustina at Miletus, early Antonine period. Marble. Detail of the neck strut. Istanbul, Archaeological Museum, inv. 1993

idea of the aesthetic as a specific area of philosophical inquiry. Nonetheless, this discussion is of the utmost importance in the study of the visual arts and specifically in the context of my argument. The issues of parergonality, extrinsicality, and margins that it raises seem most applicable there, for supports in sculpture do not have to be treated metaphorically –​their parergonality is a physical attribute of the artwork. A number of examples, however, challenge the idea that Roman statues would always have been displayed in such a manner as to conceal their struts –​ their parerga. One example may clarify this point. A slightly less than life-​size female figure identified as the muse Melpomene, found in the Baths of Faustina at Miletus, features a prominent neck support from her shoulders to the edge of the chignon which, however, did not prevent the head from breaking off (Fig. 4).51 As it was probably placed in a niche, the statue must have shown little or nothing of the neck strut. Nonetheless, the frontal arrangement highlighted a much more conspicuous device; an exceptionally robust strut, almost as large as the arm itself, curves in an arc from the Muse’s right midsection to the underside of her elbow (Fig. 5).52 51 See von Gerkan and Krischen (1928: 112–​13 no. 14), Manderscheid (1981: 95 no. 220), Schneider (1999: 8, 19–​20, 33, 77–​88 no. 1). 52 Geominy (1999b: 59 Appendix VIII), Hollinshead (2002a: 147–​8 and Fig. 6.20).

21

Figure 5  Statue of Melpomene from the Baths of Faustina at Miletus. H. 151 cm

22

22

Introduction

That such a massive support, oversized in relation to both the body and the arm, finds no parallel in other statues of the same type, seems to say more about deliberate aesthetic choices than about essential static needs. Could the arm have been secured by other methods? Would the joining of a separately carved limb present any advantages in labour and costs? I argue that struts were not always hidden and negated in the display and viewing of statues, but rather that their visibility and prominence were negotiated case by case.53 The diverse and often conflicting criteria for this negotiation will be investigated in the following chapters, by focusing on artworks, technology, and modes of viewing. Although the history of struts and supports goes as far back as the history of stone sculpture itself, the core of material that I discuss dates to the last decades of the Roman Republic and the first two centuries of the imperial period. It is then that both figural and non-​figural supports became a ubiquitous hallmark of statues of all sizes, subjects, qualities, and contexts. I  begin my study with a re-​examination of the ideas of the few scholars who, in recent years, have attempted to underline the potential of struts regarding the visual impact of a statue and the modes of its appreciation.54 The late Wilfred Geominy, who cursorily tackled the issue in a broader article about the Roman art of imitation, described non-​figural supports on Roman marbles as indicators of faithfulness.55 Geominy generally agrees with the traditional interpretation that struts were first and foremost required to translate a lighter bronze prototype into heavier stone. As universally recognised aides to the technical process of copying, struts would broadcast fundamental information such as that the supported statue was derived from a (Greek and ancient) bronze archetype and that the latter is reproduced faithfully down to the smallest detail. Struts would inform the viewer both of the sculptor’s commitment to accuracy and highlight his proficient handling of the medium. What then are we to make of extensively propped statues with no known or presumed prototype? Geominy conjectures that oversized struts may have enhanced a statue’s prestige by hinting at a fictional Greek original.56 53 Claridge (1988: 140) notes that ‘Roman eyes … appear to have accepted struts just as readily as they accepted joins’. P. Liverani (in Moreno 1995: 201) makes similar comments on the Vatican Apoxyomenos. 54 See also Anguissola (2013) on the arguments laid out by Geominy and Hollinshead as well as possible responses. 55 Geominy (1999b: 49–​51 and 56–​9 Appendices I–​VIII). 56 So Geominy (1999b: 51) on the strut under the raised arm of the helmsman from the Scylla group at Sperlonga.

23

Seeing Supports

In her landmark chapter on non-​figural supports in Roman marble statuary, which is to date the only systematic treatment of the subject, Mary Hollinshead explores an alternative reading of struts. Hollinshead acknowledges that marble statues often demanded structural assistance due to their large size or the complexity of their poses.57 In some cases, this was the consequence of copying a bronze statue. However, provided that unusually large struts ‘demonstrate how important a particular pose is to the marble carver’, the source needs not always to be a bronze prototype. Rather, shafts and bars may have been instrumental in expanding the sculptor’s choice of models and translating two-​dimensional designs inspired by table, wall, and vase painting, reliefs, and mosaics into marble. Struts, Hollinshead argues, ‘demonstrate resourcefulness in adapting compositions in other media to marble statuary’.58 Both Geominy’s general and Hollinshead’s more detailed analyses suggest that struts indicated some sort of derivation, either from a bronze or two-​dimensional model. More precisely, both agree that struts were employed to visualise such derivation and the carver’s facility. This argument provokes a chain of further considerations. It presupposes that struts are the mirror of a broader and more complex system of comparisons and discussion about the relative merits of materials (bronze vs. marble) and media (painting vs. sculpture). While addressing the problem of how struts may have allowed sculptors to translate prototypes from other materials and media, in the following chapters I will also focus on the fundamental issue of why such visual translations were relevant to those who made, purchased, viewed, and commented on art in the ancient Roman world, and why was marble so distinctive that sculptors strove to convey it by a variety of supports. This perspective allows us to reconsider the questions that have been raised in this introduction. If supports and, in particular, struts were not only accepted but sometimes even meant to trigger a chain of comparisons and appraisals about matters and media, the whole functional discourse may be reversed. What if, in other words, sculptors sometimes chose ‘extravagant’ poses 57 Hollinshead’s article, published in 2002, provides a concise presentation of the main typologies of struts and offers a wealth of insightful observations on their function and reception (for the passages quoted here see Hollinshead 2002a: 148, 152). 58 Given the poor state of documentation on ancient Greek painting (not to mention ancient drawing), it remains difficult to determine how generalised the use of two-​ dimensional models might really have been. Hollinshead (2002a: 150) correctly remarks that ‘arguing for precise prototypes in Greek wall paintings no longer extant presents an argument ex silentio comparable to that of the “lost Greek bronze original” sculpture’.

23

24

24

Introduction

precisely because of their actual or alleged need for supports? In this light, the question of whether struts are always necessary for a statue to stand safely is less relevant. Crucial, instead, is the role of technical measures which reveal the challenges of the material and the qualities of the individual craftsmen.

25

Part I

Material and History And now we come to the isolated segments –​that is, those carved parts that stand detached from the main figure all but for one single point of contact, with the result that they are almost hanging in the air. Imagine a statue with one arm stretched out and with fingers spread wide – obviously, if you carved only by chiselling these small isolated segments, you would run the risk of breaking them under the blows of your mallet. (Adolfo Wildt, The Art of Marble, 1921; tr. by the author)

25

26

27

1

Narrative and Individuality

The scholarly lack of interest in external supports has deep roots in the post-​Renaissance understanding of ancient marble sculpture. Only during the eighteenth century did scholars conjecture that most ancient statues known at the time were actually Roman copies after lost Greek prototypes.1 In his account of ancient sculpture at the Museo Pio-​Clementino, published towards the end of the century, Ennio Quirino Visconti postulated that supports in Roman statues indicate that they are copies after Greek bronze originals. Until that moment, the question whether or not supports were integral to the composition seems to have been largely ignored.2 Nonetheless, the presence and quality of supports on ancient marbles had already made problems for practitioners. Such is the case, for instance, of the Tuscan sculptor and mint master Massimiliano Soldani Benzi, one of the finest bronze casters of the late seventeenth and early eighteenth centuries. He was commissioned by the Prince of Liechtenstein to supply copies of selected statues from the Medici collections in Florence. In particular, he was asked to cast bronze replicas of two famous ancient marbles:  the so-​called Medici Venus and the Dancing Faun. Scholars later recognised that both are Roman works derived from Greek prototypes that likely did not include the supports attached to the Roman copies. The Medici Venus is supported by a tree stump and a dolphin with two tiny Cupids; the Dancing Faun, who steps with his right sandal on a metal clapper (kroupezion), also incorporates a tree stump support. Soldani Benzi, although unaware of the distinction between ‘Greek originals’ and ‘Roman copies’, took the task of faithfulness quite seriously. His patron had in fact urged him to carefully reproduce the ‘originals’, i.e. the marble statues in Florence, as they were. The letters between the two illustrate Soldani Benzi’s careful consideration on how to proceed with supports, which undoubtedly belonged to the ancient marble models and yet seemed not entirely fitting for his new bronze creations. As See Marvin (2008: 16–​167) and Anguissola (2012: 25–​30), both with earlier literature. 2 Körner (2010: 270–​3) notes the ambivalence towards supports before the ‘discovery’ of Roman copies. 1

27

28

28

Narrative and Individuality

a result, he opted for an individual strategy; he retained those supports that he regarded as integral to the composition –​the dolphin and Cupids, as well as the percussion instrument –​while excluding the apparently meaningless tree stumps.3 Obviously, the narrative content is the aspect of figural supports that has attracted the most discussion. In this chapter, I briefly address this dimension and discuss how figural supports often include elements useful to describe and situate the figure or hint at a storyline. In addition to this, I argue, figural supports sometimes contribute to a composition’s structure in terms of its symmetry, movement, balance, and visual axes.

Description and Narrative The Medici Venus was probably the first artwork to be recognised as a Roman copy after a Greek original. In the chronicle of his travels to Italy, the English man of letters and politician Joseph Addison (1672–​1719) reported the existence of many sets of ancient statues ‘made after the same design’, including those of subjects ‘such as had no relation, either to the interest or devotion of the owner’. These, he argued, must be the copies of some ‘celebrated masterpieces’. Above all, the large number of statues of the same ‘air, posture and attitude’ of the Medici Venus attests that this series derives from the best and most appreciated masterpiece of antiquity.4 Clearly, Addison is here referring in the broadest sense to the whole set of statues in the ‘Pudica gesture’ which included, notably, the sculpture presented to the Capitoline Museums in 1752 (best known as the Capitoline Venus) and the Medici Venus installed in the Tribuna of the Uffizi in 1688. Scholars today consider all these to be copies after Hellenistic creations that ultimately derive from the famous Cnidian Aphrodite made by Praxiteles. The Medici Venus, in particular, is thought to have been carved in the second half of the first century BC, as suggested by the lettering of its much-​discussed artist’s signature and, more reliably, by its support. Dating is in fact based on the general similarity between the goddess’ support –​a little Cupid on a dolphin –​and Soldani Benzi’s bronze casts are now at the Liechtenstein Museum in Vienna, inv. Sk541 (Dancing Faun, 1695–​7) and Sk537 (Medici Venus, 1699–​1702). See Zikos (1996: 132 and 134–​7 Figs. 8–​15), as well as Lankheit (1962: 326–​38) on the correspondence with the Prince of Liechtenstein. 4 Addison (1705: 339–​41), on which see Barbanera (2006: 6–​7; 2008: 38–​9), Marvin (2008: 122–​ 3), Anguissola (2012: 25–​6). The history of the reception of the Medici Venus is summarised in Haskell and Penny (1981: 325–​8). 3

29

Description and Narrative

that placed next to Augustus’ right leg in his famous portrait from Prima Porta.5 Similar questions have been raised by the conspicuous support belonging to Venus’ companion in the Tribuna, the Dancing Faun. Is the large tree stump next to the satyr merely a copyist’s addition required to stabilise the marble replica of a bronze prototype –​or is it integral to the compo­ sition since the satyr needs to prop himself while striking the foot clapper?6 Indeed, if we broaden consideration beyond the Florentine statue to embrace the whole replica series, there is much more that can be said about these satyrs’ supports. Two copies of the Dancing Faun, both dated to the mid to late second century AD, may clarify this point.7 The tall and twisted tree trunk that supports the satyr now at Palazzo Corsini in Rome seems to be much more than a neutral stability device (Fig. 6).8 Rather, its upward spiral enhances the movement of the figure’s dance step and provides a contrast to the shoulders bent to the left. A further iteration of the Dancing Faun found in the Kerameikos at Athens offers yet another solution (Fig. 7).9 The palm of the satyr’s left hand rested on the goatskin-​draped trunk, where the dancer has hung his pan-​pipes (syrinx). These objects define the figure’s character and his environment. In a manner similar to the Corsini Satyr, the trunk follows and enhances the forward bend of the figure. The long strut that runs from the right hip to the hand or arm must have stressed the satyr’s dynamic pose, while at the same time continuing the line of the branch that stretches between the trunk and body; a lower branch extends to connect to the satyr’s left knee and, in the form of a shorter strut, to his right leg. Furthermore, the support provided a cunning solution to the statue’s use as a fountain figure: a water pipe passed through a hole pierced in the back of the trunk and poured out of the goat’s mouth. These case studies raise many questions. Were supports merely a means to stabilise the marble copy of a composition conceived in bronze? Or were they also expedient in articulating new meanings, regardless of the alleged prototype? And if we accept that figural supports should be viewed as expressive elements in their own right, should we concentrate on their 5 Havelock (1995: 77–​8); on the support see also Muthmann (1951: 27–​8). 6 Habetzeder (2012: 151) considers the support integral to the narrative and therefore likely included in the prototype. 7 Geominy (1999a: 148–​9) attempts a parallel examination of the supports of the Corsini and Kerameikos statues, in order to reconstruct the intended viewpoint of both the original Hellenistic statue as well as the Roman versions. 8 Rome, Galleria Nazionale d’Arte Antica di Palazzo Corsini, inv. 710. See De Luca (1975: 77–​9 and Pls. 37–​8). 9 Athens, Kerameikos Museum, inv. 8071. On the support see Muthmann (1931).

29

30

Figure 6  Statue of a Dancing Satyr, Antonine period. Marble. H. 146 cm. Rome, Galleria Nazionale d’Arte Antica di Palazzo Corsini, inv. 710

31

Figure 7  Statue of a Dancing Satyr from the Kerameikos, late Hadrianic or Antonine period. Marble. Athens, Museum of the Kerameikos, inv. 8071

32

32

Narrative and Individuality

attributive function (i.e. the skin and pan-​pipes hanging on the trunk of the Kerameikos Satyr, which characterise the subject and environment) or consider with equal attention their contribution to the visual impact (i.e. the trunk itself, which highlights the whirling dance step) and function (the hole in the trunk that transformed the satyr into a fountain) of a statue? These questions, I  argue in the next chapters, should be addressed even more urgently with regard to non-​figural supports, which have largely been dismissed by scholars as straightforward ancillary elements. To this purpose, a closer look at figural supports, whose narrative role in the compo­ sition is more obvious, may help to lay out the complex of issues at stake. An analysis of figural supports provides a useful basis from which to introduce questions and concepts that are crucial in the assessment of struts. Supports of statues from the Roman imperial period, scholars have observed, often serve as attributes that identify the character and set the action in context. Ancient writers also understood that attributes were vital in the effort to create personal identities and set the character in a narrative. A passage from Philostratus’ Images offers an explicit explanation for the necessity of artists to use the correct attributes (phasmata) for their subjects.10 In the case of the god Dionysus, appropriate phasmata were an ivy crown, horns, or a leopard –​devices that enabled even images of the poorest workmanship to succeed in identifying their subject. Attributes were fundamental both in vesting the sculpted figure with an identity, defining the sphere of action (descriptive or situative attributes), and hinting at a storyline (narrative attributes).11 Attributes could function as regulators for the gaze and guide viewers in the process of viewing. The passage from the Golden Ass that I cited in the Introduction demonstrates that ancient viewers understood the identifying and narrative functions of figural supports. The hunting dogs that surround a statue of Diana in the fictional house of Byrrhaena function as both attributes and supports for the striding figure. Drawing on the considerations raised by the example of the Dancing Faun, the next sections explore these functions of supports, which I call ‘descriptive’, ‘situative’, and ‘narrative’, and make the case for a further raison d’être –​as complements to the human figure that enhance, individualise, or modify its movement. 10 Philostratus, Imagines 1.15.2 (p. 316). Accounting for the meanings of the term phasma (‘apparition’, ‘phantom’, ‘phenomenon’, ‘portent’, and ‘omen’ are only some of the translations provided by LSJ) exceeds the scope of this book; I refer on this to Platt (2011: 14, 238, 255, 257, 318–​19) in the context of divine epiphanies. 11 For the concept of descriptive attributes see von den Hoff (2004: 122 and n. 89), as well as Stähli (2003: 252–​8) and Giuliani (2013a) on the broader categories of ‘descriptive’/​‘narrative’. Cf. also Schade (2007: 177–​8).

33

Supports for Names

Supports for Names Marble carvers of the Roman imperial period did not invent supports, nor were they the first to employ them as descriptive implements. The sixth-​ century BC sculptor who carved the famous over-​life-​size limestone statue of a Picene warrior, found in an ancient cemetery near the town of Capestrano, seems to have been aware of the potential of supports as an additional setting for attributes. The Capestrano Warrior stands squarely between two rectangular pillars, which taper upwards and end below the figure’s armpits. Without these pillar-​like supports, the top-​heavy figure would lose its balance. At the same time, supports provide a location for the warrior’s equipment, which includes a spear sculpted in low-​relief on the outer side of each pillar. Besides, an inscription which is thought to name the person depicted runs vertically on the front of the right support.12 While essential to the statue’s balance, the pillars provide biographical details about the subject. Scholars of Roman art have long understood that supports may constitute an ideal placement for attributes. As early as 1913 Ada Maviglia hypothesised that attributes used as supports would generally correspond to the statue’s subject. The problems with her model arise in cases where supports do not meet the expectations raised by the main figure, as for example, the palm tree trunk, which is considered typical of athletes but is occasionally used for figures lacking any reference to the concept of victory. These contradictions are attributed to the Roman copyists, who are assumed to have misunderstood either the human figure or its support.13 Obviously, this exercise leads to conclusions that must necessarily remain speculative, as it does not take into account the variety of solutions employed by Roman carvers. Nevertheless, the question that it raises remains relevant. What hermeneutical tools can be employed to account for the relationship between figural supports and the figure that they sustain? Whereas Maviglia sees figural supports as extrinsic yet unavoidable elements that had to be adjusted to the figure, I argue that they could be employed by artists as bearers of meaning in their own right that triggered a process of recognition and understanding. 12 Chieti, National Archaeological Museum of Abruzzo, inv. 26117. See Holland (1956: 247), Cianfarani (1968: 13 Figs. 16–​17), Basile (1993: 11–​13 and Figs. 4–​5). The skilled use of supports conflicts with the dismissive judgement by H. Weinstock (2012: 11), according to whom they attest to ‘die provinzielle Unbeholfenheit des Bildhauers’. 13 See Maviglia (1913: 21–​5). The example of the palm tree trunk is revealing. The palm is seen as a symbol of victory and therefore appropriate for athletes. Its use to support the statue of a Pouring Satyr in Berlin (Antikensammlung, inv. Sk 257) is explained as a misunderstanding, due to the fact that this satyr wears a fillet around his head similar to that of athletes. Instead, F. Hauser (1906: 279 n. 1) hints that artistic reasons may have led copyists to choose a palm tree trunk as a support, to enhance the effects of light and shade against the nude body.

33

34

34

Narrative and Individuality

This use of struts appears most clearly in replica series. Supports may in fact play a significant role in bestowing a specific identity on conventional body types. In the many statues that employ classical or classicising ‘stock bodies’, identity was often carved into the auxiliary support.14 A famous example is a statue from Delos in the pose of the Diadoumenus by Polyclitus, the youth who ‘binds the fillet’ around his forehead as an act of victory. Unfortunately, we do not know what the subject of the original bronze statue by Polyclitus (cast around 420 BC) might have been. Greek and Latin literary sources seem oblivious to the real or fictional character whom Polyclitus had sculpted –​a god, a hero, or a winner in the Panhellenic games. More important to the Romans were the aesthetic and moral qualities connected to his figure. An extraordinarily evocative power was vested in the Diadoumenus, as well in Polyclitus’ other iconic statue, the Doryphorus, both of which came to represent rhetorical concepts in the work of Greek and Latin authors. Like Myron with his Discobolus (the ‘discus thrower’) and Lysippus with his Apoxyomenos (the ‘the man using a body-​scraper’), Polyclitus was the author of images whose gesture had become synonymous with the statue itself. The Greek title and the gesture combined to create both a visual and conceptual unity that indelibly fixed those statues in the minds of Roman viewers. The identity of the subject seems to have played a much lesser role.15 Nonetheless, at least one later copy of Polyclitus’ fillet binder seems to have attempted to give the youth a name and place. The exceptionally fine Diadoumenus found on Delos counts among the earliest surviving faithful copies of classical masterpieces (Fig. 8).16 No earlier statue, to our knowledge, attempts the precise, stylistically consistent scale-​ reproduction of a classical work that we see in the Delian Diadoumenus, carved in Parian marble at the turn of the second and first centuries BC. Unlike later copies, the Delos Diadoumenus makes almost no use of auxiliary supports. The figure, with its raised arms and the long fillet running 14 The point is raised by M. B. Hollinshead (2015) with reference to a statue at the Ny Carlsberg Glyptotek (inv. 1801), thought to reproduce Polyclitus’ Discophorus. Its support provided scholars with arguments for both a date and the figure’s identity. Its style suggests a date in the mid-​second century AD. The dragon (ketos) carved on the front has prompted identification as Perseus, who famously fought a sea-​monster to save Andromeda, e.g. by Zanker (1974: 6 no. 3) and Kreikenbom (1990: 145 no. I.7). Moltesen (2002: 305–​8 no. 102) suggests that the animal is instead an attribute of the god Neptune. According to Hollinshead, the fact that the support is incomplete reveals the sculptor’s selling strategy: the figure would be finished according to the wishes of the customer, who could choose the subject. 15 For the association between certain Greek masters and Roman visual semantics see Hölscher (2004: esp. 47–​57, 93, 97 on Polyclitus’ style). On the meaning of the Diadoumenus see Settis (1992) and Anguissola (2014a: 120–​1). 16 Athens, National Archaeological Museum, inv. 1826. See Kreikenbom (1990: 188 no. V.1) and, on the context, Nolte (2006: 284–​7).

35

Figure 8  Statue of the Diadoumenus from Delos, ca. 100 BC. Marble. H. 195 cm with the plinth (186 cm without the plinth). Athens, National Archaeological Museum, inv. 1826

36

36

Narrative and Individuality

from head to hands, was masterfully carved so as to stand safely without any assistance. The elaborate 80  cm tree trunk next to the young man is almost detached from the body and is of little consequence to the statue’s overall solidity. What we see here is nothing like the conventional palm tree trunk, generally placed adjacent to the youth’s left leg, which is characteristic of most copies of the Diadoumenus.17 The Delos statue’s stump creates an outdoor setting for the figure and holds its attributes. The naked youth’s mantle is placed on the trunk, secured by its upward branches. Casually hanging from a lower branch is a quiver, which identifies him as an archer. It is possible that, whatever the subject of Polyclitus’ original might have been, this particular copy was consciously turned into an image of the archer god Apollo, native of Delos and worshipped there in a famous Panhellenic sanctuary. That the god should be depicted as a victorious athlete fits in this context, since Delos had hosted famous festivals and games from earliest antiquity. The knotty, twisted trunk may, in addition, hint at the holy olive tree that Greek sources record near Apollo’s temple and which was thought to mark the god’s birthplace.18 Interestingly, recent analyses proved that the statue was entirely gilded, including its support, thereby reinforcing a connection with the myth of Apollo and Delos. The support not only provides a name for the subject, but strengthens the link between the body type and its (new) display context. In other words, the support both describes the image and situates it in mythical and historical space.

Supports for Stories Supports do much more than provide information about the subject. Sometimes, they also aid the reconstruction of a specific narrative related to the figure depicted. A story is, to a certain degree, implicit in any attributes that identify the subject. We may wonder, for instance, whether the Delian Apollo diadoumenos has just defeated an archer who foolishly challenged him and speculate about possible links to some of the god’s competition myths. I will illustrate the narrative potential of figural supports with two statues –​one mythical, the other historical in content. 17 F. Muthmann dated the support –​and therefore the statue –​to the second century AD (1927: 21–​ 3), but he later revised his dating to the late Hellenistic period (1951: 23–​4). On the statue’s support see also Weinstock (2012: 108–​10). F. Hauser (1906: 281) highlights the prominent role of the support and the difference in shape from the stumps typical of Hellenistic and Roman statuary. 18 For the identification of the Delos statue as Apollo see also O. Palagia, s.v. Apollon, in LIMC (vol. II, 242–​3 no. 468), as well as Kreeb (1988: 157 cat. S7.1), P. Jockey (in Marcadé 1996: 82 no. 31) and Weinstock (2012: 109). Hafner (1961: 205–​8), instead, considers both the Delos fillet-​binder and its fifth-​century BC prototype as images of Theseus, himself a boxer and the mythical founder of the Delian games.

37

Supports for Stories

Figure 9  Statue of Dionysus with a personification of grape vines (Ampelus), second century AD. Marble. H. 158 cm. London, British Museum, reg. no. 1805,0703.1 /​no. Sculpture 1636

My first example is a statue of Dionysus found near Rome and now at the British Museum, formerly in the collection of Charles Townley (Fig. 9).19 The youthful god is naked except for the leopard skin covering part of both shoulders and the breast as a sort of chlamys (cloak). Ivy crown and leopard –​Dionysus’ typical phasmata –​leave little doubt about the youth’s identity. With his left arm, Dionysus embraces the figure of a little boy, whose body seems to sprout from the supporting tree trunk placed next to the god. The boy wears a crown of vine leaves; vine leaves and grapes also spring from his cheeks and body. He offers a bunch of grapes to the god and the two exchange tender gazes. The boy’s body terminates in a vine plant, at the base of which lies a leopard, with a collar of ivy round its neck, biting a bunch of grapes. In this case, the support provides an expedient solution for the problem of depicting a specific myth about Dionysus, namely the story of Ampelus, the young satyr whom the god loved deeply and who 19 London, British Museum, reg. 1805,0703.1 (no. 1636). See M. A. Zagdoun, s.v. Ampelos, in LIMC (vol. I, 690 no. 1) and on the location of the find (La Storta) Neudecker (1988: 213 no. 54.6).

37

38

38

Narrative and Individuality

was transformed, after his untimely death, into the first grape vine.20 The statue seems to mirror the metamorphosis as described in detail by the late antique Greek poet Nonnus in his Dionysiaca, as the body changed, his belly was a long stalk, his fingers grew into top-​tendrils, his feet took root, his curl-​clusters were grape-​clusters, his very fawn-​skin changed into the many-​coloured bloom of the growing fruit, his long neck became a bunch of grapes, his elbow gave place to a bending twig swollen with berries, his head changed until the horns took the shape of twisted clumps of grapes.21

The Townley statue evokes the dimension of time, which in Nonnus’ account builds to a dramatic climax, by visualising Ampelus’ dual natures, his life as a boy and his afterlife as a grape vine. The exercise of reconstructing the narratives visualised by supports has also been attempted in the case of images with historical content. Much has been said, for instance, about the support of the over-​life-​size cuirassed statue of Augustus from Prima Porta.22 This marble image of Augustus clothed in a breastplate epitomises our collective ideas about the princeps and Roman art in the age of Augustus. Archaeologists have long recognised that the scenes depicted on the breastplate probably refer to a specific historical event of major political significance, perhaps the recovery in 20 BC of the Roman standards captured by the Parthians. It has been argued that the support next to Augustus’ right leg, a winged Cupid riding a dolphin, may also contribute to the statue’s message by adding further relevant information. Roman audiences might have understood this as a genealogical claim, referring to the gens Iulia’s alleged descent from Venus.23 Although a mortal, Augustus towers over the figure of his divine relative. Furthermore, some scholars believe that the toddler is in fact a portrait ‘in forma Cupidinis’ of Gaius, Augustus’ grandson and appointed successor. Thus, the divine child would suggest both the dynasty’s mythical past and its bright future.24 The dolphin too may embody a message that pertains, at the same time, to myth and history.25 On the one hand, it strengthens the link to Venus, by referring to the goddess’ birth from the sea. On the other hand, as an attribute of the god Neptune, the dolphin alludes to Augustus’ naval victories –​above all, the landmark success at Actium. As is well known, marine themes featured prominently 20 The narrative quality of this statue’s support is highlighted in Schade (2007: 187–​9). 21 Nonnus, Dionysiaca 12.173–​87 (tr. W. H. D. Rouse, Loeb Classical Library 1962). 22 Vatican Museums, Braccio Nuovo, inv. 2290. 23 The ideological and political implications of Augustan genealogy are discussed in Zanker (1988: 193–​210, 215–​27). In particular for the support of the Prima Porta statue see Squire (2013: 267). 24 See Studniczka (1910: 50–​5), Simon (1957: 64–​5), Pollini (1987: 41). 25 The meaning of supports in the shape of dolphins is examined by Piwowarczyk (2000).

39

To Each Copy its Support

in Augustan self-​styling and propaganda and became a distinctive feature of Roman imagery in the last decades of the first century BC.26 Myth and history overlap in this support which, together with the scenes on the breastplate, bestow on the image a narrative dimension encompassing Augustus’ ancestry, his deeds, and his legacy.

To Each Copy its Support Figural supports play a key role in describing or situating the figure and suggesting a narrative. A  further aspect, however, emerges. Supports can contribute to the aesthetic qualities of a composition by defining features such as balance, symmetry, and the juxtaposition of vertical and horizontal lines. This is, once again, most visible in the case of a replica series, where comparison between the nearly identical –​and yet strikingly different –​pieces is possible. The three full-​size versions of a statue of Silenus cradling the infant Dionysus, whose prototype has been ascribed to the fourth-​century Greek master Lysippus, are illuminating in this respect. The old, bearded leader of the satyrs is depicted in a moment of tender intimacy with the divine child, delivered into his care after birth. In order to balance his muscular body in the act of lulling the little god to sleep, Silenus leans on a tall tree trunk with his left elbow. All three copies, now in the Vatican Museums, at the Glyptothek Munich and the Louvre, exploit their support to provide information on the characters and context.27 Vine leaves, shoots, and bunches of grapes wrap around the trunk, on top of which a goatskin is placed. The copies in Munich and at the Vatican Galleria Chiaramonti, both formerly at Palazzo Ruspoli in Rome, are so similar that it has been suggested that they must be either the products of the same workshop or of ateliers working in close contact during the first century AD (Fig.  10). In particular, they share the almost vertical construction of Silenus’ figure, which follows the same line as the tree trunk. The composition is thus enclosed in a thin rectangular frame. The sculptor’s preoccupation with stability is evident in the decision to add two shorter struts which fasten Silenus’ left leg to the trunk. The slightly earlier Paris statue, found in the gardens of Sallust in Rome, develops a different concept, which entails a complex distribution of masses and weights between the human body and its support (Fig. 11). Although the statue has been extensively reworked in the modern period 26 On marine themes as related to the victory at Actium see Zanker (1988: 82–​5 and passim). 27 Vatican Museums, Galleria Chiaramonti, inv. 2242; Munich, Glyptothek, inv. Gl. 238; Paris, Louvre, inv. Ma. 922. See Moreno (1995: 252–​5 nos. 4.37.1–​3) and, on the Munich statue and the replica series, Vierneisel-​Schlörb (1979: 446–​56 no. 41).

39

40

Figure 10  Group of Silenus cradling Dionysus, first century AD. Marble. H. 199 cm with the plinth (191 cm without the plinth). Munich, Glyptothek, inv. Gl. 238

41

Figure 11  Group of Silenus cradling Dionysus, Flavian period. Marble. H. 190 cm. Paris, Louvre, inv. MR 346 /​Ma 922

42

42

Narrative and Individuality

and is in fact a complex mix of marble fragments and plaster, the position of the feet, which has not been altered, shows a pronounced off-​balance pose. This results in a large triangular negative space between the two main vertical masses. Notwithstanding the distance between body and trunk, the copyist apparently decided not to add any struts to join the two, so as to make their juxtaposition more striking. This decision had a two-​fold effect. First, it substantially increased both the technical difficulties in carving a deep, large undercut as well as the quantity of stone needed, as this compo­ sition requires a broader block of marble. Secondly, it resulted in an image in which the support is as prominent an element as the human figure itself, fundamental to the sculpture’s stability, both physical and visual. As a result the figure, whose weight is displaced across a diagonal line, offers multiple ideal viewpoints from different sides, whereas the Vatican and Munich copies encourage a static, frontal view. This discourse becomes crucial when discussing the art of Praxiteles who, in the mid-​fourth century BC, developed an array of compositions in bronze and marble in a trademark off-​balance pose, relying on auxiliary supports. It is sufficient to take a look at the marble versions of the Apollo Sauroctonos or Lizard-​Slayer to understand the variety that was available to copyists in devising the relationship between human figure and support.28 Similar considerations apply to the sculptural type of the Resting Satyr, which is also considered to be derived from a bronze statue cast by Praxiteles, and of which almost 130 replicas of both full and reduced size are known to date.29 Whereas in most full-​size specimens the satyr leans on a tall, smooth tree trunk, the support sometimes takes the form of a vine or serves as hanger for attributes such as pan-​pipes and hunting stick. The distance between the satyr and his support, too, differs greatly, and with it, the degree to which his body is bent.30 In some versions, the support repeats the body’s curved profile; in others, its plain vertical shape contrasts with it.31 Comparable variability affects the miniature copies. Although the support is part of the prototype, it seems to have emboldened the copyists’ creativity. Supports have been skilfully employed both to locate the character in

28 See Preisshofen (2002) and Martinez (2007a). 29 Lists of replicas can be found in Gercke (1968: 22–​58), Martinez (2007b: 258–​9) and Corso (2010: 123–​32 n. 221). Bartman (1992: 51–​101) provides a detailed discussion of the series with a list of miniature copies. 30 The issue of the distance between the body and its support has played a major role in the scholarly debate abut Praxiteles’ lost prototype, e.g. Gercke (1968: 59–​60) and Vierneisel-​ Schlörb (1979: 356–​7). 31 See Geominy (1999b: 52) and Bartman (1992: 53–​4).

43

To Each Copy its Support

a sylvan setting and to complete the composition as either echoing or contrasting forms. In these cases, supports provided a means to enhance the quality of each artist’s own version of this Praxitelean model. This argument has even broader implications with reference to another category of supports, those known as non-​figural supports or struts, which have so far resisted almost all attempts at interpretation by means of ­traditional iconographic and stylistic analysis.

43

44

2

Supports and Greek Marble Sculpture

Scholars have seen both figural supports and non-​representational struts as distinctive devices of Roman stoneworkers. Contrary to this expectation, however, supports and struts are recorded in Greek sculpture from the sixth century BC. They also appear in many Roman statues for which we have no Greek antecedents. The body of surviving original Greek marble sculpture from the classical period includes a limited number of freestanding statues, vastly outnumbered by reliefs and pedimental figures, which could be fixed to the tympanon wall by means of tenons or dowels.1 Occasionally, struts appear also in pedimental compositions, where multiple figures had to be fastened to each other or outstretched limbs secured to the main figure. The first arrangement is found in the symmetrical side groups from the late fifth-​century BC pediment of the temple of Marasà at Lokroi. Here, an abundance of tiny structural supports, which were not visible to the viewer on the ground, connected the figures of each group.2 Elsewhere, struts supported outstretched limbs.3 More often, however, expansive positions were achieved by skilful management of carved limbs and garments. This is the case with one of the figures from the Amazonomachy pediment of the Temple of Apollo Sosianus in Rome, considered to be a Greek original from the second half of the fifth century BC.4 The best preserved sculpture of this composition is a forward-​leaning statue of Theseus whose advanced left leg is supported by a piece of cloth sliding from his thigh.

44

1 See the comments by Richter (1931: 288–​9) on the body of surviving Greek sculpture. For technical remarks about the installation of pedimental sculptures from the Temple of Zeus at Olympia see Younger and Rehak (2009: 52–​3, 69–​70, 74–​80 and Figs. 5, 10b, 21a–​c, 22a–​b, 27a–​c, 28a–​b, 29a–​c, 37a–​b, 38a–​b). Remarks on the supporting features of classical pedimental sculpture are also in Weinstock (2012: 21–​4). 2 Reggio Calabria, National Museum, inv. 89538–​9. See Costabile (1995: Pls. 2.2 and 5.1, Group A). 3 E.g. in a fragmentary male torso from the fourth-​century BC pediment of the temple of Athena Alea at Tegea (Tegea-​Alea, Piali, Museum, inv. 1309), on which see Dugas (1924: 94–​6 no. 23 and Fig. 37, Pl. 104b) and Stewart (1977a: 19–​21 no. 13 and Pls. 9c, 11a–​c). 4 Rome, Centrale Montemartini, inv. 3529. On this set of marbles see La Rocca (1985) and Bertoletti (2006: 53–​60).

45

Fastening Movement

Notwithstanding the limited samples available for examination, struts seem to have been a familiar tactic for Greek stone carvers of the archaic and classical periods.5 Although struts only became a ubiquitous feature of stone sculpture from the first century BC, their use in bracing complex compositions and limbs detached from the core has a long tradition in the Greek sculpture.

Fastening Movement One type of support in particular seems to have emerged in the late archaic period and remained in use ever since; from the early fifth century BC large supports began to be incorporated into freestanding equestrian marble statues.6 Additionally, archaic kouroi often include struts. Unlike most later Hellenistic and Roman statues, no kouroi have struts between their legs to reinforce the calves. Instead, struts are used to fix the free-​ cut arms to the body. A  distinctive evolution in the shape of the struts between clenched hands and body accompanies the progression towards less rigid postures.7 Earlier Attic kouroi from the turn of the sixth century BC, such as the so-​called kouros of the Sacred Gate or the slightly later New York kouros, occasionally display narrow segments of stone left in place between hand and thigh. The Sounion kouros, discovered in 1906 in a deep pit near the temple of Poseidon at Sounion and dating to ca. 600–590 BC, also has its arms slightly detached from the body between armpit and hand, but these For an early enumeration of struts in Greek sculpture see Studniczka (1926: 140–​1) and more recently Hollinshead (2002a: 122–​6). On pilasters and other supports for sculpture in ancient Egypt and the Near East see Weinstock (2012: 10–​12) with bibliography. For a close parallel to the stone curtains left between the arms and body of archaic Greek kouroi cf. the free-​standing statue of the Egyptian court official Snofrunefer, ca. 2400 BC (Vienna, Kunsthistorisches Museum, inv. AE-​INV-​7506). 6 E.g. a fragmentary group of horse and rider from the Acropolis (Athens, Acropolis Museum, inv. 700), on which see Eaverly (1995: 110–​15 no. 12). The support bears traces of red paint. On the ‘unsightly’ belly supports of the Chariot Group from the Mausoleum of Halicarnassus see Waywell (1978: 17 Fig. 2, 21). For comparable evidence from the Antikythera shipwreck see P. C. Bol (1972: 84–​9 nos. 92–​5 and Pls. 50.4, 52, 53) and E. Vlachogianni (in Kaltsas, Vlachogianni, and Bouyia 2012: 112–​13 no. 61). For Roman parallels cf. Bergemann (1990: passim). 7 The outstretched forearms of archaic korai, the female counterpart of the kouros male youth type, did not require any struts for stability. As Claridge (1990: 137–​42 and Figs. 2–​4, 6a) explains, the arms were carved separately and provided with a tenon, which was inserted into a corresponding mortice with metal dowels. Drapery was often extended in order to secure the lowered arm to the body. See Richter (1968: Figs. 340, 365, 393). 5

45

46

46

Supports and Greek Marble Sculpture

are fastened to the upper thighs by narrow screens of stone.8 Later kouroi, instead, have arms that stand almost free from the armpits to the wrists. Connectors are made often in the shape of short bars, similar in both function and treatment to the ubiquitous ‘Roman’ struts. The statue of a youth in Parian marble from the mature archaic period, found at Attic Myrrhinous (modern Merenda) together with the Phrasikleia kore, preserves traces of two short, thin struts between hips and palms.9 Two semi-​c ylindrical struts connected the upper thighs to the now missing hands of the Aristodikos kouros, which epitomises the transition from late archaic to early classical sculpture. This element highlights the dynamic movement of the youth and the novelty of his pose (Fig. 12).10 The plasticity of the muscles and the movement of the arms, bent at the elbows, as well as the hairstyle of short shell-​like curly locks, place this statue, which was executed in Parian marble and found in the Attic interior, at the end of the kouroi series in the years around 500 BC. The next phase in the history of struts is illustrated by the marble statue from the Athenian Acropolis known as the Kritios Boy, which seems to embody the stylistic transformation of freestanding statuary in the first quarter of the fifth century BC. Although smaller than life-​size, this statue employed struts in the form of oval bridges to support the movement of the youth’s arms.11

The Invention of Leaning Bodies The use of struts and supports remains fairly marginal throughout the fifth century BC. Most freestanding marble statues seem not have required any external support, relying instead on the arrangement of limbs and garments 8 Respectively Athens, Kerameikos Museum, inv. 1700; New York, Metropolitan Museum of Art, inv. 32.11.1; Athens, National Archaeological Museum, inv. 2720. See Richter (1970: 41–​2 nos. 1–​2), Kaltsas (2002: 39 no. 17), Niemeier (2002: 40–​6), Brüggemann (2007: 169 no. 180, 199 no. 299, 201 no. 305). 9 Athens, National Archaeological Museum, inv. 4890 (ca. 540–​530 BC). See Kaltsas (2002: 49 no. 46) and Brüggemann (2007: 208 no. 331). Cf. the slightly later kouros 12 from the Boeotian sanctuary of Ptoon Apollo, dated to ca. 530–​520 BC (Athens, National Archaeological Museum, inv. 12, 2005) on which see Richter (1970: 122–​3 no. 145), Kaltsas (2002: 62 no. 80) and Brüggemann (2007, 161–​2 no. 142). 10 Athens, National Archaeological Museum, inv. 3938. See Richter (1970: 139 no. 165), Kaltsas (2002: 66 no. 94), Maderna-​Lauter (2002: 227–​9) and Brüggemann (2007: 206–​7 no. 326). 11 The debate on date and style is summarised by J. M. Hurwit, who also comments on the struts (1989: 61–​2 and 78–​80, on the struts 71, 80).

47

Figure 12 Aristodikos kouros, ca. 510–​500 BC. Parian marble. H. 198 cm. Athens, National Archaeological Museum, inv. no. 3938

48

48

Supports and Greek Marble Sculpture

alone. Nonetheless, attributes with a supporting function became commonplace. For example, Heracles’ club and Asclepius’ rod became essential elements of their iconography. The image of Aphrodite, too, shows a significant evolution during the last decades of the fifth century. This resulted in a broader choice of designs, such as leaning poses that relied on figural supports.12 In a limited number of cases, the prominent position and structural function of supports have raised questions about the narratives and the visual impact of compositions initiated in bronze or other materials lighter than stone. The Wounded Amazon of the so-​called Sciarra-​ Berlin-​Lansdowne type, which is known from multiple Roman marble versions of the first and second centuries AD, is a good example.13 The statues in this replica series, which are thought to derive from a bronze original from the third quarter of the fifth century BC, depend on a support for both their balance and the arrangement of limbs.14 Exhausted and wounded under her right breast, the mythical warrior rests her left elbow on a post. This support enables her to shift her weight to the right leg, while her right hand is raised to her head as if about to faint. Baffled by the prominence of this support, scholars occasionally suggested a later dating of the prototype to the Hellenistic or Augustan age, which seemed more compatible with this ‘virtually unnecessary’ implement.15 Others, however, have stressed the narrative role of the pillar and argued that, while stabilising the composition, it may represent a boundary marker of the sanctuary of Artemis where the wounded warrior sought shelter.16 As a pars pro toto, the pillar sets the character in its mythological context. The Varvakeion Athena statuette, regarded as the most faithful –​ or rather the only complete –​surviving copy after the chryselephantine statue of Athena Parthenos erected for the Parthenon has posed similar questions.17 Athena extends her right arm forward, supported by a 12 Schoch (2009) and Koçak (2013) explore this issue with regard to the most relevant sculptural types. 13 Hiller (1976: 37–​40). Similar questions have been raised by another late fifth-​century BC type, the so-​called Narcissus, who also leans on a pillar. 14 A summary of the scholarly debate is in R. Bol (2004: 145–​58). For a list and discussion of the replicas see R. Bol (1998: esp. 35–​49, 173–​87 for the Sciarra-​Berlin-​Lansdowne type). 15 According to B. S. Ridgway, the Sciarra-​Berlin-​Lansdowne type is a classicising creation, inspired by fifth-​century prototypes but executed in the first century BC with a ‘virtually unnecessary support, unprecedented for bronzes before the Hellenistic period’ (1974: 7–​8, 17). 16 On this reading see Hafner (1961: 210), Arnold (1969: 60 n. 249), Stewart (1990: 162). See also León (2008: 249). 17 Athens, National Archaeological Museum, inv. 129. See Karanastassis (1987: 408–​10 no. BI.12) and Kaltsas (2002: 104 no. 187).

49

The Invention of Leaning Bodies

column. A Nike stands in her palm, about to take off. Although ­scholars have often argued that a support was necessary to sustain the Nike of the statue created by Phidias, others consider such a static solution unlikely to originate in a fifth-​century BC Greek sculpture and attribute the ­addition to the Roman copyist.18 Figural supports seem to have represented a meaningful yet marginal phenomenon in Greek freestanding sculpture of the fifth century BC, and the use of non-​figural struts remained exceptional.19 Beginning in the fourth century BC, the work of Praxiteles seems to have initiated a new phase in the development of struts. In a number of types attributed to this prolific Athenian sculptor, supports are integral in the construction of the piece and its visual effect. This compositional choice, as scholars have often remarked, goes hand in hand with a new conception of the human figure, which now relied largely on external elements for balance. Among the types attributed to Praxiteles, which are known from multiple Roman copies, it seems that both figures initiated in bronze, such as the so-​called Apollo Sauroctonos or Lizard-​Slayer, and sculptures ­originally carved in marble, like the Cnidian Aphrodite, were equipped with prominent supports to match the human figure.20 Exceptionally relevant to our argument is the group of Hermes and the infant Dionysus, which was found in 1877 in the ruins of the Heraion at Olympia and whose attribution to Praxiteles has been the subject of fierce controversy among art historians ever since (Fig. 13).21 The figure’s weight rests on his right leg, while his left foot touches the ground lightly, in an off-​ balanced stance which is made possible by a bulky support on his left side. Based on a remark by the second-​century Greek traveller Pausanias, who mentions a ‘marble Hermes carrying the baby Dionysus’ by Praxiteles in the Temple of Hera, it has been conjectured that this may be the original statue 18 These arguments are summarised by Hiller (1976: 35–​6). Langlotz (1947: 102) suggested that the pillar may hint at some ancestral practice of aniconic worship, whereas Schrader (1924: 38–​9) noticed that, in its original setting, the statue’s support would remain invisible against the columns of the cella. 19 E.g. in the Nike carved by Paionios of Mende (Olympia, Archaeological Museum, inv. 46–​8), whose advanced left foot is connected to the rocky surface of the base only by a small strut. On the statue see Hölscher (1974) and, for a summary of the scholarly literature, Anguissola (2012: 156 and 169 nn. 114–​16). 20 For the Cnidian Aphrodite and the Apollo Sauroctonos see Martinez (2007a), Pasquier (2007b: 139–​46), and Corso (2007: 9–​187; 2013: 22–​65). On the prototype of the Resting Satyr and the disagreement that persists about its material, see Martinez (2007b: 241–​8, 258–​9) and Corso (2010: 42–​69). 21 The arguments supporting the attribution to Praxiteles are laid out by Corso (1996; 2013: 165–​ 72). A succinct overview of the question can be found in Pasquier (2007a: 97–​103).

49

50

Figure 13  Statue of Hermes with infant Dionysus. Marble. H. 215 cm. The statue has either been considered an original carved by Praxiteles of the fourth century BC or a copy made in the late Hellenistic or Roman period. Olympia, Archaeological Museum

51

Supplementing the Legs

carved by the Greek sculptor.22 The prominence of the support and, especially, the presence of a rectilinear bar-​like strut that connects the polished body of the god to the massive trunk at his side, have however prompted widespread rejection of this attribution in favour of a much later dating as a copy of Praxiteles’ lost original.23

Supplementing the Legs The argument that a strut between hip and tree trunk would be ‘offensive to the aesthetic sense of a fourth-​century Hellenic sculptor’ conflicts with other, more securely dated evidence from the same period.24 Although it remains difficult to determine with certainty whether struts connecting the trunk of a human figure to outstretched limbs or attributes were in fact viable options for Greek sculptors of the fifth and fourth century, one particular type of non-​figural support seems to have enjoyed some success in this period of Greek art. Both the Kallithea monument now at the Piraeus Museum and the Daochos dedication at Delphi, two family monuments erected in the second half of the fourth century, attest to the use of vertical struts located behind statues. Vertical struts are an important feature of the Daochos monument at Delphi.25 The ensemble consisted of a long, rectangular base, on which nine statues stood in a row, portraying the donor, his son and ancestors. An inscription naming the individuals represented appears on the pedestal. Six 22 The passage from the Description of Greece (5.17.3) is listed in DNO (vol. III, 144–​8 s.v. Praxiteles no. 36, 1945). 23 According to Antonsson (1937: 26, 77), the ‘offensive’ strut proves that the statue, carved in the fourth century BC, was substantially reworked centuries later. According to Adam (1966: 124–​8, esp. 125–6) the strut does not militate against a date in the fourth century BC; other technical details, however, seem to point to a later date. 24 As evidence for the irrelevance of the strut for determining the date, Adam (1966: 102–​4) offers a comparison with a statue of Asklepios from Eleusis (Archaeological Museum, inv. 50), with the remains of a strut between body and stick. This statue is accepted as a Greek creation of ca. 320 BC on account of the inscription on the base (IG II2 4414), on which see Clinton (2005–​8: no. 681). See also B. Holtzmann, s.v. Asklepios, in LIMC (vol. II, 882 no. 234), Kranz (1989: 119), and Papangelē (2002: 272). 25 See Dohrn (1968) on the sculptures and Jacquemin and Laroche (2001) for their context. Dating is connected to Daochos’ title as hieromnemon at Delphi (delegate from Thessaly to the Amphictyonic council), the title he carries in the inscription on the base. From other sources we know that he held that office from 337/​6 to 333/​2 BC. Because this is his only known connection with Delphi, it has been assumed that the monument was dedicated during this period. For a later date, shifted about two generations to the early third century BC, see Geominy (1998; 2007).

51

52

52

Supports and Greek Marble Sculpture

generations of the family were included, among which two ancestors who had won crowns in the Pythian Games and were therefore directly linked to the sanctuary at Delphi.26 Moving from right to left the group depicts the progenitor Aknonios and his three sons, Agias, Telemachos, and Agelaos, followed by the offspring of Agias in generational order: Daochos I, Sisyphos I, the donor Daochos II, and his own son Sisyphos II at the left end. Six figures survive in their entirety or as substantial fragments. A pair of sandal-​clad feet is the only surviving element of the final statue, the donor himself, Daochos II.27 All of these figures have prominent vertical figural or non-​figural supports.28 The right lower leg of the two athlete brothers, Agias and Agelaos, is augmented by a vertical mass of stone with a coarse and irregularly picked surface. Both the progenitor Aknonios and his grandson Daochos I have almost identical pillar-​like vertical supports running from the plinth to the hem of their heavy chlamys (Fig. 14). A thin vertical support, almost oval in section, is placed behind the left foot and calf of Daochos II. The statues of Sisyphos I and II, instead, are both equipped with figural supports, a tree stump and a herm respectively.29 The support plays a minor role in the composition of Sisyphos I and allows the figure, dressed in a short tunic, to stand almost fully upright. In contrast, a markedly off-​ balance stance was chosen for Sisyphos II, requiring a support in the form of a herm. The presence in the same group of figural and non-​figural supports, of different forms and in different relationships to the human body, has sometimes been seen as an indication of different sculptors’ hands.30 The Kallithea monument offers comparable evidence.31 Two of the three figures included in its imposing architectural frame are 26 The figure of Agias is to be discussed in relationship to the art of Lysippus. From a base found in Pharsalos, the family’s home town, we know that Lysippus had made there a bronze statue of Agias. It is thus possible that the marble in Delphi depended on the bronze set up at Pharsalos. The question is summarised by Anguissola (2012: 158–​60). 27 Delphi, Archaeological Museum. On the individual figures see Dohrn (1968: 34–​42). 28 See Hollinshead (2002a: 124) and Weinstock (2012: 34–​40). 29 Ridgway (1990: 47–​8) discusses the difference and suggests a slightly later date for the statue of Sisyphos II. See also Geominy (2007: 94–​5). Dohrn (1968: 41) notes that if this statue is contemporary with the rest of the monument, which he dates no later than 332 BC, it is the first evidence for a supporting herm associated with a mortal. Wrede (1986: 63) and Schoch (2009: 150–​2) consider the herm a reference to the athletic sphere of the gymnasium. 30 So Hollinshead (2002a: 124). See also Adam (1966: 97–​102) for the identification of different hands. 31 The monument of Kallithea, named after the find-​spot between Athens and Piraeus, was erected around 330–​320 BC by the Istrian metic Nikeratos for himself and his son Polyxenos. See Ridgway (1990: 31–​2), Steinhauer (2001: 307 and 350–​1 Figs. 458–​9 = 1998: 305–​6 and 348–​9 Figs. 458–​9) and, on the context and artistic influences, Israel (2013). Geominy (2007: 93–​4) proposed a later date, based on a comparable shift in the chronology of the Daochos group.

53

Figure 14  Statue of Aknonios from the so-​called Daochos monument, seen from the left side. Marble. H. 180 cm. Delphi, Archaeological Museum

54

54

Supports and Greek Marble Sculpture

supplemented by sturdy pillar-​like supports along their leg that are similar to those of the Daochos group. Due to the placement of the statues on a high platform, however, these struts must have remained almost invisible to the viewer below. As plain, squared blocks of stone, such struts properly belong to the category of non-​figural supports, unrelated to the subject and content of the composition. Although their similarity to the much-​discussed strut between body and tree-​trunk support of the Hermes at Olympia is obvious, in terms of both abstraction and non-​involvement in the narrative, their visual effect is radically different, placed as they are behind the legs or partly hidden behind the folds of the heavy drapery.32 The Daochos and Kallithea monuments demonstrate that in the late fourth century BC a variety of supports and struts was used to meet the structural needs of a composition initiated in marble. In particular, pillar-​ like struts behind or beside the foot and calf of a freestanding figure seem to be a successful invention of this period, as both the Kallithea monument and the Daochos dedication attest, as well as a handful other isolated works.33 That similar supports occur in similarly dressed (or undressed) figures indicates that the choice followed coherent formal criteria, at least within the same workshop or in the same monument.

The Emergence of Struts By the late Hellenistic period supports and struts were widely used elements of freestanding marble figures. The set of large-​scale marble statuary from Delos from the late second and early first centuries BC illustrates this evolution in technique and taste.34 Two of the best-​known specimens of Delian statuary from the turn of the second and first centuries BC  –​the group of Aphrodite, Pan, and Eros from the Hall of the Poseidoniast known as the Slipper-​Slapper group (Pantoffelgruppe) and the male portrait known as the pseudo athlete –​famously include 32 In comparing the Daochos group (in particular, Sisyphus I) and the Hermes group from Olympia, Casson commented: ‘We are not talking about supports, but about visible supports … In the Hermes we meet for the first time a strut that hits the eye, damages the general appearance of the group and is, in fact, an inartistic intrusion’ (1931: 266). 33 E.g. a late fourth-​century BC fragmentary statue from Chalkis (Archaeological Museum, inv. 10) that preserves the stub of a strut behind the left calf: Dohrn (1968: 46–​7 and Pls. 36–​7), Ridgway (1990: 49), Weinstock (2012: 40–​1). 34 See also Weinstock (2012: 86–​102, 108–​21) on tree trunk supports in the marble statuary from Delos.

55

The Emergence of Struts

prominent stone connectors. In the Slipper-​Slapper group a large strut links the goddess’ left thigh to Pan’s right leg. This element is carefully rounded and contrasts strongly with the thick wool that covers Pan’s legs.35 The pseudo athlete displays a striking number of more or less evident supports, in addition to the unusually large tree trunk behind the figure’s right leg.36 Struts are located between the legs and between the left hand and thigh, while short branches connect the massive tree trunk to the falling edge of the cloak. It is in this context that we should examine the statuary recovered from a shipwreck off the islet of Antikythera, south of the Peloponnese. It is generally believed that the ship had sailed towards Italy from the eastern Mediterranean and sunk off Antikythera’s northeast coast at some point in the second quarter of the first century BC. The marble statuary carried by the vessel seems to have been produced at the very beginning of this century, perhaps in Delos.37 The evidence from Antikythera confirms that all the main types of struts employed by Roman marble carvers were already a common feature of late Hellenistic sculptures made in Greece. From the point of view of sculptural technique, the most relevant feature is the variety of bulky quadrilateral struts that are attached to several human figures and connect the figure to its base or the outstretched limbs to the core.38 Whereas some struts might have functioned as precautions for shipping, other conspicuous props are an integral part of the composition. A primary example is the statue of a nude boy in Parian marble, bent over with his head raised, often interpreted as a pancratiast or wrestler just before the match (Fig. 15).39 35 Athens, National Archaeological Museum, inv. 3335. See Marcadé (1969: 393–​6), Marquardt (1995: 227–​36 no. 1), Kaltsas (2002: 294–​5 no. 617), Zimmer (2014: 113–​22). On the sculptural decoration of the so-​called Hall of the Guild of the Poseidoniasts (worshippers of Poseidon) from Berytos (Beirut), see also Kreeb (1988: 105–​19 cat. 7). For the group’s supporting tree trunk and its role in the composition see Weinstock (2012: 110–​12). 36 Athens, National Archaeological Museum, inv. 1828. For the statue see Marcadé (1969: 279), Kaltsas (2002: 312 no. 654), and Vorster (2007: 282–​4). On the location of the find, which is the same building where the famous Fillet-​Binder was found, see Kreeb (1988: 155–​60 cat. 7) and Nolte (2006: 284–​7). For the athlete’s support and struts see also Hollinshead (2002a: 128–​9, 143) and Weinstock (2012: 88–​90). Adam (1966: 107) thinks that the pseudo athlete was left unfinished and that the sculptor would have cut away all struts before the final finish. 37 On the historical context and the hypotheses on the provenance of the cargo see Bouyia (2012). 38 See P. C. Bol (1972: 56–​7 no. 36, 60 no. 49, 68–​9 no. 46, 78–​83 nos. 26–9), Kaltsas, Vlachogianni and Bouyia (2012: 102–​8 nos. 48–​54). Almost all statues of horses from the Antikythera shipwreck show the remains of belly supports or struts that connect hooves and plinth. See P. C. Bol (1972: 84–​91 and Pls. 50.4, 52, 53) and Kaltsas, Vlachogianni, and Bouyia (2012: 112–​15 nos. 61, 64, 66). 39 Athens, National Archaeological Museum, inv. 2773. See P. C. Bol (1972: 69–​72 no. 25 and Pls. 38–​40) and E. Vlachogianni (in Kaltsas, Vlachogianni, and Bouyia 2012: 104–​5 no. 50).

55

56

56

Supports and Greek Marble Sculpture

The gap between the thumb and forefinger is bridged by a very thin strut. More visible in the composition are the large, quadrilateral struts that almost create a cobweb around the human figure. These comprise a huge shaft, now largely corroded by sea water, joining the plinth to the statue’s left thigh, a bar between the right elbow and thigh, as well as a third such

Figure 15  Statue of a boy from the shipwreck of Antikythera, early first century BC. Parian marble. H. 111.5 cm with the plinth. Athens, National Archaeological Museum, inv. 2773

57

The Emergence of Struts

squared, roughly picketed strut beneath the right knee. Clearly, this abundance of supports was required by the peculiar stance of the young athlete whose off-​centre, crouching pose required additional measures for stability. If such struts were instrumental to the general scope of the composition, their visual share must have been acknowledged and accepted. Although slightly different solutions could have been devised that would limit the need for external support, the sculptor who created this statue and his prospective customers evidently valued the effectiveness and impact of the youth’s stance more than they disapproved of the related shortcomings in stability and balance. Sometimes, the presence of large struts has been, on the one hand, fundamental to the reconstruction of a marble composition and, on the other hand, an obstacle to accurate dating. An intricate group of Artemis and Iphigenia, found in Rome in the area of the Horti Sallustiani and now at the Ny Carlsberg Glyptotek in Copenhagen, is a perfect case in point. What remains of the group, reconstructed in the early twentieth century by Franz Studniczka thanks to its many struts, are actually the torsos of two women and the fragmentary head of a deer.40 The rear torso has been recognised as that of the goddess Artemis. In front of her, the figure of Iphigenia is preserved as a torso of a semi-​recumbent young woman, whose chiton is ripped aside exposing her right breast, hips, and right leg. Dating has ranged from the late fourth century BC to the Roman imperial period. Some have considered the group to be either a late classical or Hellenistic original; others think instead that it is a Roman copy after a lost bronze composition made at some point between the late fourth and the mid-​first century BC.41 One of the reasons for a later dating of the Copenhagen statues has been the number and prominence of their struts, thought to be incompatible with a work of Greek art.42 But the freestanding Hellenistic statuary from Delos and the marble statues recovered from the shipwreck of Antikythera constitute a reliable body of evidence about the widespread use of both figural and

40 Copenhagen, Ny Carlsberg Glyptotek, inv. IN 481–​482a. 41 According to A. M. Nielsen the group dates to the third to first centuries BC (in Nielsen and Østergaard 1997: 38–​9 no. 12), whereas E. Simon considers it to be a product of the first-​ century BC ‘classicising’ taste (LIMC, vol. II, 837–​8 no. 337 s.v. Artemis /​Diana). Smith (1991: 120 Fig. 139) and Moreno (1994: vol. II, 584–​5) believe that the group is a Roman copy after a Hellenistic original; see also Ridgway (1990: 283). 42 For the struts of the Artemis and Iphigenia group see Studniczka (1926: 141–​2). The many struts are also visible in the bronze relief depicting the archaeologist Franz Studniczka in absorbed contemplation of his reconstructed group of Artemis and Iphigenia made by Franz Hackebeil in 1928, on which see Cain (2009: 6 Fig. 1).

57

58

58

Supports and Greek Marble Sculpture

non-​figural supports by late Hellenistic marble carvers in Greece. Although this can hardly be accepted as a decisive indication that the Artemis and Iphigenia group dates to the mid to late Hellenistic period, the possibility of framing the technical device of struts within a larger and long-​term development in both workshops’ practices and, evidently, current taste, allows us to rule out one major argument that has so far prevented scholars from focusing on other, more reliable stylistic features. Struts were employed by the artists who created some of the most striking archaic kouroi and continued to be used throughout the classical period in order to secure slightly outstretched limbs or the projecting parts of a composition. It is in the late fourth century that both figural and non-​ figural supports became fundamental and often optically prominent features. By the time the Slipper-​Slapper group and the pseudo athlete had been installed in Delos, struts seem to have become a popular and accepted expedient to guarantee the stability and integrity of freestanding marble statues. Struts, in themselves, can hardly be presented as a Roman invention. Rather, stone carvers of the Roman era may have adopted the concept and practice of structural supports from the Greeks along with their ­tradition of marble sculpture. Non-​figural supports became commonplace in Roman marble statuary from the late republican period onwards.

59

3

The World of Struts

In the Vatican’s Galleria Chiaramonti stands the statue of a paunchy Silenus found in the ruins of a Roman villa near Ariccia in 1791 (Fig. 16). The statue, which dates to the age of Trajan but may stem from an early Hellenistic prototype, displays a sort of anthology of the most common typologies of struts used by Roman marble carvers during the first centuries AD. Although modern restorations made by Antonio d’Este have altered individual details of the composition (most notably the position and gesture of both hands), one important feature has remained unaffected: the many bars that connect the arms and the hanging animal skin to the body, the legs to each other, and the leopard’s belly and paw to the ground.1 The form and treatment of these struts differ remarkably. Some are plain, roughly picked masses of stone, whereas others were carved as faceted rods. Extensively propped statues like this continue to puzzle modern viewers, who have assumed either that their carvers lacked the necessary skills to create a stable composition unassisted by stone supports or that they wished to express by means of the struts some bizarre sense of humour  –​a wild, drunken Silenus restrained in his expansive move forwards by marble braces.2 Non-​figural struts, as I have shown, were well-​known devices to ensure stability and balance from the earliest Greek marble sculpture. It is however only in the late Hellenistic period that they feature prominently in many compositions, rapidly establishing themselves as almost ubiquitous elements of marble statuary. Compositions such as the Silenus from Ariccia are primary examples of how supports had become a pervasive feature of Roman stone sculpture, to secure body parts and attributes alike. In this and the next chapter, I  introduce the material that lies at the core of my argument and provide the coordinates to navigate the subject and grasp its Vatican Museums, Galleria Chiaramonti, inv. 1441. See Amelung (1903–​8: vol. I, 671–​3 no. 544) and Andreae (1995: Pls. 660–​1). The shepherd’s staff (pedum) and the strut connecting to the head are modern. Instead, on the left shoulder are the remains of a roughly picked strut, which must have supported the raised arm. 2 Hollinshead (2002a: 144) explores the possibility that this abundance of struts may have humorous effects. However, she inaccurately identifies the strut ‘sprouting’ from Silenus’ head as ancient rather than a modern addition. 1

59

60

Figure 16  Statue of Silenus from a Roman villa at Ariccia, Trajanic period. Marble. H. 175 cm. Vatican Museums, Galleria Chiaramonti, inv. 1441

61

The Places for Struts

broader implications for Roman visual culture. The first step in this exercise must be to describe the placement, form, and size of struts, taking particular notice of choices that bear little relationship to tectonic function.

The Places for Struts Struts take many forms and seem almost commonplace in Roman marble sculpture. Often, as is the case with the Silenus statue from Ariccia, non-​ representational struts occur together with figural supports like tree trunks or attributes. Struts could be used either to support the entire statue or to fasten certain parts and projections to the main axis of the figure.3 In this section, I address the placement of struts in Roman marble statues. Having a clear picture of where struts could be located is an essential premise in order to address their static function and their visual impact. One of the most common devices to reinforce a freestanding statue  –​ and yet the one which most often escapes notice –​is the triangular wedge of stone under the heel of a slightly raised foot. This occurs on many figures standing on a load-​bearing leg with the other slightly bent or raised.4 A small prop under the sole of the left foot, for instance, is the only stabilising ­element in a high-​quality late Hellenistic statue such as the Diadoumenus from Delos, which relies only on its own balance and on the small, barely visible mass of stone under the raised left foot. Broad, flat bars commonly join  –​and thereby mutually stabilise  –​the vertical elements of multi-​figure compositions. This is notably the case in several versions of the Three Graces.5 An assessment of two replicas of the Three Graces from the Baths of Cyrene indicates that the presence of such struts did not necessarily depend on the scale of a group but on a variety of reasons that concerned, first and foremost, arrangement and style.6 Both groups are less than life-​size. They both date to the reign of Hadrian, and The pyramidal support under the knee of two famous crouching ­figures –​the Atlas Farnese (Naples, National Archaeological Museum, inv. 6374) and the Crouching Venus from the same collection (inv. 6293) –​were added by modern restorers. See S. Pafumi (in Gasparri 2009: 155–​ 8 no. 71 and 69–​72 no. 29). 4 In one case, the Sciarra Amazon now in Copenhagen (Ny Carlsberg Glyptotek, inv. 1568), traces of red paint have been detected on such a heel strut. 5 Cf. also the group called Orestes and Electra from the macellum of Puteoli (Naples, National Archaeological Museum, inv. 6006), on which see Anguissola (2012: 83–​4, 107 n. 62). The large strut between the figures is omitted in Clarac’s Musée de sculpture (vol. V, Pl. 836 no. 2093). 6 Shahat (Cyrene), Museum, inv. 14.348 (h. 140 cm) and 14.346 (h. 116 cm). See H. Sichtermann (in LIMC, vol. III, 209 nos. 129–​30), as well as Paribeni (1959: 108–​9 nos. 301, 303) and Manderscheid (1981: 103 nos. 285–​6). 3

61

62

62

The World of Struts

they were displayed together at the sides of the monumental entrance to the frigidarium.7 However, they seem to be sculpted in different types of marble and rely on different strategies for their supports. The sculptor who made the larger group in Pentelic marble exploited the maidens’ sinuous pose and their leaning on each other’s shoulders to stabilise his compo­ sition (Fig. 17). Small curtains of stone only remain between the calves of the outermost right figure, as well as between this and the middle Grace. Two additional rope-​like struts join the raised arms of the outer Graces to their hips. The figures of the smaller group in Parian marble, instead, stand upright in a frontal pose. Their bodies only touch at shoulder level and are linked to one another by large flat shafts (Fig. 18).8 The sculptor also added long marble struts between the hips and outstretched hands, one of which is elegantly curved. Vertical elements of single-​figure compositions, too, were typically joined by horizontal struts. Struts connect, for instance, the human body and its figural support or attributes –​think of the many Venuses derived from Praxiteles’ Cnidian Aphrodite that deploy an array of bar struts to connect the goddess to her hydria-​cum-​drapery. Stone bridges generally link the calves, thighs, or hips of a figure to a nearby pillar or tree trunk. In many cases, one or more connectors were left in place to secure attributes such as a club, a quiver, or a rudder to its holder.9 Solid struts, of course, were all the more urgent in the case of long, thin attributes such as a spear or stock, which could potentially be compromised by the weight of the human figure leaning on it. This is particularly true for large statues whose mass poses particular challenges for the sculptor.10 Like any other item detached from the main vertical figure, drapery itself sometimes requires robust struts to secure hanging folds. Occasionally, the need to fasten the hanging drapery 7 Manderscheid (1981: Pl. 7.1) and Zanker (1992: 18 Fig. 15), in the intercolumniations between rooms G and G1. 8 In a third group of the Graces from the temple of Isis at Cyrene (Shahat, Museum, inv. 14.347, 96 cm) sturdy bars connect the upright figures. However, this group harks back to a different prototype for the outer Graces, represented in the pudica gesture. Therefore, no struts are needed to support their raised arms. See H. Sichtermann (in LIMC, vol. III, 209 no. 131) and Paribeni (1959: 109 no. 302 and Pl. 144). 9 The examples are numerous. E.g. a statue in the type of the Tiber Apollo at Cherchell (Museum, inv. S.30), on which see Landwehr (2000: 1–​12 no. 67 and Pls. 1, 6d) or a Fortuna in the type of the so-​called Dresden Artemis in Madrid (Prado, inv. 20-​E) discussed by Schröder (2004: 379–​ 82 no. 182). In the case of a statue in the pose of the Venus of Capua, from Perge (Antalya, Museum, inv. 8.29.81), a bulky horizontal strut is required by the size of the attribute, a heavy round shield. See Özgüt (1996: no. 17), E. Özgen and İ. Özgen (1988: 202 no. 89). 10 This happens also with much smaller statues, like a third-​century AD Bacchus in the National Museum of Algiers (inv. 2) which, although only 128 cm high, has two long struts to secure the thyrsus. See Landwehr (2006: 4–​5 no. 177).

63

The Places for Struts

to the body must have been so acutely felt that the strut was almost as large as the figure’s own limbs and as thick as the fold itself.11 In addition to securing vertical masses such as attributes or garments, struts often fasten the limbs of a statue to its torso, thereby providing other structural axes to bridge the voids left by splayed arms or legs. Very often, a horizontal bar connects the calves of a standing male figure, a device that reflects concerns about supporting the whole statue on a pair of slender ankles. The late Hellenistic pseudo athlete from Delos, the second-​century AD Silenus from Ariccia, as well as many other works from all periods and places of the Roman world include such a bar to join the engaged and free legs. Struts that fasten the figure’s spread or bent arms to its sides occur with equal frequency. These struts are sometimes so large that they become relevant to both the structural and visual dimension of a composition. This is clear when considering the two diagonal bars of stone that supported the right arm of the over-​life-​size wounded warrior now at the Metropolitan Museum of Art, identified as the Greek hero Protesilaos, the first who died in battle on Trojan soil.12 The two struts, one of which was longer than the arm that it supports, featured among the most prominent parts of the composition. Although possibly dismissed by viewers in the act of deciphering the subject, they still impact the visual presence of the figure. Particularly widespread are robust struts that connect outstretched hands supporting an object –​a cup, box, or another attribute –​to the body.13 Struts of this type, too, are sometimes oversized and well beyond what was required by the figure’s considerably smaller limbs. This is most apparent in the example of the wine-​pouring satyr from Torre del Greco, now at the Museo Salinas in Palermo, with a long and finely smoothed bar of stone connecting the youth’s extended left hand to his thigh. I will return to this sculpture several times in this and the next chapters. 11 E.g. the statue of a boy wearing the Phrygian cap (Ganymedes or Parides) in the Capitoline Museums (inv. S 1080, h. 134 cm) and a second-​century AD statuette of Mercury found near Cherchell (Algiers, National Museum, inv. 579, h. 90 cm). See respectively Stuart Jones (1926: 163–​4 no. 15) and Landwehr (2000: 61–​2 no. 105). 12 New York, Metropolitan Museum of Art, acc. no. 25.116, Hewitt Fund 1925. A smaller strut fastened the left arm to the torso. It is possible that the only other copy of this figure, now in the British Museum (no. 1876,0903.1; Sc. 1538), shared the same criterion for the placement and size of supports. However this torso, found in Cyzicus, is preserved only from the neck to the knees. On this sculptural type see Chapter 6. 13 This is equally the case in large, monolithic figures such as a 180 cm high matron from Puteoli, garbed as a priestess in an act of sacrifice (Boston, Museum of Fine Arts, inv. 34.113), as well as miniatures such as a seated female statuette now in Munich (Glyptothek, inv. 471). See respectively Comstock and Vermeule (1976: 224–​5 no. 355) and Vierneisel-​Schlörb (1979: 508–​14 no. 46).

63

64

Figure 17  Group of Three Graces from the frigidarium in the baths of Cyrene, Hadrianic period. Marble. H. 140 cm. Shahat (Cyrene), Museum, inv. 14.348

65

Figure 18  Group of Three Graces from the frigidarium in the baths of Cyrene, Hadrianic period. Marble. H. 116 cm. Shahat (Cyrene), Museum, inv. 14.346

66

66

The World of Struts

There is one circumstance, however, under which the size of the strut supporting an outstretched hand with a cup also serves a practical purpose, that is to allow the drilling of a water channel. This peculiar arrangement may explain the unusual width of the cylindrical shaft beneath the cup proffered by the statue of Ulysses from Punta Epitaffio, depicted in the act of offering undiluted wine to Polyphemus. Thanks to a channel leading from a hole at the bottom rear of the strut to an opening inside the vessel, the hero’s cup could be made to overflow (Fig. 19).14 A second century AD literary description of an overflowing cup illustrates the intended effect. In The Story of Leucippe and Clitophon, Achilles Tatius introduces the wonder of a precious vase made of rock-​crystal:  ‘vines crowned its rim, seeming to grow from the cup itself, their clusters drooped down in every direction: when the cup was empty, each grape seemed green and unripe, but when the wine was poured into it, then little by little the clusters became red and dark, the green crop turning into the ripe fruit’. Both the literary text and the sculpture at Baiae highlight the performative dimension of filling the cup, ‘little by little’, until it overflows with wine.15 Of course, it is impossible to know whether the channel was part of the original design or if it was drilled through the strut at a later stage. Nonetheless, the existence of other examples of figural and non-​figural supports used to disguise water channels reinforces the possibility that here, too, the support and the conduit were planned together from the beginning.16 Neck or nape struts often occur on statues from the Hadrianic period onwards produced in the eastern Mediterranean and in North Africa. I will examine these in detail in the next chapter. These generally take the form of simple masses of unworked stone at the back of a figure’s neck. Neck struts necessarily occupy a particular place in the effort to account for supports in Roman marble statuary. For one, they go far beyond any other supporting device in defying our expectations about the ancient Romans’ concepts of naturalism, completeness, and visual congruence. Besides, their regional concentration may trigger a chain of considerations concerning local taste and conventions. 14 Baiae, Museo Archeologico dei Campi Flegrei, inv. 222736. See Andreae (1983: 49–​50 and Pls. 80–​4), Hollinshead (2002a: 132), Valeri (in Miniero and Zevi 2008: 156–​8). The statue of Ulysses’ companion (inv. 222737) also holds a wineskin that was pierced to recreate the illusion of pouring wine. 15 Achilles Tatius, The Story of Leucippe and Clitophon 2.3 (tr. S. Gaselee, Loeb Classical Library). J. Elsner (2013: 108) compares the colourful effects of the cup described in the novel with the famous Lycurgus Cup (on which see also Chapter 8). For the effects of light on Ulysses’ cup see the image in Miniero and Zevi (2008: 158). 16 Cf. the multiple versions of the Boy Strangling a Goose found in the Villa of the Quintilii (Introduction) and the statue of a dancing faun from the Kerameikos (Chapter 1).

67

Figure 19  Headless statue of Ulysses from the nymphaeum at Punta Epitaffio, first half of the first century AD. Marble. H. 175 cm. Baiae, Museo Archeologico dei Campi Flegrei, inv. 222736

68

68

The World of Struts

An overview of the most-​widespread types of non-​representational supports should also include those found on Roman sarcophagi. Although functionally and conceptually different, they bear distinctive similarities in shape and treatment with those used for freestanding marble statues. Sarcophagi often include struts that secure virtually three-​ dimensional figures to the background plane. This specialised use of struts allowed the design to include an array of projections that skilfully detached human and animal bodies from the framework, thereby enhanc­ ing their three-​dimensionality.17 Although struts mostly remain hidden behind the figures that they fasten, some received a careful surface treatment and seem designed to be noticed and examined along other stylistic features.

The Shapes of Struts Struts range in size from small connectors to large bars of marble, sometimes over one metre long and larger than the limbs that they (purportedly) support. Most struts are shaped as irregular solids with flat surfaces. The occurrence of different forms cannot be linked to any chronological or geographical trend, except for a few idiosyncratic types. Only by generalising and with an ample degree of approximation can the impressive variety of struts used in Roman statuary be accounted for in a number of loosely defined categories. Further, no specific shape seems to be strictly confined to one particular place in the composition; often, the same statue includes more than one strut of different size and form. In the same way, compa­ rably placed struts often present entirely different forms even in statues that belong to the same replica series and were produced only a few decades apart. This is clear when comparing the strut-​work among the items of certain replica series, including smaller propping devices such as the wedge under the fingers of the left hand of the Lancellotti discus thrower and the longer polygonal strut that fastened the left hand to the torso on another discus thrower from Castel Porziano.18 17 Hollinshead (2002a: 137) remarks that the warriors of the Amazonomachy sarcophagus from the mid-​second century AD in the Capitoline museum (inv. 726) project forward as much as 12 cm from the background thanks to their struts. On this sarcophagus see B. Andreae (in Helbig4, vol. II, 82–​4 no. 1228). Cf. the Medea sarcophagus in Basel (inv. BS 203), dated to the late Antonine period, where small connectors link body parts and attributes, while longer struts fasten the coils of the winged snakes to the background. See M. Schmidt (1969: Pls. 19, 21). 18 Respectively Rome, Museo Nazionale Romano, inv. 126371 and 56039. See Giuliano (1979: 180 no. 117, D. Candilio, and 184–​6 no. 120, E. Paribeni) and Anguissola (2005: 318–​20 nos. 1–​2).

69

The Shapes of Struts

The great majority of non-​figural supports from the late Hellenistic period and into the Roman imperial age take the form of a four-​sided rod with a rectangular or square profile (rectangular prism), straight or oblique. The statue of a nude boy from Antikythera, the reconstructed group of Artemis and Iphigenia from the Horti Sallustiani and the Silenus from Ariccia on display in the Galleria Chiaramonti are all supported by more than one such strut, which is variously placed beneath the off-​balance limbs and between the body and an extended arm or an attached attribute. This form is the most commonly used –​although by no means the only one –​in the case of exceptionally large struts. Almost as frequently, struts take the shape of flat rectangular prisms, such as those that often connect the members of multi-​figure groups. Other more elaborate types are unique, or almost so, such as the finely smoothed hexagonal prism between the left hip and wrist of one of the late first-​century AD ‘pouring satyrs’ found in the imperial villa at Castel Gandolfo, to which I will return in the next chapters.19 In a number of cases, struts are carved as rectangular prisms that slightly taper towards one end. These are generally of modest dimensions and fasten an outstretched arm to the body.20 Occasionally, however, an exceptionally large tapering strut may be used to support a figure from below, as seen in the so-​called ‘first companion’ of Ulysses from the Polyphemus group from the grotto at Sperlonga.21 This figure, who strides forward as he lifts the pointed olive tree trunk used to blind the cyclops, is supported by an impressive number of struts, today preserved only in fragments: two tapering struts supported the lower body by linking the left thigh and the right knee to the ground, two bars must have fastened the pole to both thighs, whereas one further connector sprouted horizontally from the man’s groin to secure the pole to his stomach. 19 Malibu, J. Paul Getty Museum, inv. 2002.34 (ex Dresden, Albertinum, inv. 101). The satyr belonged to a series of four identical copies, carved in the last quarter of the first century AD and displayed in the villa at Castel Gandolfo. One of them, now in the British Museum (inv. 1838, 1231.1, Sc. 6848), bears traces of a strut in the same position. For the other two (Dresden, Albertinum, inv. 100 and 102), each of which also had a similar strut polished away by modern restorers, see C. Vorster and W. Geominy (in Knoll, Vorster, and Woelk 2011: vol. II, 863–​9 no. 207, 870–​6 no. 208). On the group see also Gercke (1968: 2–​4 nos. St. 2–​4 and 8 no. T.3) and Neudecker (1988: 141 no. 9.2.a–​d). For an elaborated polygonal strut see the Tyche from Prusias-​ad-​Hypum, a second-​century AD version of an early Hellenistic type (Istanbul, Archaeological Museum, inv. 4410). 20 E.g. the above-​mentioned sacrificing matron at the Museum of Fine Arts in Boston (inv. 34.113) and a Dionysus at the National Museum of Algiers (inv. 2). A further example is a statue of Poseidon from Sagalassos (Burdur, Archaeological Museum, inv. K.42.02.04), on which see Mägele, Richard, and Waelkens (2007: 483–​4 no. 3). 21 Sperlonga, National Archaeological Museum. See Conticello (1974: 21–​2 no. 1.1.1 and Pls. 1–​3). Conticello (1974: 21) describes the number and size of struts in this figure as ‘esorbitanti, nel numero e nelle proporzioni, le effettive necessità statiche del marmo’.

69

70

70

The World of Struts

Another famous Homeric group, from the Claudian nymphaeum of Punta Epitaffio at Baiae, depicts a slightly earlier moment in the Blinding of Polyphemus, when Ulysses presents a cup of wine to the Cyclops. This group also features a rather unusual type of prominent strut. The statues of Ulysses and his wineskin-​bearing companion include long cylinders that connect the hero’s raised cup to his knee and supported his companion’s left arm and right leg.22 The struts here seem to have been crafted by the sculptor to resemble the large tree stumps between the two figures’ flexed legs. Also uncommon within the corpus of Roman statuary are struts with an elliptical cross-​section, which only occur in a handful of examples.23 Furthermore, in a number of cases, the expansive movement of an arm extended from the body or raised to head height is accompanied by a curved strut. One such strut, shaped as a faceted rod, supports the elegant movement of a standing nude figure of a youth found in the Gardens of Maecenas on the Esquiline in the nineteenth century (Fig. 20). The statue is a variation on the type of the so-​called Centocelle Eros, which is thought to derive from a fourth-​century BC prototype. The quiver hanging on the tree trunk has suggested a possible identification as the young Apollo.24 In this reading, the object that the youth holds in his right hand, terminat­ing in a small apple-​shaped knob, may be a plectrum used to play the lyre. Others, instead, read the figure as a representation of Thanatos or the Genius of Death holding an inverted torch in his right hand. Occasionally, as is the case with the famous statue of Diana at the Albertinum in Dresden or with the Melpomene found alongside other figures of Muses in the Baths of Faustina at Miletus, struts curved in an arch can reach impressive dimensions, larger than the limb supported and more visible than any attributes.

22 Baiae, Museo Archeologico dei Campi Flegrei, inv. 222736 (Ulysses) and 222737 (Ulysses’ companion). 23 E.g. the oval trace on the left leg of the first-​century​ BC statue of Aristogeiton at the Centrale Montemartini to be discussed in Chapter 6. Cf. the slightly tapering cone attached to the so-​ called Choiseul-​Gouffier Apollo in the British Museum (reg. no. 1818,0801.1, Sc. 209), on which see Smith (1892–​1904: vol. I, 86–​7 no. 209). In discussing a nude male torso in Berlin (Staatliche Museen, Antikensammlung, inv. Sk 1486), B. Fröhlich (1998: 153 and n. 452) maintains that oval struts were preferred by Greek sculptors, as opposed to the quadrangular ones chosen in later periods. The evidence from Greek freestanding marble statuary, however, seems inconclusive. 24 Rome, Musei Capitolini, inv. 1092. See Stuart Jones (1926: 156 no. 4), H. von Steuben (in Helbig4, vol. II, 388–​9 no. 1586) and, for the tree trunk support with a quiver, Muthmann (1951: 70–​1). On the replica series see Zanker (1974: 108–​9).

71

Figure 20  Statue of a youth, interpreted as Eros, Thanatos, or Apollo, second century AD. Marble. H. 164 cm. Rome, Musei Capitolini, inv. 1092

72

72

The World of Struts

The character’s expansive pose is highlighted by the curved strut, which follows the upwards movement of the arm.

Expanding Movement The Dresden Artemis and the Melpomene from Miletus, both works of the mid second century AD, belong to a group of Roman statues that are characteristically overpropped. Mostly supplemented by long quadrangular bars, less often by curved rods of quadrangular section or by cylindrical struts, such compositions stand out from the corpus of Roman statuary because of the striking discrepancy between the actual static needs and the exorbitant size of their supports. Typically, sculptors placed bulky supports under a raised arm or forearm that had to be fastened to the body. Although several solutions existed to underplay the effects of a strut –​such as placing the device at a point where the gap in the undercut is minimal –​the carvers of these statues do not seem to have considered the struts’ visibility a problem. That the visibility of struts was not a major concern for the Roman sculptors and their customers is evident when observing works such as, for example, the first-​century AD pouring satyr from Torre del Greco, now in Palermo. The statue boasts an oversized strut that can hardly be accounted for in terms of statics, balance, and marble technology (Fig. 21).25 At 39 cm long, which is more than a quarter of the statue’s height, the strut between the youth’s left hand and thigh is carefully shaped in the form of a quadrangular, slightly tapered shaft. This bar originates from the left thigh, forming a sharp angle with the body. Where it joins the hand, the strut broadens into a thick cushion of marble, such that the hand itself is carved in relief on the top. Together with the tree stump, here unusually tall and located in a different position, the strut creates a composition that stands out among other replicas. Apparently, the presence of large struts appealed to the taste of certain sculptors and their public.26 This impression is strengthened by another 25 Palermo, Archaeological Museum A. Salinas, inv. 1556. See Gercke (1968: 4–​5 no. St. 5) and J.-​L. Martinez (in Pasquier and Martinez 2007: 272–​3 no. 66). The figure measures 146 cm (163 cm with the right arm and oinochoe), the tree stump on the satyr’s right side is 86.5 cm high and the strut between the left hand and thigh is ca. 39 cm long. 26 E.g. the larger than life-​size statue of Venus of the Arles type at the Centrale Montemartini in Rome (inv. 2139) on which see H. von Steuben in Helbig4 (vol. II, 504–​5 no. 1725) and the miniature version (h. 54 cm) of the Lateran Poseidon at Eleusis (Museum, inv. 5087), which E. Bartman (1992: 132–​3) ascribes to an Attic classicising workshop of ca. AD 140–​60. See also the statue of a Niobid found in the frigidarium of the Roman baths at Aeclanum, in ancient Samnium (Avellino, Museo Irpino, inv. 94), displaying an impressive array of sturdy quadrangular struts. In particular, a bulky bar sprouting from the boy’s upper right arm and running parallel to his lower arm seems to be intended merely to support the tips of the fingers. See Soleti (2010: 54–​5 no. 2.a.AE.1 and Pls. XXVI–​XXVII).

73

Figure 21  Statue of a Pouring Satyr from Torre del Greco near Naples (Villa Sora), mid-​first century AD. Marble. H. 146 cm (163 cm with the raised right hand and oinochoe). Palermo, Archaeological Museum A. Salinas, inv. 1556

74

74

The World of Struts

solution, which was occasionally adopted by sculptors of the first two centuries AD: to use multiple struts to support raised arms. This becomes clear in examining a few case studies, among which the most eloquent is the statue of a diadoumenos at the Metropolitan Museum of Art, variously dated from the Flavian to the Antonine period but probably to be assigned to the last quarter of the first century AD (Fig. 22).27 Unfortunately, when the statue of the youth reached its new home in New York in the early twentieth century, it was in a much poorer state of preservation than what we see today. Essentially, it consisted of a few fragments: the head, both arms (but only the left hand), both calves and feet and most of the palm trunk support. The missing torso and the upper part of the legs had to be supplied from a plaster cast of a Hellenistic replica of the same work, the famous Diadoumenus of Delos, dating to the turn of the first century BC. Although separated by about two centuries, the two statues are similar to the point that the plaster patchwork fits almost perfectly with the fragments.28 Few other examples show as conspicuously how the form of certain classical Greek works of art had survived almost unaltered for centuries after their original, which, in the case of this athlete, dates to the second half of the fifth century BC. Notwithstanding their striking similarity in size and proportion, the New York and Athens diadoumenoi differ in a manner that should not pass unnoticed. The New  York statue bears traces of a peculiar abundance of heavy supports. These joined the left hand to the shoulder, both elbows to the body, the right shoulder to the fillet, and the two legs to each other (Fig. 23).29 The fact that the dimensions coincide with those of the marble from Delos, fully preserved with only a tiny strut under the left foot, rules out static constraints. Moreover, the fine quality of the New York diadoumenos suggests that it is the work of an accomplished marble carver, one not necessarily anxious about working without extensive props or unable to find a more discreet placement for them, had he wished to do so. Clearly, both the carver and contemporary viewers were not as puzzled by these ‘inartistic intrusions’ as we are. 27 New York, Metropolitan Museum of Art, Fletcher Fund 1925 (acc. no. 25.78.56). See Richter (1935; 1954: 30–​2 no. 38) and Kreikenbom (1990: 111–​12, 114, 188 no. V.2). 28 See Kreikenbom (1990: 111–​12). Richter (1954: 31) explains that ‘it was found that the different parts of the two statues corresponded so exactly that nothing had to be added or subtracted at the junctures; only the directions of the limbs and of the head differed slightly and made some adjustment necessary’. 29 On the struts of the New York Diadoumenus see Geominy (1999b: 57 Appendix IV), Hollinshead (2002a: 146), Anguissola (2013: 1–​2, 14, 16). In the drawing by M. G. Winkes published by Hollinshead (2002a: Fig. 6.20), however, the strut under the right arm, of which clear traces remain, as well as the bar connecting the calves are missing.

75

Expanding Movement

Other compositions follow a similar strategy for the duplication of struts, in the form of hefty diagonal shafts stretching both from the figure’s hip to the underside of the upraised elbow and from the top of the same shoulder or biceps to the wrist. The strategy of double supports above and below the raised arm is employed by the warrior statue known as Protesilaos in the Metropolitan Museum of Art. Similarly, another statue set up in the Baths of Faustina at Miletus, a colossal headless statue of Venus, which elaborates on the type of the so-​called Venus of Ostia, is supplemented by two long bars from right hip to elbow and from right biceps to wrist. For the outstretched left forearm, instead, the sculptor adopted a different technique and carved the limb from a separate block of marble.30 Sometimes the strut placed under the upraised arm, instead of linking hip and elbow, reached from the hip or thigh to the wrist of an outstretched arm, thereby connecting two even farther points of the composition. The athlete with a body-​scraper in the Vatican Museum, considered to be a copy of Lysippus’ bronze Apoxyomenos, provides, again, a perfect case in point to illustrate this strategy. The athlete’s right arm, which projects straight forward from the body, was supported by a long diagonal strut running from the right thigh to the wrist. The broken-​off stubs remain of this bar, which was noticeably longer than the athlete’s arm and approximately the same length as his leg. The startling effect of unexpectedly large struts is evident if we try to recover the ancient appearance of two marble sculptures of the same ­series, the Discobolus (Discus Thrower): the torso from Castel Porziano at the National Museum of Rome and a similar piece found in Carcassonne in the bed of the river Aude and now in the Museum Saint-​Raymond in Toulouse.31 Both count among the versions that best preserve the ‘contorted and elaborate’ posture of the athlete created in bronze by Myron around 460–​450 BC. The figure is crouched, ready to throw the discus ‘as 30 Istanbul, Archaeological Museum, inv. 2004. See Mendel (1912–​14: vol. I, 332–​3 no. 126), K. A. Neugebauer and Th. Wiegand (in von Gerkan and Krischen 1928: 121–​2 no. 16), Manderscheid (1981: 94 no. 213). On the struts see also Hollinshead (2002a: 147–​8) and on the technique of piecing Schneider (1999: 43). 31 On the discus thrower from Castel Porziano (Rome, Museo Nazionale Romano, inv. 56039) see D. Candilio (in Giuliano 1979: 180 no. 117) and Anguissola (2005: 319–​20 no. 2). The date of this statue is a matter of debate. Most scholars accept a dating to the reign of Hadrian, while others consider it to be an Augustan copy that was later restored. See Rizzo (1907: 7), W. Fuchs (in Helbig4, vol. III, 176–​80 no. 2269), and Neudecker (1988: 240 no. 70.1). The Toulouse statue (Musée Saint-​Raymond, inv. 30013) is discussed by Slavazzi (1996: 29–​30, 72, 90–​1) and Anguissola (2005: 322–​3 no. 7) with earlier bibliography.

75

76

Figure 22  Statue of the Diadoumenus, last quarter of the first century AD. Restored with casts of the Delos Diadoumenus. Marble and plaster. H. 185 cm. New York, Metropolitan Museum of Art, Fletcher Fund 1925, acc. no. 25.78.56

77

Figure 23  Drawn reconstruction of the Diadoumenus at the Metropolitan Museum. Plaster integrations are in grey; the reconstructed right hand and wrist, fillet and struts are represented with a dotted line

78

78

The World of Struts

if he would spring up all at once with the cast’.32 The rotated and upwards movement of the athlete was, in both cases, supported by a heavy diagonal strut, extending from the right hip to the discus (Fig. 24). It is possible that other replicas of this series originally included a bulky strut to hold up the discus, the stubs of which may have been removed and smoothed away by modern restorers.33 Clearly, the presence of a bar of considerable size at a point which is so essential to visualise the spinning movement of Myron’s athlete, his momentous tension, and the thrust upwards, must have affected the composition in a way that the present state of these torsos fails to convey. The Homeric groups of Scylla and of the Blinding of Polyphemus at Sperlonga demonstrate especially well how struts were instrumental in ­staging and highlighting expansive poses and dramatic gestures. In the former, long bars connect the raised and outstretched arms of Ulysses’ companions attacked by the sea-​monster. In the Blinding of Polyphemus, instead, the sculptors chose massive tapering struts for a composition aimed at stressing the tense and precarious position of the men, almost frozen in terror, who stealthily approach the sleeping giant.

Carved Details Unlike the Greeks surreptitiously approaching him, the colossal Polyphemus of Sperlonga, who lies asleep on a rock, does not require any massive strut to support his arms and legs. Instead, this statue is characteristic in that it features numerous tiny struts to connect some of its smallest parts, such as fingers and toes.34 The sculptor fastened the cyclops’ right hand to the body via two rectangular struts linked to the thumb and little finger. Another 32 Quintilian (Institutes of Oratory 2.13.10) defines Myron’s Discobolus as ‘contorted and elaborate’, whereas Lucian (The Lover of Lies 18) describes the athlete’s pose in further detail. 33 Such is the case of the Lancellotti Discobolus, the sole full-​size complete replica of Myron’s discus-​throwing athlete. Guattani (1784: XII–​XIII) reports: ‘un tal soggetto eseguirsi non potea, che col porre un grandissimo puntello sotto il braccio destro (come di fatti avea prima, che il braccio fosse impernato)’ (= Cancellieri 1806: 31–​2). See also Welcker (1849: 423), E. Paribeni (in Giuliano 1979: 184–​6 no. 10) and Geominy (1999b: 58–​9 Appendix VI). A hand with a discus at the Museum of Ancient Sculpture G. Barracco (inv. 98), which probably belonged to a replica of Myron’s Discobolus, preserves the stub of a thick spiral strut (see Chapter 4). 34 For Polyphemus’ hands and feet see Conticello (1974: 16–​17 nos. 8–​11 and 12–​14), who provides the inv. no. Bellante of the fragments (1974: 24). See the images in Conticello (1974: Pls. 20–​2), Andreae (1999: 198–​9).

79

Figure 24  Drawn reconstruction of the Discobolus from Castel Porziano (now in Rome, Museo Nazionale Romano). Reconstructed parts are in grey

80

80

The World of Struts

bar runs between thumb and index finger. Three shorter bridges join all the other fingers to one another (Fig.  25). Similar struts  –​a rectangular prism between thumb and forefinger and smaller bridges for all other fingers –​secure the left hand, which rests open along the sloping profile of the rock. The smallest struts link all the toes of Polyphemus’ surviving left foot, whose exceptional quality is apparent in the detailed carving of the giant’s bones, nails, and the tufts of hair on the foot and on each toe. Significantly, no other item in the group shares this feature with the Polyphemus statue, which remains unique in its broad array of small connectors between the extremities.35 Interdigital struts seem to be infrequent in the late Hellenistic and Roman marble statuary.36 Although it is impossible to put forward reliable statistical data covering the whole period and area under consideration, we can nonetheless observe that struts for the fingers occur repeatedly in statues carved during the first century BC and the first century AD. One of the most common types of interdigital struts is small bars between the thumb and forefinger. One such bar appears on the early first-​century BC statue of a boy from the Antikythera shipwreck, together with smaller connectors between the other fingers, as well as on another fragmentary hand from the same cargo.37 A statue from ancient Messene, a fragmentary copy in Parian marble of the colossal Resting Heracles of the Caserta type, found in the gymnasium, includes a comparable strut to connect the thumb and forefinger of a hand.38 The poorly preserved tabula ansata carved on the rock beneath Heracles’ club bears the signature of the sculptors, Apollonios son of Hermodoros from Alexandria and Demetrios son of Apollonios. Based on prosopographical considerations, scholars believe that the two masters from Alexandria were active in the late Augustan period. A small number of other works feature less regularly shaped connectors between the fingers; among these

35 Two other statues from the villa at Sperlonga, however, include interdigital struts. These are a draped maiden (the so-​called Circe) and a female figure interpreted as Andromeda chained to the rock (or perhaps Hesione). See Neudecker (1988: 221–​2 nos. 62.6 and 62.27). 36 Dainty finger struts appear again in late-​antique miniature sculpture, which is often characterised by lacy strut-​work, as described by Stirling (2005: 40, 45, 108, 125). See also Chapter 8. 37 For the Antikithera hand (Athens, National Archaeological Museum, inv. 15550) from a larger than life-​size male figure, see E. Vlachogianni (in Kaltsas, Vlachogianni, and Bouyia 2012: 110–​11 no. 58). 38 See Themelis (1998–​9: 79–​81; 2013: 153–​60). A strut connecting the thumb and forefinger of a hand is on another well-​known sculpture from Messene (Museum, inv. 330), the statue of Artemis Laphria, dated to the Antonine period, on which see Themelis (2003: 139–​40).

81

Figure 25  Statue of Polyphemus from Sperlonga, right arm and hand. Marble. Sperlonga, National Archaeological Museum

82

82

The World of Struts

are the pseudo athlete from Delos and the mid-​first-​century AD Vatican Apoxyomenos.39 Miniature struts are also occasionally used to support a statue’s penis when it is carved from the same block of stone as the figure itself. This device was used for two of the sculptures from the first-​century AD nymphaeum of Punta Epitaffio at Baiae, the wineskin-​bearing companion of Ulysses and a smaller than life-​size statue of Dionysus with a panther, both of which include a small strut from penis to scrotum.40 A smooth, curved strut between penis and scrotum also features on the famous headless statue of a discus thrower from Castel Porziano, considered by some an Augustan copy, later restored and partially reworked. Within the category of genital struts, a particularly intriguing case is that of ­figures –​typically satyrs –​ with an erect phallus fastened to the lower abdomen by a small strut.41 Obviously, erect members could be carved much more efficiently either separately from the body or in relief over the lower abdomen. These struts suggest, therefore, that the choice of a daring undercut at this delicate and difficult-​to-​reach point of the statue may be meant to highlight the detail

39 The pseudo athlete is equipped with several small struts for his left hand; these occur between the thumb and the index finger, between the other fingers, between the upper thigh and both the thumb and the little finger. The Apoxyomenos bears traces of a small stone bridge on the palm of the right hand, which likely secured the bent little finger; see P. Liverani (in Moreno 1995: 201). Few other statues include interdigital struts. One of these is the Kassel Apollo (Museumslandschaft Hessen Kassel, inv. Sk 3), on which see Gercke (1991: 92). Interdigital supports also appear in two marbles from the nymphaeum at Baiae –​the figures of Antonia Minor (inv. 222738; traces of struts on the reattached fingers of the left hand) and Dionysus with a panther, on which see Zevi (1983: Figs. 149–​50) –​as well as in a colossal left hand from Pergamon, probably belonging to a portrait of Trajan or Hadrian (Berlin, Antikensammlung, inv. AvP VII 282.8; V3.1–​26), published by A. Fendt (in Grüßinger, Kästner, and Scholl 2011: 574–​5 no. 10.6). Cf. three nude male figures at the National Archaeological Museum of Formia (inv. 15423, 15424, and 147633), all dating to the Augustan and Julio-​Claudian periods, with interdigital struts that probably served to fasten small or thin attributes such as a sword, a spear or a roll; see Cassieri (2013: 37–​44 nos. 14–​16). 40 See Zevi (1983: Figs. 94, 142). Andreae (1991: 264) considers the genital strut of the wineskin-​ bearer as evidence that the statue is the copy of a bronze prototype. Another statue of Dionysus with a panther including a genital strut is in Dresden (Albertinum, inv. Hm 292); S. Oehmke (in Knoll, Vorster, and Woelk 2011: vol. I, 587–​92 no. 133) dates this marble to the second half of the second century AD or the first half of the third century. Genital struts appear also in the small statue of a young boy now in Chicago, Art Institute (Katherine K. Adler Memorial Fund, 1976.426), dated to the first century AD, as well as in a nude male figure at Formia (inv. 15424) from the Julio-​Claudian period. 41 E.g. the Capitoline version of the Townley group (Rome, Centrale Montemartini, inv. 1729), on which see Stähli (1999: 365–​6), as well as two version of the so-​called Dresden symplegma of a satyr and Hermaphroditus –​in Dresden (Albertinum, inv. Hm 155) and in Eichenzell (Schloss Fasanerie). See respectively C. Vorster (in Knoll, Vorster, and Voelk 2011: vol. II, 922–​ 9 no. 221) and Stähli (1999: 314–​15 no. 1.6).

83

Carved Details

itself and the skill of its carver, and perhaps the humorous character of the composition. The use of these small connectors was not limited to fingers, toes, and the male member. They frequently appear supporting a variety of other delicate elements of a composition.42 A  particularly widespread device is the use of bridges between folds of drapery, which appear, again, on the statue of Ulysses from the nymphaeum of Punta Epitaffio.43 The technique of connecting separate folds with struts seems to have flourished in the Hadrianic and, especially, Antonine periods, with the purpose of increasing the movement and texture of the marble surface rather than for reasons of stability.44 Indeed, from the mid-​second and throughout the third century AD the taste for small connecting struts seems to have particularly affected the treatment of hair, which is often enlivened by bridges that span the drill channels of voluminous coiffures.45 This feature, which is especially frequent in the reliefs of sarcophagi, can hardly be described as a by-​ product of the sculptural techniques of drilling. Instead, hair struts seem to have been left in place by choice and should, therefore, reflect precise expressions of artistic taste related to texture and the contrast between light and shadow. According to Mary B. Hollinshead, hair struts are indicative of ‘Roman appreciation for the aesthetic value of pattern’. The sequences that they form in a head of hair create a ‘visual rhythm’ which succeeds in introducing a dimension of variety and a ‘tactile’ quality of texture.

42 In a portrait bust of Caligula now in Copenhagen (Ny Carlsberg Glyptotek, inv. 1453) two tiny rectilinear struts stabilise both ends of a crisp ribbon around the emperor’s head. On the portrait see Boschung (1989: 118–​19 no. 43) and Johansen (1994: 134–​5 no. 55). On the struts see also Hollinshead (2002a: 134 note 53). 43 An image of these folds is in Zevi (1983: Fig. 80). 44 E.g. for the cloak of a late Antonine statue of Hermes from Side (Museum, inv. 266) see Inan (1973: 71–​7, Pl. 17). 45 See Strong and Claridge (1976: 200), as well as the seminal discussion in Hollinshead (1998: esp. 127 for the quoted passage).

83

84

4

Surface and Colour

The male portrait from Delos, now in the National Museum of Athens, known as the pseudo athlete displays a striking number of more or less prominent supports (Fig.  26). In its combination of a realistic portrait head with a nude ideal body, the pseudo athlete constitutes an exceptional document for the society of late Hellenistic Delos and its eclectic visual culture. The prominence of supports may reflect the carver’s concerns about the stability of such a large figure, which was cut out of a single block of marble (with the minor exception of a section at the top of the head) and which rested on slender ankles with the left leg drawn to the side and rear. All the struts of this statue display a rough surface treatment.1 The sculptor treated struts and the tree trunk in the same way with long, mainly oblique strokes of the point. In contrast, he carefully finished the rest of the statue with a variety of instruments –​drills to prepare the way for more detailed carving, flat and rounded chisels to smooth over the surfaces, a rasp to complete the work. The evident lack of interest in giving the struts a high level of finish has induced some scholars to believe that these were meant to be removed before the final smoothing.2 Indeed, the tapered ends of the strut between the figure’s calves seem to suggest that the sculptor had begun to remove the bar. However, this uneven surface of figural supports is characteristic of sculptures produced by Delian workshops of the late Hellenistic period.3 This suggests that the treatment of both trunk and struts may be a matter of stylistic choice and the workshop’s practices. It has been argued that the sculptor who made the pseudo athlete was sensitive to the placement of struts; horizontal bars in peripheral positions, such as between the calves, have a less finished surface, whereas other and more visible struts were shaped as faceted rods. This reading is inaccurate for two reasons. First, it depends on a misunderstanding about the statue’s For a view of the trunk and the struts that connect it to the drapery see Adam (1966: Pl. 53c). 2 So Adam (1966: 107); contra Weinstock (2012: 88). 3 For tree trunk supports in the late Hellenistic statuary from Delos see Weinstock (2012: 86–​102). 1

84

85

Figure 26  Standing male portrait from Delos, best known as pseudo athlete, early first century BC. Marble. H. 225 cm. Athens, National Archaeological Museum, inv. 1828

86

86

Surface and Colour

number of struts.4 Secondly, it implies that auxiliary supports in the lower section are per se of lesser visual impact than those above, that is, that struts providing horizontal support to the lower half of the figure were simultaneously seen and ignored while those bracing the upper body required additional care to please the eye. This suggestion poses an incorrect causal link between the position of supports and their treatment. There is little evidence, in fact, that placement influenced the care that the sculptor put into carving supports. A more relevant question, as I argue in this chapter, is whether the surface treatment of struts is indicative of their role in the statue’s composition. Smaller struts, which were required for the statue’s stability, often have a rough surface. This may indicate that the sculptor was not interested in these devices from an artistic point of view. It is equally possible, however, that rough surfaces served to distinguish the strut from the more finely carved flesh of the figure. The examples where the surface of struts is finely polished or carved in a decorative fashion also provoke conflicting explanations. Struts may be carefully finished or decorated in order to enhance their visibility. Equally however, sculptors may have chosen this treatment in order to harmonise the strut with the adjoining surfaces.

Surface Treatment The surface treatment of struts does not conform to any consistent criteria. Most struts were worked solely with the point. Often, the strut combines flat planes and sharp angles with uneven surfaces.5 Coarsely carved struts in the form of a more or less regular prism are found on statues from different periods and sites. They are striking in their contrast with the polished surfaces of skin and drapery. In general, all of the struts in a single figure receive the same surface treatment. The choice not to give a strut a polished Hollinshead (2002a: 117 n. 2, 128–​9) speaks of ‘horizontal bar-​like struts’ between the figure’s calves and ‘a faceted rod-​like strut from hip to arm’. No such rod-​like strut exists; the misunderstanding derives from a deceptive plate in Michalowski (1932: Pl. 15). Hollinshead (2002a: 127) suggests a similar explanation for the satyr with infant Dionysus from the Farnese collection, now in Naples (inv. 6022), displaying a ‘roughly finished bar … worked solely with the point, between the legs, while a strut on the same piece located closer to the more finely detailed head exhibits a more finished (i.e. chiselled) surface’. As reconstructed by S. Pafumi (in Gasparri 2009: 137–​9 no. 61), the bar between the satyr’s calves is ancient, but the finished strut close to the satyr’s head is a modern restoration. 5 E.g. the coarse strut between the torso and right arm of a marble satyr from the Hadrianic nymphaeum at Sagalassos (Burdur, Archaeological Museum, inv. K.171.44.03), on which see Mägele, Richard, and Waelkens (2007: 484 no. 4). 4

87

Surface Treatment

surface may serve to distinguish skin from extrinsic elements. Often both tree stumps and non-​figural supports share a surface texture similar to the bark of a tree.6 Struts that receive the same surface treatment as other parts of a statue are less frequent but by no means an exception.7 Both the bulky strut of the Pouring Satyr in Palermo and the hexagonal prism between the hip and wrist of a youth of the same type from Castel Gandolfo, now in the J. Paul Getty Museum, were carefully smoothed on all sides, apparently with the same instruments used for the skin. I  have already mentioned the curving strut that connects the right hip and hand of the youth from the Gardens of Maecenas (variously interpreted as Eros, young Apollo, or Thanatos). This, too, was given not only a faceted shape, but also a refined surface. Only occasionally did sculptors treat the forward and rear facing facets of struts differently –​the former smooth, the latter coarse.8 This treatment applies sometimes also to figural supports. There is evidence of supports in the shape of tree stumps that attempt a naturalistic imitation of the bark on the front side, while being left undifferentiated on the rear.9 The available evidence suggests that the surface treatment of struts did not follow any consistent patterns. Neither placement in the composition nor any other compositional constraints can be associated with the choice to leave struts rough or to give them a smooth surface. The degree of finish seems rather to be a matter of individual choice, possibly as a consequence of specific expressive needs or workshops’ traditions. 6 See also the statue of Meleager found at S. Marinella (Civitavecchia) and now in Berlin (Antikensammlung, inv. Sk 215). Although extensively restored, it preserves an ancient support between left hip and wrist. The rugged surface, distinct from the polished skin, is identical to both the wooden spear and the tree stump. See H. Heres in Scholl and Platz-​Horster (2007: 198–​9 no. 118). The statue was found together with another, fragmentary copy of the same type, now at the Fogg Art Museum (Cambridge, MA, inv. 1926.48), on which see Neudecker (1988: 217 no. 58.2) and Vermeule and Brauer (1990: 45 no. 30), also including a rough-​hewn strut on the left side. 7 The statue of the Muse Euterpe from the Baths of Faustina at Miletus (Istanbul, Archaeological Museum, inv. 1999) includes a broken-​off strut on the left hip that continues the folds of the drapery. See Schneider (1999: 9 and Pl. 7c). 8 E.g. the torso in the style of Polyclitus now at the J. Paul Getty Museum (inv. 73.AA.93) on which see Picón (1995: 232–​3 and Figs. 13.9–​10) and a statue of Athena from Perge (Antalya, Museum, inv. 1.22.82) presented in Özgüt (1996: no. 19), E. Özgen and İ. Özgen (1988: 202 no. 91). Cf. the pair of marble images of Titus and Vespasian from Misenum (Baiae, Museo dei Campi Flegrei, inv. 153350 and 153951), on which see S. Adamo Muscettola (in Miniero and Zevi 2008: 201–​2). 9 For a Dionysos with Pan at Ny Carlsberg Glyptotek (inv. 2080) see L. Leegard (in Moltesen 2005: 120–​2 no. 50).

87

88

88

Surface and Colour

Neck Struts The category of the so-​called neck struts or nape struts, which I  have already mentioned with regard to the statue of Melpomene from the Baths of Faustina at Miletus, provides further examples to address the question of surface finish. A neck strut is a mass of stone left at the back of a figure’s neck, apparently as a support for its neck and head.10 Neck struts are a widespread feature of marble statuary from the mid to late Hellenistic period through late antiquity, for sculptures of various subjects and sizes.11 Generally, neck struts are shapeless masses of stone. They are only occasionally fashioned into rectilinear solids.12 Even when neck struts have an almost regular shape, however, their surface remains rough, worked with the pick alone or, occasionally, with the point. Dimensions play no role in the use of neck struts, which were added to statues irrespective of their size.13 Even miniatures sometimes include a strut almost as large as the head itself. They occur on many sculptural types, including both nude and draped figures, as well as cuirassed statues and togati.14 The catalogue of statues with neck struts includes many male and female portraits, both of members of the imperial house and private persons. One series, in particular, is repeatedly associated with neck struts, the female portraits combined with the ideal body type known as Large Herculaneum Woman.15 This type, which takes its name from 10 This material is cursorily discussed by Hollinshead (2002a: 132–​3, 138–​41), Stirling (2005: 117–​19), Friedland (2012: 24–​5, 31 n. 14, 32 n. 23). Occasionally it is impossible to tell apart actual struts and the irregular bumps of coarsely carved shoulders or the raised edge of a cape. See Rosenbaum (1960: 43 no. 16 and Pl. XVI.1–​2), Inan and Alföldi-​Rosenbaum (1979: 61 no. 5 and Pl. 5.1–​2), Stirling (2005: 43, 68). 11 E.g. the colossal Juno Cesi (Rome, Capitoline Museums, inv. 731), a work of the second quarter of the second century BC. See Stuart Jones (1912: 340–​1 no. 2), H. von Steuben (in Helbig4, vol. II, 233–​4 no. 1427) and, on the strut, Papini (2010: 259; 2015: 108). Other early examples are in Kane Trimble (1976–​7: 322–​3 no. 8), Inan and Alföldi-​Rosenbaum (1979: 150 no. 97). For Late Antiquity see Stirling (1997a; 2005: 116–​19, 136). 12 See Inan and Rosenbaum (1966: 115 no. 124, 183–​4 no. 248), Inan and Alföldi-​Rosenbaum (1979: 216–​17 no. 188, 261–​2 no. 244). Cf. a statuette of Ganymedes in the State Museum of Egyptian Art in Munich (inv. ÄS 4198), on which see Müller (1975). 13 For a portrait head from Caesarea Philippi, 6 cm tall with a strut of 4 cm, see Friedland (2012: 134–​7 no. 24). Other examples are in Kane Trimble (1976–​7: 322–​3 no. 8) and Stirling (1997a; 2005: 117–​19). 14 E.g. the statue of a togatus from the East Baths at Jerash (NEB.5) discussed by Friedland (2003a: 338; 2003b: 442–​4 no. 4) or the one found at Koçhisar (Afyon, Museum, inv. 3240) on which see Inan and Rosenbaum (1966: 183–​4 no. 248). Cuirassed statues with neck struts are equally widespread: e.g. Antalya, Museum, inv. 3730 and Antakya, Museum, inv. 11113, on which see Inan and Alföldi-​Rosenbaum (1979: 97–​9 nos. 46–​7). 15 On this replica series and its use in portraiture, see the landmark study by Trimble (2011). Neck struts appear on two portraits in the guise of the Large Herculaneum Woman found

89

Neck Struts

an over-​life-​size statue found in Herculaneum in the early eighteenth century, represents a woman in a pose of solemn modesty, clad in a long drapery. Roman sculptors often combined copies of the Large Herculaneum Woman body type with individual heads to create portrait statues that hinted at the sitter’s virtue and social status. A portrait of Faustina the Elder in the pose of the Large Herculaneum Woman, formerly in the Pembroke collection at Wilton House and now in the J.  Paul Getty Museum, seems to defy all modern expectations about realism and completeness in its use of a neck strut (Fig.  27).16 The carved folds of the mantel, which is raised to cover the empress’ head, stop abruptly behind her neck to allow for a large, shapeless protrusion of unworked marble. The contrast between the coarsely picked strut and the adjoining surface of the veil, where the artist painstakingly singled out the individual folds, is startling. The naked bodies of well-​known athletes or gods types, derived from classical models, also occasionally include neck struts between the shoulders, such as a fragmentary Apoxyomenos found in the so-​called building M at Side and two statues from Perge –​a Diadoumenus torso and the Sandalenbinder from the South Baths. In a number of cases, neck struts appear on statues with unfinished backs.17 Sculptures such as the Muses group from Miletus, which have a coarsely carved back and are sometimes supplemented by a neck strut, were probably displayed in niches.18 The available evidence of rough backs-​ cum-​strut, however, does not allow us to draw universally valid conclusions about the visibility of supports. The Melpomene from Miletus is exemplary in this respect. Although the placement in a niche probably hid the large neck strut, the front view included a large arched strut under the right arm, nearly as large as the arm itself. near the city gate at Perge. One of them probably represents a member of the imperial house, Sabina (Antalya, Museum, inv. A3086 /​A3066); the other is Plancia Magna, whom inscriptions identify as a city benefactress (inv. A3459). See Inan and Rosenbaum (1966: 72–​3 no. 36), Inan and Alföldi-​Rosenbaum (1979: 248 no. 225), and Trimble (2011: 166–​ 70, 401–​4 nos. 84–​5). The portrait of Faustina the Elder from the nymphaeum of Herodes Atticus at Olympia (inv. Λ 155; see Chapter 8) combines an inset head with a redundant strut at the nape. 16 Malibu, The J. Paul Getty Museum, inv. 70.AA.113. See Vermeule and Neuerburg (1973: 32 no. 68), Inan and Alföldi-​Rosenbaum (1979: 106–​7 no. 55), Schneider (1999: 32–​3), Trimble (2011: 444 no. 173). 17 See Kane Trimble (1976–​7: 322–​3 no. 8) and Friedland (2012: 83–​6 no. 4, 134–​7 no. 24). 18 For the statue of Melpomene see Schneider (1999: 8 and Pl. 3b); the Muse of the ‘Münchner Klio’ type (Istanbul, Archaeological Museum, inv. 1994), on which see Schneider (1999: 8–​9 and Pl. 7a), also has a coarse back and the neck strut. On the display context see Manderscheid (1981: 44 Fig. 14) and Schneider (1999: 50–​4).

89

90

90

Surface and Colour

Figure 27  Portrait statue of Faustina the Elder with the body type known as Large Herculaneum Woman, mid-​second century AD. Marble. H. 210 cm. Detail of the veiled head with the neck strut. Malibu, The J. Paul Getty Museum, inv. 70.AA.113

Spiral Grooves Assertions that struts were not seen by ancient viewers or that they are not meant to be ‘seen’ in the same way as the statue’s limbs and figural supports are contradicted by the high level of surface finish and decorative carving displayed by some non-​figural props.19 But why did some sculptors go to the length of decorating the (purportedly) inessential parerga of their artwork? By doing this, sculptors may have intended to draw greater attention to the struts, indicating that they were desirable elements. Alternatively, 19 Stirling (2008: 92, 108–​13 no. 6, 151 and Figs. 14–​15) shows that the struts of a statuette of enthroned Roma from the Panayia Domus at Corinth (No. S-​1999-​007), dating to the mid-​ third century AD, are decorated with an incised row of Vs. The strut underneath the stool is finished with a scallop pattern and has traces of red paint. Incised decoration is visible on the outside struts of the larger group of the Three Graces from the Baths at Cyrene (inv. 14.348), as well as on the strut of a statue of Aphrodite restored as the Muse Melpomene in St Petersburg (Hermitage, inv. A378), on which see Waldhauer (1936: 49–​50 no. 293). Here, however, it is impossible to determine whether the pattern was incised by the ancient carver or by the modern restorer, who replaced the missing part of the strut.

91

Spiral Grooves

this choice may betray an attempt to embellish –​and therefore improve – features considered unappealing. These considerations apply especially to a strut form characteristic of the Hadrianic and Antonine period, cylindrical or slightly tapered bars with a spiral motif.20 The set of examples is large and includes many statues of different subject and size.21 The most startling example of the use of decorated struts is a fragmentary hand with a discus from the Baths of Caracalla. This hand, made of Greek island marble, probably belonged to a high-​quality copy of Myron’s Discobolus. The bones, sinews, and muscles of the hand are meticulously carved. Under the discus, the remains of a broken strut, neatly shaped as a spiral groove, emerge.22 This strut continues behind the discus and, as is clear if we compare the fragment with the Lancellotti statue, must have reached down to the athlete’s right side. The effect of such a long spiral bar must have surpassed that of the Castel Porziano and the Toulouse replicas, with their heavy shafts between hip and discus. At once large and elegant, such a support challenges our assumptions about ancient visual conventions. The use of spiral struts in this period is not exclusive to freestanding sculpture. Spiral struts appear on a small number of sarcophagi, such as two mythological sarcophagi now in the Capitoline Museum and in the Berlin Antikensammlung that depict, respectively, an Amazonomachy and episodes related to the myth of Medea. The similarity of their struts supports other stylistic evidence that these sarcophagi came from the same workshop in Rome.23 The Amazonomachy sarcophagus, which scholars date to around AD 140–​50, is equipped with both quadrilateral and spiral struts. The Victory figure in the left corner holds a trophy, 20 See Hollinshead (2002a: 129–​30) and Anguissola (2013, 9–​10). 21 E.g. a spiral motif decorated at least one of the struts on the famous portrait group of Marcus Aurelius and Faustina as Mars and Venus in the Capitoline Museums (inv. 652), uncovered in 1750 in the Isola Sacra necropolis. For the modern restorations, which include part of the upper strut and the part of the lower strut joined to the wrist, see E. E. Schmidt (1968: 85, 86 n. 9). On the statue see also Stuart Jones (1912: 297–​8 no. 34), H. von Heintze (in Helbig4, vol. II, 199 no. 1394) and K. Fittschen (in Fittschen and Zanker 1985: 69–​70 no. 64). For the struts see Hollinshead (2002a: 129) and Schade (2007: 194 n. 111). A comparable design was chosen for the struts on two very fine mid-​second-​century AD versions of the wine-​pouring satyr (Paris, Louvre, inv. 2333; Rome, Palazzo Altemps, inv. 8597). 22 Rome, Museo di Scultura Antica G. Barracco, inv. 98. See W. Fuchs (in Helbig4, vol. II, 622 no. 1858) and Anguissola (2005: 327 no. 17). 23 Rome, Capitoline Museums, Palazzo Nuovo, inv. S 726; see B. Andreae (in Helbig4, vol. II, 82–​4 no. 1228), Grassinger (1999: 237 no. 94), A. Ambrogi (in La Rocca and Parisi Presicce 2012: 292–​3 no. II.2). Berlin, Antikensammlung, inv. SK 843 b; see Zanker and Ewald (2012: 354–​7 no. 20, esp. 356 for views of the the struts = 2004: 337–​9 no. 20). For comments on the workshop see Russenberger (2015: esp. 165 for the struts). Hollinshead (2002a: 130) cites another example of a spiral strut, on the so-​called Ammendola sarcophagus (Rome, Musei Capitolini, inv. 213) with battle scenes against the barbarian Gauls, dated to ca. AD 160–​70, on which see L. Musso (in La Rocca and Parisi Presicce 2010: 310–​23; 2012: 293 no. II.3).

91

92

92

Surface and Colour

secured to the background surface by a large spiral strut, carefully worked in the round (Fig. 28). On the viewer’s right, the figures of an Amazon and Victory have simple bar struts that connect the projecting limbs to the main figure or to the background.24 This difference has raised the question of whether the same sculptor executed the two sides. Since, however, all of the figures are stylistically very similar, it is unlikely that different stone carvers worked separately on this object. Equally unconvincing is the conjecture that a single designer selected different forms according to location, a twisted ‘conspicuous’ strut under the trophy, undecorated ‘invisible’ connectors behind the relief figures.25 The strut attached to the thigh of the last Amazon on the right, in fact, would by no means be invisible or, potentially, less noticeable were it finished with a spiralling groove too. We can question whether the spiral pattern embellishes an impossible-​to-​conceal strut, or if it is rather a strategy for drawing attention to an intentionally visible strut. The Berlin sarcophagus with episodes from the story of Medea raises further observations along this line. On the left-​hand side, Jason gazes with satisfaction at his happy bride Creusa, for whose sake he abandoned Medea. Jason’s right elbow is secured to the background by a cylindrical strut decorated with a spiral groove (Fig. 29). In the middle of the scene, Creusa is seen again in agonising pain, after putting on the fatal wedding gift from her rival Medea, a cursed dress. Only close inspection reveals another thin spiral strut between her left thumb and the background. On the right, Medea is fleeing in a snake-​drawn chariot. A flat quadrangular strut with incised parallel lines links the lower serpent’s wing to Medea’s flying d ­ rapery. The sculptor fastened the coils to the background at the right corner of the sarcophagus by means of both undecorated quadrangular bars and slightly tapering struts with incised lines. All the struts of this sarcophagus, albeit different in shape and treatment, are equally invisible from the front and can be seen only upon close inspection. These examples confirm that the treatment of struts did not depend primarily on their placement or necessity for stability. As I argue in Chapter 6, spiral patterns were chosen in response to idiosyncratic expressive needs, in order to either harmonise the figure with its struts or to mark the distinction between adjoining surfaces.

24 Views of the quadrangular struts are in Sichtermann and Koch (1975: Pl. 24) and Hollinshead (2002a: 131 Fig. 6.9). 25 The two hypotheses are laid out and discussed by M. Hollinshead (2002a: 130), who supports the latter.

newgenrtpdf

93

Figure 28  Amazonomachy sarcophagus from Rome, mid-​second century AD. L. 247 cm. Rome, Capitoline Museums, Palazzo Nuovo, inv. S 726

94

94

Surface and Colour

Figure 29  Medea sarcophagus from Rome, mid-​second century AD. L. 227 cm. Detail of the left corner. Berlin, Antikensammlung, inv. SK 843 b

Colour The evidence for painted struts is scant.26 Unfortunately, the disappearance of any remains of paint from most ancient marbles deeply affects our understanding of colour on Greek and Roman statues. Nonetheless, advanced scientific methods, such as the use of ultraviolet fluorescence, infrared reflection, and raking light, now allow us to detect evidence of paint that is invisible to the naked eye. There exists a relatively large sample of figural supports that still bear traces of paint. Occasionally, a tree trunk support was painted in a dark colour, so as to loosely resemble the bark of a tree.27 The attributes of painted 26 Hollinshead (2002a: 127) remarks that ‘there is no record of paint on struts’. For a general introduction to the polychromy of Roman sculpture, its techniques, and the state of research see Abbe (2015). 27 See one of the figures of a Polyphemus Group at the Archaeological Museum of Selçuk (inv. 1558–​61), reused in a Flavian nymphaeum, as well as the so-​called Slipper-​Slapper group of

95

Colour

figures were often included in the colour scheme. Both the portrait statue of Augustus from Prima Porta and the statue of a pudica Venus carved in marble of Aphrodisias and found in Nysa-​Scythopolis (Bet Shean) preserve abundant traces of colour, which extend to their figural support in form of a Cupid riding a dolphin.28 Two famous under-​life-​size statues of Venus found in Pompeii, best known as Lovatelli Venus and the ‘Bikini’ Venus, also display a colourful polychromy for both the goddess and her auxiliary figure.29 A replica of the Sciarra-​type Amazon at Écija, in Andalucia, has copious remains of the original polychromy, including red pigment on the pillar that supports the woman’s leaning pose.30 Significantly, only the front of the pillar was entirely painted, whereas no traces of colour were detected on the back. Gilding was an even bolder solution. It seems that the Diadoumenus statue found in Delos was entirely gilded, including the tall and knotty tree trunk at the youth’s feet.31 The gilding must have greatly affected the appearance of this figure, beautifully carved in Parian marble. The artwork articulated an elaborate interplay between material and narrative. First, gilding increased the cost of the statue, both because of the intrinsic value of the raw material and of the time-​consuming process, which requires expert artisans.32 Scholars have also observed that, in Hellenistic sculpture, the choice of golden skin may have served to emphasise the divine or heroic status of the subject.33 The use of gold leaf on a sculpture also complicates the relationship with the classical original –​in this case, the Fillet-​Binder that Polyclitus

28

29

30

31

32 33

Aphrodite and Pan. See Blume (2015: vol. I, 208, 244–​5 no. 63; vol. II, Pls. 123 Figs. 33.26–​7, 208 Fig. 63.50). On the colour pattern of the Prima Porta Augustus see Liverani (2003: 188 = 2004: 239) and Bradley (2009: 447–​50). For the Venus from Bet Shean (Jerusalem, Israel Museum, inv. IAA 2001–​2987) see Foerster (2005: 6, 18–​21). Naples, National Archaeological Museum, inv. 109608 and 152798. See Østergaard (2008: 46–​ 8), as well as Antonelli, Pollini, and Cancelliere (2017) for the ‘Bikini’ Venus. Cf. Blume (2012: 757 Fig. 6; 2015: vol. I, 278–​80 no. 86; vol. II, Pls. 286–​91 Figs. 86.1–​34) for an alabaster statuette of Aphrodite and Eros from Byblos in the British Museum (inv. 1914.10–​20.1), with traces of colour everywhere but the skin. Écija, Museo Historico Municipal, inv. X 1804-​197. See León (2008), Sargent, and Therkildsen (2010: 47–​8 and Fig. 18), Østergaard, Sargent, and Therkildsen (2014: 58–​9). For examples of statues with traces of red paint on the upper face of the plinth see Knoll, Vorster, and Woelk (2011: vol. I, 161–​9 no. 10, 288–​92 no. 43; vol. II, 704–​11 no. 164). See Bourgeois, Jockey, and Karydas (2011: 647–​51, 653–​4). See also Blume (2015: vol. I, 221–​3 no. 45) for the traces of gilding. It is possible that another version of the Diadoumenus was in part gilded, a miniature terracotta produced in Smyrna at about the same period as the Delos marble (New York, Metropolitan Museum of Art, inv. 32.11.2); see Murray (1885: 245), Zanker (1974: 14–​15 no. 12.1), Kreikenbom (1990: 196 no. V.32). The gilding techniques used at Delos are examined in Bourgeois, Jockey, and Karydas (2011). So Blume (2014: 178, 180).

95

96

96

Surface and Colour

cast in bronze. The resemblance of gold to the original’s shiny bronze surface probably did not escape skilled viewers. At least some of them must also have been aware that the Delian bronze alloy enjoyed a century-​old fame, such that, according to later accounts, Polyclitus himself chose it as his favourite material.34 Lastly, the choice to cover the figure in gold may be instrumental in establishing its new identity as Apollo, who stands here next to his famous olive tree, where he has hung his quiver. The association between Apollo, his Delian cult, and gold is deeply rooted in the mythographic tradition reflected by Callimachus’ hymns that elaborate on the image of a polychrysos god, ‘rich in gold’, whose body and attributes shine with the precious metal.35 The use of gold for this statue and its support may therefore depend on the wish to lend a local flavour to the composition. The effort to describe the use of colour on struts is much more complicated. It is commonly asserted that Roman sculptors painted their struts in order to conceal them, either in a dark colour or in tones that helped disguise them against the background.36 Two main obstacles prevent us from systematically addressing this hypothesis. First, the evidence available is scattered at best, with only a handful of statues having been examined for traces of colour. Additionally, traces of paint are more likely to be observed on the hair or drapery of a fi ­ gure –​surfaces with numerous cavities, where pigments may linger longer than on smooth planes. The Melpomene of Miletus, for instance, still bears traces of ochre, yellow, and reddish-​brown on her hair and garments, as well as on the lion skin and on the beard of the mask that she holds in her left hand, but none has been detected on the large, plain strut.37 Similar considerations apply to the group of Artemis and Iphigenia in Copenhagen. Although traces of pigments and gilding have been detected in several points of the human figures and the animal, there is no indication that the struts were painted too. The instrumental examination carried out by the museum’s research team has failed to provide any evidence that, as Franz Studniczka believed, the group’s heavy struts were painted in a dark colour.38 34 Pliny the Elder, Natural History 34.9–​10. 35 Callimachus, Hymn to Apollo 32–​4 and Hymn to Delos 261–​3. See the commentary by Stephens (2015: 87–​8, 221–​2). For a fragment from the Aetia mentioning a gilded cult statue of Apollo at Delos see Harder (2012: vol. I, 311–​12 no. 114; vol. II, 892–​902). 36 So Treu (1897: 202), Studniczka (1926: 142), Richter (1931: 288), Blinkenberg (1933: 23–​4), Yfantidis (1984: 77). 37 See von Gerkan and Krischen (1928: 112), Geominy (1999b: 59 Appendix VIII), and Schneider (1999: 8). 38 See Studniczka (1926: 142). On the colour scheme see Sargent and Therkildsen (2010: esp. 14, 18 Figs. 5–​6).

97

Colour

Second, any effort to reconstruct a statue’s original setting with its interplay of light, shade, and wall decoration must involve a significant degree of guesswork. Often, the lack of context prevents us from drawing any conclusions about the effects that painted struts could generate. In most cases, it remains impossible to determine to what extent the colour stood out from or blended in with the surrounding environment. The relationship between paint and context has famously provided arguments for the assessment of struts in the colossal groups of Scylla and Polyphemus from the grotto at Sperlonga. Scholars imagined that their heavy struts would be disguised in the gloomy environment of the cavern, hardly noticed by the guests gathered in the summer cenatio. Baldassarre Conticello, who published the sculptures in 1974, advocated this view by hinting at traces of dark paint on the struts of the ‘first companion’ of Ulysses in the Polyphemus group.39 Regrettably, his account of polychromy in the Sperlonga groups is unsystematic at best and it fails to explain where exactly the remains of pigment were detected and on how many figures.40 Besides, the spelunca of this luxury villa could hardly have been a gloomy cave, covered as it was with multi-​colour marbles and bright mosaics that intensified the sparkling reflections of the water. The path around the basin, the seats arranged in the interior, as well as the lavishly decorated rooms at the rear, suggest that the whole space was used for some form of reception and entertainment, perhaps involving music and theatrical performances, close to the sculptural groups. It seems that all main groups were arranged to be viewed from different angles, both distant and close.41 Even the few statues that were particularly rich in polychromy, fully revealed by instrumental analyses, cannot provide any conclusions due to the lack of comparative evidence.42 The layer of red paint preserved on the stone wedge between the left foot and the base of the eponymous Sciarra 39 See Conticello (1974: 22, 47) and the comments in Stewart (1977b: 89) about visibility inside the grotto. 40 On the traces of colour on other fragments from Sperlonga (i.e. the head of the helmsman from the Scylla group and the cloak of Ulysses in the Polyphemus group) see Cassieri (2000: 123, 134–​5), Østergaard (2010: 51, 56), Østergaard, Sargent, and Therkildsen (2014: 63–​4). 41 The architectural history and decoration of the grotto is presented in Cassieri (2000: 44, 49–​ 51); see also H. A. Weis (2000: 135 and 160 n. 147) for the display and the multiple viewpoints of the statues. J. Higginbotham (1997: 161) believes that the platform in the middle of the pool was arranged so that the couches would face the grotto –​and the people who occupied them would see its display. 42 It is likely that the long strut of a fragmentary equestrian group from the Athenian Acropolis (Acropolis Museum, inv. 700), dating to the turn of the sixth and fifth centuries BC, was painted red to imitate a wooden fence or a trunk. See Eaverly (1995: 111 and n. 155).

97

98

98

Surface and Colour

Amazon in Copenhagen, for instance, may have integrated the strut into the surface of the base.43 In general, colour could set the strut apart from the adjoining parts painted in other, more delicate tones, such as the skin of the foot, and the straps and thong of the sandal. Recent research has indeed demonstrated that for Hellenistic small-​scale terracotta figures dark tones were sometimes chosen for ‘neutral’ stability devices, such as the flat surface between the legs of a kline.44 The problems that the questions of coloured supports raises are most clearly demonstrated by the Slipper-​Slapper group from Delos (Fig. 30).45 The almost life-​size group dedicated by a Dionysios of Berytos to the ancestral gods and found in the Hall of the Poseidoniasts depicts a playful sexual interaction between Aphrodite and Pan. In the raised right hand, the goddess brandishes one of her sandals, threatening to spank her indiscreet companion. The goat-​footed, horned would-​be lover who has seized her left wrist leans on a high tree trunk covered with an animal skin, against which his hunting stick (lagobolon) is placed. The smooth strut between the two figures bears traces of blue pigment. According to excavation reports, the group was recovered among the debris of the collapsed upper floor in the House of the Poseidoniasts of Berytos. Together with the statue, the archaeologists found many fragments of red, blue, and black plaster, which likely belonged to the wall decoration of the room where the group was set.46 Although we lack enough information to both reconstruct the display context in detail and to determine the absolute and relative quantity of each colour, scholars rather arbitrarily decided that blue must have prevailed in the wall decoration and that the strut must have been painted the same colour in order to conceal it against the background.47 Albeit legitimate and per se plausible, nothing in the archaeological record warrants this reading of the evidence, based as it is on the premise that the strut and wall decoration must have corresponded to conceal the former. It remains equally likely that blue was chosen with an eye to differentiate the various parts of 43 Copenhagen, Ny Carlsberg Glyptothek, inv. IN 1568 (mid-​second century AD). See Østergaard (2010), Sargent and Therkildsen (2010: esp. 47 and Figs. 1–​4, 16–​17), Østergaard, Sargent, and Therkildsen (2014: 52–​7). For a similar strategy to integrate the stone wedge under the sole into the basis see Blume (2015: vol. I, 42, 185). The statuette of a nude male from a nymphaeum near Porta San Lorenzo in Rome (Centrale Montemartini, inv. 998) has traces of red-​brown paint both on the basis and on a strut linking the left arm to the body. 44 See Blume (2015: vol. I, 41–​2). 45 For the polychromy of this group see the excellent analysis by Blume (2015: vol. I, 207–​9 no. 33 and esp. 209; vol. II, Pls. 119–​24 Figs. 33.1–​30 and esp. Pl. 123 Fig. 33.28). 46 The finding context is presented by Bulard (1906: 610) and Zimmer (2014: 113–14). 47 K. Yfantidis’ (1984: 77) account is quintessential in this respect.

99

Figure 30  Group of Aphrodite, Pan, and Eros (so-​called ‘Slipper-​Slapper group’ or Pantoffelgruppe) found in the House of the Poseidoniasts of Berytos at Delos, ca. 100 BC. Marble. H. 155 cm with the inscribed base. Athens, National Archaeological Museum, inv. 3335

100

100

Surface and Colour

the group and mark a contrast to the goddess’ skin, Pan’s woollen thighs, and the tree trunk. Blue paint distinguished the struts from the skin and garments also in the famous marble statue of Artemis in archaic style found in Pompeii. That the statue was painted in antiquity had already been noticed at the time of the excavation in 1760.48 Writing in the late nineteenth century, Franz Studniczka noticed that the statue’s accessory features, the plinth, the tree stump behind the figure, and the rough strut under the right heel, all bore traces of black paint.49 Recent instrumental analyses allowed scholars to reconstruct in full the figure’s lavish polychromy. Indeed, the examination conducted in 2010 confirmed Studniczka’s observations that the lower part of the composition was covered in black paint, in striking contrast to the white and pastel drapery. The use of infrared luminescence led to further, unanticipated results; both the goddess’ diadem and the short struts at her wrists revealed traces of Egyptian blue. In consequence, scholars assumed that the blue painting was intended to disguise the struts.50 Certainly, blue bars would stand out from the adjoining white and pink surfaces. Whether they also stood out from the surrounding environment or merged with it is hard to determine. The statue stood in a shrine in the rear viridarium. Although the excavation reports indicate that this space was decorated in a colourful manner, very little of the wall paintings and stuccowork remains.51 Only Giuseppe Spano’s records from the early twentieth century provide a detail that may be directly relevant to our question; blue was dominant in the pediments of the shrine.52 We cannot say, however, if blue prevailed on the shrine as a whole, or if the background was painted with that colour too. Besides, it remains impossible to know if the marble left the workshop as a painted piece, if it was produced with this precise setting in mind, or 48 Naples, National Archaeological Museum, inv. 6008. On the statue, found in House VII.6.3, see Fullerton (1990, 34–​5 no. II.A.1) and Sepe (2008: 127–​9 no. C.18). Early excavations and the discovery of the Artemis are discussed by Primavesi (2008: 31–​3; 2011: 25–​7). The statue of the same type now at the Museo Nazionale Romano (inv. 568647) bears traces of two similar struts to connect the wrists to the draped body. Unfortunately, hands and wrists of the other known versions of this figure, now at the archaeological museums of Venice (inv. 59) and Florence (inv. 10701), are missing and it is therefore impossible to compare the treatment of struts throughout the series. See Fullerton (1990: 35 nos. II.A.2–​3 and 42 Figs. 11–​12). 49 See Studniczka (1888: 291) and Primavesi (2008: 40–​2; 2011: 33–​4, 63 nn. 77–​8). 50 See Brinkmann, Koch-​Brinkmann, and Piening (2011: 70–​2) and Primavesi (2011: 63 n. 77). 51 The architectural context is presented in detail by Primavesi (2008: 33–​8, 43–​9; 2011: 27–​31, 34–​9). 52 G. Spano records the finding of two fragmentary pediments (Notizie degli Scavi di Antichità, 1910: 442). Primavesi (2008: 45, 49; 2011: 36) assesses Spano’s report in relationship to previous descriptions of the house.

101

Colour

if it was first painted upon its installation in the shrine. In the absence of more contextual information, there is little reason to think that the same blue colour should serve opposite purposes on the struts and on the diadem –​ to disguise the former and highlight the latter. One further difficulty is that microscopic finds of minute particles of a certain pigment would not necessarily indicate that the pigment detected was the final colour. Instead, it could constitute a foundation layer or have been mixed with other pigments to achieve more refined tonal effects.53 In conclusion, the present, fragmentary state of the research on ancient polychromy prevents us from drawing general conclusions about the role of colour and its contribution to either highlight or downplay a statue’s struts. However, the number of objects that have been analysed scientifically is rapidly increasing. A larger dataset will hopefully provide a more complete picture of the pigmentation of struts.54 Nonetheless, the evidence of struts with carved patterns supports the theory that the decoration may have served different purposes in different contexts.

53 Egyptian blue could be used to create particular tones in a complex mixtures of pigments, as the results of recent investigations conducted on the ‘Treu Head’ at the British Museum have shown. See Verri, Opper, and Deviese (2000) and Verri, Opper, and Lazzarini (2014). 54 Hopefully, scientific examination of other marbles from well-known architectural contexts, e.g. Villa A at Oplontis, will provide new insights in the polychromy of Roman sculpture.

101

102

103

Part II

The Limits of Stone One must require from a sculptor not only the interest that results from the artwork as a whole, but also the interest of each individual part of the whole. Most often, the work of a sculptor is composed of a single figure, in which it is impossible for him to incorporate the different factors that cause the interest of a painting. (Étienne-​Maurice Falconet, Reflections on Sculpture, 1761; tr. by the author)

103

104

105

5

The Quest for Solidity

In the history of sculpture, few works epitomise the sculptor’s command of his technique better than Gianlorenzo Bernini’s Apollo and Daphne. Like the other works that Bernini made for Cardinal Scipione Borghese in the early years of his activity, the Apollo and Daphne was intended for a cultivated audience familiar with both ancient mythology and sophisticated sculptural technique. The entry in Paolo Alessandro Maffei’s 1704 Raccolta di statue antiche e moderne indicates that Bernini’s Apollo and Daphne, considered to be a ‘sovrano miracolo dell’arte’, was also appreciated for the bold use of a single block of stone for such a dynamic, theatrical composition.1 The treatment of Daphne’s hands, in particular, is a testament to artistic ingenuity (Fig. 31).2 The sculptor used a sophisticated method to carve Daphne’s splayed fingers.3 His first step was to drill out the spaces that would separate the digits. He then proceeded to hollow out the area with a file or small chisel. Short stone bridges were left in place to brace the stone while he worked on the fingers. Instead of removing these supports once the statue was finished, however, the sculptor integrated them into the work’s fiction; he disguised the struts as short branches, part of Daphne’s process of transformation into a laurel tree. The leaves and branches that grow out of and around her fingers visualise Daphne’s arrested movement, her body suddenly stiffened. As Michael Cole has observed, part of the fiction of the Apollo and Daphne ‘is that fingers pre-​existed the struts’. By means of this technical detail, Bernini added a further layer to the group’s narrative, staging a conflict between the time of the fiction and that of creation. Whereas in the story the fingers of the running nymph exist before the branches that sprout out of them, the process of carving requires that stone bridges are left in place prior to shaping the fingers in the round. The statue, in Cole’s words, ‘creates a false memory of fingers with nothing between them’.4 See Maffei (1704: col. 73) and Hermann Fiore (1997: 76). 2 On the possibility that the finishing touches were put on the group by Bernini’s gifted assistant Giuliano Finelli rather than by the master himself see Montagu (1985: 26), Coliva (1998: 268–​ 9), and Cole (2007: 60, 62, 64). The standard reference on Finelli is the monograph by Dombrowski (1997, esp. 27–​9 on the Apollo and Daphne). 3 See in particular Rockwell (1997) and Cole (2007: 60–​1). 4 See Cole (2007: esp. 63). 1

105

106

Figure 31  Gianlorenzo Bernini, Apollo and Daphne, 1622–​5. Marble. H. 243 cm. Rome, Galleria Borghese, inv. N. CV. Detail

107

From Bronze to Marble?

The case of Bernini’s Apollo and Daphne raises compelling questions about sculptural processes and the utility of struts during and after carving, as both safety braces and stability implements. Here, I address issues of technology and production as well as the principal explanations that scholars have put forward to account for the pervasive use of struts in Roman marble statuary. The primary arguments on this point are that struts were necessary to copy a bronze prototype into marble, to brace protruding extremities during the work, to guarantee the safety of a statue during transport, and to ensure stability and balance once the statue was in place. Most struts certainly had an immediate functional utility of this sort. Yet these explanations are not all equally persuasive.

From Bronze to Marble? Since Giuseppe Antonio Guattani’s comments on the discus thrower found in the Villa Palombara, the idea that struts are characteristic of Roman marble statues that copy Greek bronze prototypes has become a tenet of ancient art history. Sculptors would have avoided supports that created a ‘svistamento notabile’ for a statue initiated in stone, ‘di prima intenzione in marmo’.5 A more self-​contained composition would instead be chosen, one that would not require the struts necessary to sustain the expansive postures of the lighter bronze prototype. The variation in the placement of struts on different versions of the same sculptural type has been considered further evidence of a prototype with no struts at all, that is, a prototype in bronze.6 Occasionally, scholars have reversed the argument and argued that replica series lacking large struts must derive from prototypes initiated in marble, which were obviously created with a view to avoiding external supports.7 A life-​size figure in marble with an approximate volume of one cubic metre weighs about 2,710 kilograms (i.e. has a density of about 2.7 metric tons or about 170 pounds per cubic foot).8 Due to its weight, marble certainly poses greater challenges to the artist than bronze. The higher tensile strength of bronze and the lighter weight of hollow-​cast forms ease the task of creating statues in active poses. Nonetheless, the relationship 6 7 8 5

For these remarks see Guattani (1784: XII = in Cancellieri 1806: 31–​2). E.g. see Ridgway (1966: 35); cf. Andreae (1982: 176). See Stähli (1999: 340) on the ‘Dresden symplegma’ type of Hermaphroditus and a Satyr. There is variation among different types of stone. Most limestones weigh marginally less than 2,710 kilograms per cubic metre, while granites weigh more. See Russell (2015: 189).

107

108

108

The Quest for Solidity

between bronze and marble, as well as between Roman artworks and their alleged Greek antecedents, requires subtler and more flexible interpretations. Scholars have recognised that Roman sculptors were able to introduce striking creative modifications to the types that they imitated. It is ultimately impossible to determine whether Roman copyists were always so constrained by the form of their model that they could not make limited changes in order to avoid the use of struts, had they wished to do so. The belief that struts invariably indicate that a given statue derives from a bronze prototype is contradicted by two considerations. First, the Romans also produced marble copies of Greek statues made in the same material. Marble copies of marble prototypes also include supports for outstretched extremities or attributes. We may push the argument to a reductio ad absurdum. If we assume that struts indicate derivation from bronze, we should therefore conjecture that the Greek prototypes behind the Caryatids from Hadrian’s Villa, which include bar-​like struts for their arms, were bronze statues instead of the marble maidens of the Erechtheion.9 Second, not every statue with struts derives from one specific Greek antecedent. Large struts also feature on ideal statuary that cannot be connected to any recognisable prototypes, as well as on Roman togati and cuirassed portraits.10 Bars similar to what is called a ‘strut’ when it occurs on a marble statue are also occasionally found on Roman bronze statues. Sometimes, these are the remains of the casting process. On occasion, the artist chose not to remove the sprues used to channel the molten bronze into his moulds, leaving them in place after the piece was finished.11 From a functional point of view, these devices hardly compare to the bridges that connect the limbs of a marble statue. Nonetheless, their impact on the composition is similar, as elements unrelated to the narrative and hinting at the technical process. An example almost contemporary with Bernini’s Apollo and Daphne clarifies this point. The late mannerist bronze sculptor Adriaen de Vries is known to have often left the sprues of his bronze compositions in place, disguising the metal tubes as branches, vines, and other motifs rather than sawing 9 See also Geominy (1999b: 49). The struts for the attributes and drapery of the Caryatids from Tivoli (inv. 2236, 2238, 2239) are visible in the plates published by E. E. Schmidt (1973: Pls. 6, 7a, 8a, 11, 13a–​b, 27a–​b). 10 This point has been raised compellingly by Hollinshead (2002a: 120, 147–​8). 11 E.g. a miniature group of wrestlers at the Ny Carlsberg Glyptotek in Copenhagen, dated between the first century BC and the first century AD (inv. 848, H. 8.4 cm with the base), on which see Østergaard (1996: 269 no. 173). For other bronze statues that include connecting rods, possibly the gates or channels through which the molten bronze flowed into the mould see Thouvenot (1927: 40 no. 172 and Pl. XIII) and Klatt (1995: 549–​50 no. A14).

109

From Bronze to Marble?

them off the finished piece. His Theseus and Antiope, the Wrestlers, and the Hercules Pomarius all rely on this peculiar use of channels.12 Roman bronzes also include other types of struts. This is the case, for instance, in a life-​sized bronze statue of a boy said to come from the island of Rhodes and now at the Metropolitan Museum of Art (Fig. 32). With his broad face and distinctive hairstyle, the boy resembles the young princes from the family of Augustus.13 He wears a fine himation with woven decoration, gracefully draped over his left arm and worn at the hip. This statue was cast in sections using a combination of the direct and indirect methods of the lost wax process. Metallurgical joins are visible, as well as two bronze struts that connect the folds of the drapery on the boy’s left side. Scholars have debated whether these were merely ornamental or intended to secure the statue’s safety. It has even been suggested that this bronze is the copy of a marble work and therefore retained the struts of its model. More likely struts were used here as spacers when the drapery was cast. They would keep the wax model in the correct position and stabilise the lower section of the figure, with its elaborate arrangement of interlocking folds.14 The choice not to remove the struts after the casting and joining process had been completed, however, probably depends on aesthetic factors. By the time this statue was cast, struts had already become a familiar feature and a commonplace addition to marbles of various size and quality.15 Struts between the calves of a standing figure, such as those that feature on many Roman naked marble statues, are also occasionally found on bronzes, where they are probably meant to increase the figure’s stability by bracing its extremities.16 Interdigital connectors, albeit infrequent in 12 See Cole (2007: 62). On Adriaen de Vries’ Theseus and Antiope (Royal Collection Trust/​Her Majesty Queen Elizabeth II, inv. 57961), Wrestlers (Stockholm, Nationalmuseum, inv. Drh Sk 73), and Hercules Pomarius (Prague, National Gallery, inv. VP400), see F. Scholten (in Kommer 2000: 232–​4 no. 12, 300–​5 nos. 34–​5) and Bassett (2008: 11–​16, 207–​12). 13 New York, Metropolitan Museum of Art, acc. no. 14.130.1, Rogers Fund 1914. See Richter (1915: 149–​52 no. 333), Lahusen and Formigli (2001: 83–​4 no. 35), S. Hemingway (in Daehner and Lapatin 2015: 260–​1 no. 35). On the identification of this figure as a young prince from the family of Augustus (Lucius or Gaius) see Zanker (1989). 14 The use of struts in relation to the casting techniques of this statue is thoroughly examined by Hemingway, Milleker, and Stone (2002: esp. 204–​5). 15 For a parallel see the strut in the folds of the marble statue of Ulysses from Baiae, which is clearly visible in Zevi (1983: Pl. 86). 16 See the short metal bridge between the calves of a fine classicising figure of an ephebe at the National Archaeological Museum of Athens (inv. X 26087), produced between the end of the first century BC and the first decades of the following century. See R. Proskynitopoulou (in Kaltsas 2004: 101–​3 no. 7; 2012: esp. 9, 20 n. 15) for the strut (measuring 2.1–​2.6 cm) between the youth’s legs.

109

110

Figure 32  Statue of a boy, Augustan period. Bronze. H. 132.4 cm. New York, Metropolitan Museum of Art, acc. no. 14.130.1, Rogers Fund 1914

111

Carving

absolute terms, are the struts that occur most frequently on bronzes.17 These metal connectors probably served to ensure the grip and fasten an attribute –​a hypothesis reinforced by the fact that fingers are occasionally modelled on the outside, but left undifferentiated on the inside of the hand, so as to create a hollow with smooth sides. The likely use of interdigital struts as attribute holders on bronzes encourages us to consider the possibility that similar devices may sometimes have served to append a tool or hold a vessel by its handle also on marble statues.18 These examples demonstrate that connecting bars were by no means exclusive to marble statuary. Although rare, they also occur in bronze statues, especially those produced around the age of Augustus or in the following decades. They could serve various functions, ranging from stabilisers for delicate parts of the composition to attribute holders.

Carving This evidence encourages us to reconsider the issues of production and technology in closer detail. In this and the following sections I concentrate on the production stages of marble statues and consider whether and in which measure devices such as struts could be helpful to each of them. The life of a marble figure starts with its actual carving, either in the sculptor’s workshop or in its intended context of display. Pieces such as the boy from Rhodes now at the Metropolitan Museum of Art indicate that connecting devices may be functional to the safety of a bronze during the casting process. Regardless of whether it is a copy of a bronze prototype or not, a marble statue also requires that the sculptor leave some supports in 17 E.g. a bronze statue of Dionysus said to be from Syria and dated to the late Hellenistic or early imperial period, which includes a strut in the left hand between the thumb and the small finger (Washington, DC, coll. Susanne K. Bennet). The god’s left arm hangs at his side, the fingers curled, probably around a large object such as a thyrsus or a kantharos. See Bennett (1994: esp. 9–​11, 80), Mattusch (1996: 224–​31 no. 23, esp. 224, 227, 231 n. 3 with detail view at p. 226 Fig. 23h), Hemingway, Milleker, and Stone (2002: 204–​5). Close parallels are the bar between the thumb and forefinger of a hand in the Tampa Museum of Art (inv. 86.142, L. 10.9 cm), discussed by Mattusch (1996: 226 Cat. 23 Fig. 1), and the double metal bridge between the thumb and forefinger and between the small finger and the palm of a life-​size right hand with arm recently seen on the antiquarian market in Munich. The latter probably belonged to the statue of a youth used as a lamp holder, with its hand prepared to hold the attribute. In the catalogue Gorny & Mosch, 214 (2013: 41 no. 23) the piece is dated to the beginning of the imperial period or earlier. 18 For interdigital struts as attribute-​holders on marble statues see N. Cassieri’s discussion (2013: 37–​44 nos. 14–​16) of three nude male figures at the National Archaeological Museum of Formia (inv. 15423, 15424, and 147633).

111

112

112

The Quest for Solidity

place until the work has been completed. A number of unfinished marble statues are representative of a process that requires the use of struts. There, the sculptor worked from a frontal plane through the depth of his block, gradually carving the figure free of the marble matrix. Struts were left in place to secure portions of the figure to the background.19 This measure helps to avoid damage from chisel strokes while the rest of the statue is being hewn from the stone. Of course, not every freestanding statue has to be worked in this way. Nonetheless, it is likely that sculptors who followed a different procedure, gradually progressing around all sides, also left connectors in place until most of the figure had been roughed out. This explanation may apply to several cases, especially to the small interdigital struts that had no practical function as attribute holders. The minute struts between the toes of the colossal Polyphemus at Sperlonga, for instance, may have been left in place as a precaution during the carving (Fig. 33).20 Not only was the cyclops’ left foot finished with exceptional care, highlighting the bones and sinews, but a tuft of hair was carved in painstaking detail on both the upper surface of the foot and on each toe. The tiny connectors would protect the toes from the tool strokes as the artist finished this delicate detail.21 These considerations, of course, do not 19 See the unfinished mid-​second-​century AD funerary statue of an athlete found in the Kerameikos (Athens, National Archaeological Museum, inv. 1662), with a large strut left in place to secure the right hand to the background. See Kaltsas (2002: 352 no. 743) and Hollinshead (2002b: 225–​6). Other scholars, like Conze (1911–​22: 74 no. 2023) and von Moock (1998: 138 no. 276), suggested that this piece is a high relief in which the background has not yet been worked. A comparable example is the unfinished athlete (a version of the Discophorus by Polyclitus) from the sculptor’s workshop at Aphrodisias, whose extended lower left arm is still embedded in a large marble projection. See Rockwell (1991: 134, 135 Figs. 11–​12). For unfinished statues carved front to back see also Russell (2013a: 320). 20 This explanation has been put forward by Hollinshead (2002a: 150). P. C. Bol (1972: 96) also acknowledges this function in the Antikythera corpus. For more views of Polyphemus’ foot see Conticello (1974: Pls. 19–​20). An intriguing case of interdigital connectors is attested in a different geographical and cultural context, Gandhara art. There, the statues of Buddha often display screens of stone between the fingers, which scholars believe were protection measures required by the stone. Mention in the texts of the ‘webbed fingers’ among the marks of Buddha may derive from the experience of observing and describing the many statues that shared this technical feature. See Bussagli (1959: 218). 21 Apparently, the safety of fingers was also a concern for the restorers of ancient sculpture, who often left one or more connectors in place between the fingers or the toes. See Kalveram (1995: 206–​7 no. 90) for the group called Orestes and Pilades at the Louvre (inv. MA 81), S. Pafumi (in Gasparri 2009: 77 no. 32, 271 Pl. XXX.5; 80–​2 no. 35, 277 Pl. XXXIII.5) for two versions of the Capitoline Venus in Naples (inv. 6286 and 6299), Waywell (1986: 76 no. 11, Pl. 50) for a version of the Medici Venus at the National Archaeological Museum of Athens (inv. 3524), Fendt (2012: vol. II, 370–​4 no. 89) for the Amazon type Sciarra in Berlin (Antikensammlungen, inv. Sk7), W. Geominy (in Knoll, Vorster, Woelk 2011: vol. II, 673–​7 no. 156) for a dying Niobid in Dresden (Albertinum, inv. Hm 124).

113

Carving

Figure 33  Statue of Polyphemus from Sperlonga, left foot. Sperlonga, National Archaeological Museum

explain why so many struts were not removed once the statue was finished, including those that were not placed at particularly fragile points. Once the statue of Polyphemus was finished, securely placed at the rear of the grotto at Sperlonga, it would be relatively easy for a skilled artist such as the one who had carved the feet to remove the struts with a file or rasp. One reason the struts were left in place is probably the general indifference to struts that also accounts for the silence of our ancient written sources,

113

114

114

The Quest for Solidity

which apparently found these devices unworthy of mention. It is also possible that struts were sometimes left in place to highlight the difficulties of a composition and draw attention to its delicate carving. After all, the daring carving of the splayed toes does not enhance the subject matter or meaning of the Polyphemus; the feat is irrelevant to the mythological narrative. The evidence that struts could serve as safeguards against the percussive strokes of the sculptor’s tools, however, does not apply to the whole sample. As sculptors of any period know well, large undercuts in the stone count among the most challenging aspects of their art. In the case of statues such as the Vatican Apoxyomenos or the discus thrower from Castel Porziano, the strut, albeit large compared with the limbs, is ultimately only a thin stone line enclosing a large triangular undercut that had to be carefully hollowed out. Struts like these are not functional during the process of carving. Rather, their shaping and smoothing poses a further threat to the safety of a statue.

Transport It has often been argued, in very unspecific terms, that struts were first and foremost security measures for transportation of the statues.22 According to this argument, even minute connectors such as the interdigital struts between the toes of the Sperlonga Polyphemus are evidence that the statue was carved far from its context of display.23 This assumption has had meaningful consequences in the regional surveys of Roman statuary. The large number of struts on the statues from a certain site or area has sometimes been thought to demonstrate that the region mainly relied on shipping from distant regions of the empire for the sculptural décor of its buildings.24 Of course, virtually every statue that was not carved in situ had to be moved at some time, whether the workshop was distant or nearby. Although marble is remarkably resilient compared with other stones, it lacks tensile strength. This means that the thinner and more distant from the main core the carved forms are, the less the figure is able to withstand shocks and vibrations.25 According to a letter written in 1581 by Simone Fortuna to 22 E.g. Treu (1897: 202), Lippold (1923: 43, 72–​3, 133–​4), Studniczka (1926: 142), Richter (1954: 31), Linfert (1979: 781), R. Bol (1984: 21). The transport argument is reviewed by Hollinshead (2002a: 140–​1). 23 For instance, according to Stewart (1977b: 89) and Rice (1986: 234 n. 48). 24 See Slavazzi (1996: esp. 140) about sculptural copies in the province of Gallia Narbonensis. 25 For comments on this point see Claridge (1990: 136) and Hollinshead (2002a: 122).

115

Transport

the Duke of Urbino, Giambologna discussed this point, commenting on the relationship between material and ingenuity. Giambologna complained that, unlike bronzes, marble statues ‘run the risk of getting broken, whenever they are moved about from place to place, as well as from the least accident or mishandling. And a man cannot guarantee to produce in marble anything with as extraordinary a composition as his imagination suggests (capricci fuori dall’ordinario, come egli ha fantasia)’.26 Evidence from the early modern period indicates that the apparatus for moving statues, which seems to have changed little over time and remained essentially identical from the Cinquecento to Canova’s period, was based on a rudimentary technology. Accidents during transportation occurred frequently, and even the most famous pieces, moved with particular attention, risked damage.27 Thus Giambologna’s preoccupation with the material limits of the stone. Although it is indisputable that movement would endanger a marble statue, addressing the role of struts requires that we define the set of operations that go under the label of ‘transport’ in closer detail.28 Statues could be moved either over long distances or locally, loaded onto boats or transported by land. At the destination, the statue had to be hoisted into position. Furthermore, the vast majority of statues that have survived likely changed their owner and location several times. The base of the Vatican Apoxyomenos, for instance, bears traces of the sort of reworking required by a new placement.29 The platform was narrowed at the sides and sockets were prepared for three or four sets of clamps that would fasten the work to a pedestal. In general, it is difficult to determine whether a given statue was imported. Sometimes, archaeologists can demonstrate that both the stone and the style of a statue point to the same origin for a work, far from the place where it was found.30 In Roman times, however, the movement of craftsmen and the long-​distance supply of raw material were closely related.31 Therefore, 26 The quotation is taken from Avery (1987: 251–​2); the document is also cited by Helmstutler Di Dio (2015: 179). 27 See Bourgeois (2003: 150), with particular reference to the famous transport of the Medici Venus, which reached Florence from Rome in 1677, and of the Farnese Bull brought from Rome to Naples in 1788. 28 The transport of marble in the Roman world has been examined at length by B. Russell (2013a: 95–​104; 2015). For calculation of the costs of carving, transport, and set up of the marble floor and wall decoration at the Villa A at Oplontis see Fant and Barker (2016) with the Appendix by Barker (2016) for an estimate of the required man-​power. 29 See P. Liverani (in Moreno 1995: 201 no. 4.29.4). The base of the Discobolus from Castel Porziano was also extensively repaired. 30 Russell (2013a: 337–​8) lists examples of statuary in marble of Thasus and Docimium which are likely to have been imported as finished works. 31 See the summary in Russell (2013a: 329–​36; 2015: 190–1).

115

116

116

The Quest for Solidity

the distinction between imports and the work of travelling craftsmen is often a matter of guesswork. The shipping of a bulky marble statue was much more demanding than for other artworks. Panel paintings or mosaic emblemata, for instance, were light and easy to dispatch, stored flat in thin cases.32 Recent research in the field of early modern art has demonstrated that the transport of a marble statue could instead easily require months of effort even for relatively short distances. Even the route between the marble quarries at Carrara and the city of Florence required a complex set of operations. The block of marble was carried to the closest port, loaded on a flat-​bottomed boat by cranes, and transferred on a larger sea-​faring ship to either Pisa or Leghorn. There, the block would have to wait until the river Arno was high enough to set sail for Signa, near Florence, on a smaller boat. At Signa, it would be unloaded and brought over land to Florence.33 Regrettably, we do not know the technical details of how marble statues were packed and stowed in antiquity. It is also impossible to determine whether significantly different practices were followed in different periods or regions. Spectacular evidence for the trade of artworks is provided by shipwrecks, such as the Antikythera and the Mahdia cargoes.34 The Antikythera ship was probably travelling from the eastern Aegean towards Italy in the second quarter of the first century BC, carrying a large set of marble and bronze statues, both old and new pieces. Recent examination of the corpus of marble statuary from Antikythera has led to the suggestion that the statues may perhaps have been stowed upright in the ship’s hold, secured to low orthogonal bases with a depression on the upper sides for the inset of the statues’ plinths.35 The Mahdia wreck, discovered off the east coast of Tunisia, also yielded a large quantity of sculpture, as well as architectural pieces and ­durable luxury goods, including furniture, lamp-​stands, and vessels. The last port of call was probably Athens and the ship was also en route to Italy before being blown off course, probably about 80–​70 BC.36 In theory, the new pieces of statuary, produced with a view to long-​distance trade, could 32 At Kenchreai, one of the two harbours of Corinth, archaeologists found a set of opus sectile glass panels still in their wooden crates. Each crate held two panels, packed face to face with perhaps a little straw or seaweed between. See Ibrahim, Scranton, and Brill (1976: esp. Figs. 9–​12 and Drawing IVB). 33 See Helmstutler Di Dio (2015: 171), as well as Jones (2015: 123–​4) for the transport of Istrian stone in sixteenth-​century Venice. For the logistics of marble transport in antiquity see Russell (2013a: 100–​8, 111–​12). 34 For a review of the shipwreck evidence see Pensabene (2013: 147–​95) and Russell (2013a: 112–​ 18; 2013b). 35 See Vlachogianni (2012a: 43). 36 For the Mahdia wreck and its materials see Hellenkemper Salies (1994).

117

Transport

throw light on questions about transport technology. Unfortunately, closer comparison between the two cargoes is hardly informative to our purpose. The Mahdia ship, in fact, carried only a limited sample of large-​size freestanding marble statuary. It remains therefore impossible to determine whether the use of many struts observed in the Antikythera marbles was widespread practice for travelling statues. Not all sculpture was moved by water, and even when it was, part of the journey still had to be conducted overland.37 Few quarries were located on the coast and landlocked production centres had to transport their commodities initially by land. Land vehicles could not carry the huge loads that the naves lapidariae managed, so marble sculptures must have been moved as single items or small sets. Travel over uneven roads is likely to be even riskier for such vulnerable goods than travel by water. A wall painting from Stabiae seems to depict a statue, with its base, as having just been unloaded from a two-​wheeled wagon pulled by oxen.38 The statue had clearly been held upright on the cart during transportation. However, it seems reasonable that over long distances statues would be stowed in crates, probably lying down. The question at stake is whether any of these operations required or recommended the use of struts. The evidence that most struts were never removed once the statue was in place argues against the idea that they were primarily intended as safeguards for transport, to be removed at ­destination. However, we should not forget that chiselling a strut away could also damage a newly installed statue. More significantly, in many cases struts exceed the size needed for mere safety requirements. Some  –​for example, the long strut formerly under the arm of the Vatican Apoxyomenos –​are even longer and thinner than the limbs that they purportedly support, thereby posing a further threat to the overall solidity of the statue.39 Most struts do not indicate that a work was left unfinished. Rather, they occur on completed statues, with carved details and smoothed surfaces. The struts themselves were in most cases given the form of rectangular prisms or 37 Raepsaet (2008: esp. 591–​2) and Russell (2013a: 96–​105; 2015: 196–​8) provide overviews of the land transport of stone. Its challenges are examined in Raepsaet (1984) with regard to the transport of column drums from the Pentelic quarries to Eleusis in the fourth century BC. 38 For the construction site painted at Stabiae see Adam and Varène (1980: 218 and 216–​17 Fig. 2), Di Pasquale (2004: 190–​1) and Russell (2015: 197). P. Miniero Forte (in Conticello, Franchi dall’Orto, and Varone 1990: 240, 242–​3, 245 no. 169) thinks that the object unloaded from the cart is instead a block of stone. A late antique mosaic in Tunis, on which see Russell (2013a: 99 Fig. 4.1), depicts a similar cart, pulled by horses, for the transport of a small column. 39 For similar comments see also Geominy (1999b: 50 and n. 56) and Hollinshead (2002a: 142).

117

118

118

The Quest for Solidity

other regular shapes, with flat surfaces and sharp angles. Obviously, there would be no point in a sculptor tidily shaping a quadrilateral bar or an arched connector if he expected it to be removed once the sculpture had been set up. Also, the percussive strokes of the sculptor’s pick and chisels used to rough out and shape the long strut added considerable strain to the figure as a whole and to its outstretched extremities. Struts, in fact, do not seem to be the safest choice to ensure that a statue with extended limbs or daring undercuts would survive transport intact. Alternative and more effective methods exist to reduce the risk of breakage during transport. One is that the statue could be shipped unfinished, whether as a roughed-​out or near-​finished piece. In the case of figures with limbs, drapery, or attributes projecting outside the main core, their safety could be increased by delaying work on the most delicate parts and leaving rough screens of marble in place where the undercuts would later be carved. The archaeological evidence indicates that, in the Roman world, stone intended for sculpture was often quarried and dispatched as roughly squared blocks, a format that allowed it to be used for a ­variety of projects.40 A handful of finds indicate that detailed work was occasionally also undertaken at the quarry, with the statue leaving the site as a near-​finished piece with only limited sections awaiting completion upon installation.41 Of course, we also have plentiful evidence that finished marble statues were transported. Shipwrecks, for instance, have yielded finished statuary, both older works and new products. The letters that Cicero wrote to Atticus in 68–​66 BC describe the long-​distance commissioning of sculpture more or less of the kind that the Mahdia and Antikythera ships carried.42 Probably, sculptors catering to distant markets were aware of the challenges of transportation and worked accordingly. Although undocumented, it cannot be excluded that the sculptor was responsible for delivery

40 See Russell (2013a: 317–​18; 2015: 191–​3). For the unshaped blocks in fine-​grained Parian marble found in and around the Fossa Traiana, probably destined for statue production, see Pensabene (1994: 121–​2 esp. nos. 54–​6) and Bruno et al. (2002: 347–​54). 41 See Russell (2013a: 324–​6). 42 Letters to Atticus 1.8.2, 1.9.2, 1.10.3. In a passage from his Periplus (1.4–​2.1), Arrian requests Hadrian to dispatch ‘a statue of Hermes about five feet tall … and another, four feet tall, of Philesios’ for a temple in Trapezus on the Black Sea. Philostratus (Life of Apollonios of Tyana, 5.20) describes a merchant setting sail from Piraeus with a cargo that included votive statuettes in gold, stone, and ivory. Robert (1962: 42 and n. 13a) mentions an inscription from the region of Prusa in Bithynia recording the dedication of a statue ‘brought from Docimium’. On the transport of finished statues see Russell (2013a: 313, 315, 336–​8; 2015: 193–​6) and the examples listed by Arata (2005).

119

Transport

to the closest port or handling centre, as was customary in later periods.43 Sculptors working with a view to long-​distance transport could adjust the style and composition so that their figures would withstand the journey. Self-​contained forms without large extremities could limit the risk of damage.44 Later sources are sometimes explicit about this choice. In 1674 the Spanish noblemen Don Iñigo Melchor Fernández de Velasco, Duke of Frias, wrote from Madrid to Genoa requesting the execution and shipment of eleven statues. The duke not only included a drawing of the pieces, but he also gave instructions that they should be carved ‘so that no point of the sculpture sticks out, because if a part comes out it will be broken or damaged’.45 Another case study from a later period elucidates this point. Giambologna’s group of Samson and the Philistine was commissioned in 1562 and o ­ riginally set up on a fountain in Florence.46 The pyramidal block of marble is pierced in the lower part with a series of interstices. Samson’s raised right arm and the jawbone that he holds as a weapon are cut entirely away from the core. His bag too protrudes from his side and hangs at the extremity of the composition. In 1601 the group was sent to Spain as a diplomatic gift to the Duke of Lerma, the favourite and chief minister of Philip III of Spain, from the Grand Duke Ferdinando I of Tuscany. The Grand Duke and his entourage had carefully considered the choice of a suitable gift. The statue was to be installed in Lerma’s gardens at Valladolid, where skilled craftsmen were unavailable. Therefore, a work that was fully finished and ready for shipment had to be chosen.47 We do not know whether Giambologna’s bold creation survived the journey intact. Pope-​ Hennessy reported that the statue was made of multiple pieces of marble and noted ‘old fractures’ perhaps caused by its travels.48 Soon afterwards, Lerma requested another 43 Burford (1969: 188) and Russell (2013a: 103) mention a specific provision, albeit not referring to statues, in one of the Epidaurian contracts stating that the block of stone is to be delivered undamaged. A passage by the second-​century AD jurist Gaius, cited in the Digest (19.2.25.7), deals with provisions for the transport of columns. The men entrusted with the task, whose relationship to the carvers’ workshops remains unclear, would be held responsible for any damage caused by lack of diligentia. See Pensabene (2013: 148–​9). For the early modern practice see Avery (2015: 101–​2) and Helmstutler Di Dio (2015: 179). 44 See the remarks by Ridgway (1966: 31). 45 See Silva Maroto (1995: 216 doc. 11); the translation is taken from Helmstutler Di Dio (2015: 179). 46 London, Victoria & Albert Museum, no. A.7-​1954. See Avery (1987: 75–​7). 47 In a letter to Grand-​Duke Ferdinando I written in August 1601, cited by Goldberg (1996: 529 n. 6), Francesco Guicciardini voices his concern that in Valladolid there are ‘né marmi né maestri, che li sappiano lavorare’. 48 See Pope-​Hennessy (1964: vol. II, 460).

119

120

120

The Quest for Solidity

fountain figure as a pendant to Giambologna’s group. In this case, a new piece was carved, purposely designed to be sent on the long voyage to Spain. Cristoforo Stati’s Samson and the Lion was ready to leave Florence in late 1607. This group is self-​contained, with limbs close to the body and the lion strategically placed in the centre of the composition, to provide a further point of balance to the lower section.49 Kelley Helmstutler Di Dio has recently argued that the striking stylistic difference between the two statues meant to be displayed as pendants is due to the different circumstances of their production.50 Stati’s work was composed in such a way as to resist the shocks and strains of a long journey. Another way to increase the chances that fully carved sculptures would survive transportation intact is that of carving the projections as separate pieces, to be joined to the main figure at the destination.51 The technique of joining is recorded as early as the archaic period, notably for the korai from the Akropolis.52 In most cases, sculptors added the head from the bottom of the neck, or the forearm where it emerged from a sleeve with the socket-​and-​tenon method. The early classical torso of a warrior now at the Acropolis Museum indicates that an outstretched arm, which projects from the core, could also be fastened to the shoulder by means of a socket-​and-​ tenon join with vertical cross-​pin.53 The corpus of classical and Hellenistic sculpture offers a large sample of projects that made extensive use of joins.54 Most of the freestanding statues from the Mausoleum of Halicarnassus were in fact carved in more than one piece, joined together by means of sockets, dowels, or clamps, often secured by lead. This technique becomes extremely common in the Hellenistic period, and seems to be prevalent in Rhodes and Delos. In the sample of Hellenistic sculpture from Rhodes, joins are observed not only in extended limbs, but also in small details such 49 Chicago, Art Institute, Chester D. Tripp Fund, no. 1997.335. See Wardropper (1999). That Stati was capable of creating more dynamic figures, with limbs stretching out of the core, is demonstrated by his Orpheus (New York, Metropolitan Museum of Art, acc. no. 41.100.242), on which see Wardropper (2011: 105–​7 no. 33). 50 See Helmstutler Di Dio (2015: 175–​9). 51 On the Greek and Roman techniques of making joins in marble statuary see the fundamental body of work by A. Claridge (1988; 1990; 2015: 109–​10). See also the overview in Burnett Grossman (2003: 4–​5, 59, 77–​8, 98). 52 For the Acropolis korai see Lechat (1903: 227–​42), Dickins (1912: 37–​9) and Claridge (1990: 137–​42). 53 Athens, Acropolis Museum, inv. 599. See Claridge (1990: 140 Fig. 5) and Brinkmann (2003: no. 51). 54 See Adam (1966: 80–​2) and Stewart (1977a: 44). A number of publications have addressed the technique of making joins in specific geographical contexts: Waywell (1978: 63–​5) on the Mausoleum at Halicarnassus, Hofter (2011) on Pergamon, Merker (1973: 8–​9) on Rhodes, Marcadé (1969: 109–​12, 318–​19, 415–​16) and Jockey (1999) on Delos.

121

Transport

as fingers.55 The general impression, however, is that this technique reached its peak in the late Hellenistic and early imperial periods and became less common by the beginning of the second century AD, when the size and quality of marble blocks improved substantially.56 This method presents several advantages. First, it allows sculptors to adapt stock bodies to the requirements of individual commissions. Secondly, it drastically cuts down the costs of both the raw material and the labour –​a point to which I will return in Chapter 7. Thirdly, a statue made of separate pieces can be more easily packed, stowed, and shipped.57 It is impossible to determine whether piecing separate parts was actually considered a more suitable option by the carvers who worked for distant markets. Certainly, the choice depended on the availability of skilled craftsmen at the destination and on the internal organization of the workshop, which could have local branches or rely on travelling sculptors. The largest sample of marble sculpture purposely produced for the overseas market that we know of, from the Antikythera shipwreck, used both large struts and joins for outstretched extremities. A number of pieces were prepared with square sockets on the contact surface, to receive the tenon of the adjoining part.58 We have seen already that the statue of the nude boy in a bent-​over position includes large struts that connect the limbs, small interdigital struts, and sections that were worked as separate pieces, such as the top of the head and possibly the right forearm with the hand. The three largest pieces recovered from the wreck, a poorly preserved version of the Hermes Andros-​Farnese type and two colossal figures of Heracles rely on joins instead of struts.59 The Antikythera cargo also includes statues that combine the techniques of struts and joins. The so-​called Achilles, part of a set of sculptures of Homeric subjects, was made of two blocks joined at the hip, while the hero’s right arm and hand are linked to the torso by two struts.60 The same strategy is used for an over-​life-​size statue of Ulysses. Two 55 For instance in Merker (1973: 8, 26–​7 no. 12 and Pl. 4 Fig. 10, 28 no. 48). 56 The chronology of this technique is examined by Claridge (1988: esp. 148–​51). 57 P. C. Bol (1972: 95–​6) and Merker (1973: 8–​9) discuss joins as a means to limit the risk of breakage during work and transport. 58 See P. C. Bol (1972: 94–​6) and E. Vlachogianni (2012a: 39–​41; in Kaltsas, Vlachogianni, and Bouyia 2012: esp. 112–​13 nos. 60–​1). 59 The head and shoulders of the Hermes were worked separately. The left arm and the cloak were attached to the upper part of the body as individual pieces. The surface between the shoulders of one of the Heracles statues, a version of the Farnese Heracles, was prepared for an inset head. Both arms of the other colossal Heracles were worked as separate pieces. See P. C. Bol (1972: 47–​8 no. 38, 48–​9 no. 23, 53–​5 no. 24 and Pls. 24, 25, 28). 60 For the statue of Achilles (Athens, National Archaeological Museum, inv. 5746) see P. C. Bol (1972: 79–​80 no. 27 and Pls. 46–​7) and E. Vlachogianni (in Kaltsas, Vlachogianni, and Bouyia 2012: 107 no. 52).

121

122

122

The Quest for Solidity

blocks are joined at the hip and a short strut connects the right arm to the body. Another strut is placed between the tree trunk support and the right leg.61 Both these statues show a relatively self-​contained pose; the limbs do not project far outside the core. A second, poorly preserved statue identified as Ulysses follows a different design, with both arms detached from the body and the right arm extended below and outwards in an emphatic gesture, perhaps of bewilderment and surprise.62 The sculptor entrusted the safety of both the raised right arm and the left hand to dowels. A single, short strut is located between the leg and the tree stump. This indicates that both methods –​those of struts and joins –​could be used for statues made for long-​distance trade in the same material (Parian marble), in the same period (early first century BC) and probably in the same production centre (Delos).63 The best-​preserved marble works from the Antikythera cargo, the nude boy and the Achilles, show that struts attached to statues made for transport overseas could also be finished and given a regular shape. They were, in other words, treated as part of the composition rather than as functional elements to be removed at a later stage. I conclude this examination by returning to the case of Sperlonga. Recent analysis has proved that the fine-​grained white marble used for both the Scylla and the Polyphemus groups is Docimium marble from the Phrygian quarries of İscehisar. Its homogeneity suggests that almost all the blocks were extracted from the same area of the same quarry.64 The inscribed signature states that at least the group of Scylla, located in the middle of the pool, was the work of three Rhodian masters. Notwithstanding the fact that the marble was imported, that the sculptors came from the eastern Mediterranean, and that an array of struts of impressive number and size were used, there is increasing consensus that the groups were carved or finished on site as part of a unitary project, as suggested also by the many marble chips discovered around the supporting plinth of the Scylla group.65 61 Athens, National Archaeological Museum, inv. 5745. See P. C. Bol (1972: 78–​9 no. 28 and Pls. 44–​5) and E. Vlachogianni (in Kaltsas, Vlachogianni, and Bouyia 2012: 106 no. 51). 62 See P. C. Bol (1972: 80–​2 no. 29 and Pl. 48). 63 For a summary on the issues of material, date and provenance, see Vlachogianni (2012b: esp. 63–​4, 70–​2). 64 See Bruno, Attanasio, and Prochaska (2015: esp. 377, 384–​90). This article also includes a brief discussion of the state of research on the date, style, and provenance. Table 2 indicates that all blocks used for the mythological groups come from the same source within the quarrying site, with the exception of Polyphemus’ right arm, the headless Ulysses from the Palladium group, the head from the Pasquino group, and the statue of Ganymedes. 65 See the fundamental work by Kunze (1996: esp. 158–​9, 181–​2). In particular on the marble chips discovered around the plinth of the Scylla group see Andreae (2003: 56) and Bruno, Attanasio, and Prochaska (2015: 377 n. 13).

123

Transport

Figure 34  Drawn reconstruction of the ship of Ulysses from the group of Scylla at Sperlonga

The sculptures were set up with an eye to their respective positions and the view from the triclinium.66 The intricate group of Scylla arose drama­ tically from the centre of the pool (Fig. 34). The Blinding of Polyphemus, a ­colossal sculptural tableau featuring numerous individual figures, was set in a rocky alcove at the back of the main cave (Fig. 35). The task of adjusting the figures to their setting, inevitably in consultation with the other craftsmen at work in the grotto, must have required at least some carving on site. The evidence from Sperlonga contradicts the assumption that struts indicate long-​distance shipment, even when both the stone and the craftsmen are known to come from a different region. 66 See Kunze (1996: 159, 181–​2).

123

newgenrtpdf

124

Figure 35  Drawn reconstruction of the group of Polyphemus at Sperlonga

125

Stability

Stability Struts seem to have limited utility for transportation compared with other techniques, such as joining or, more simply, dispatching unfinished pieces. Unfortunately, we know too little of ancient techniques for packing and stowing to examine this question in closer detail. The last stage of the transport, the actual hoisting into place of a statue, is equally elusive. At times, the ropes or hawsers used to lift a piece left traces on the marble surface.67 At Sperlonga, archaeologists were able to identify the recesses that held the beams that supported the machina tractoria, the pulley used to lift the blocks or figures in place.68 The challenges to the structural integrity of a statue remain, even when it has reached its intended location. Even in the unlikely case that the sculpture would never be moved again, the low tensile strength of the material puts the figure at risk of breaking. The more slender and isolated the forms, the less able they are to withstand the tension created by their own weight. Figures that are heavily out of balance, such as for instance the Antikythera boy, bent over in a precarious position, would hardly stand safely without supports. Clearly, their removal would endanger the statue’s overall stability. Most struts are undoubtedly useful to guarantee the long-​term safety of a statue. The triangular wedge connecting the free leg of a figure to the base is fundamental to stability and balance.69 Struts that connect vertical elements (such as the two elements of a group, a statue and its attribute, the calves of a figure) provide a further axis of balance for the sculpture as a whole. In the case of an outstretched arm holding a heavy object, a strut strategically located under the hand is expedient to arrange and manage the weight. In general, features that project out from the core benefit from underpinning. The basic principles of cross-​bracing for the distribution of weights and forces were, of course, well known to Roman craftsmen and were widely employed in architecture as well. The method of joining offers no decisive advantages in this respect. A join that is not gravity assisted needs to be carefully reinforced. First, a sculptor 67 See Waywell (1978: 65) on the marbles from Halicarnassus. Ropes are unlikely to have been of standard manufacture in Antiquity, but rather were specialised products. For an introduction to the ancient technology of ropes see Waetzoldt (2007: 113); on their use to move sculpture in the early modern period see Wallace (2015: esp. 50). 68 Cassieri (2000: 50) and Andreae (1999: 199) mention traces of the pulley on the ceiling of the Sperlonga cave, above the location of the group of Polyphemus. For the lifting technologies and equipment in Roman building sites see Di Pasquale (2004: 125–​240), as well as Adam and Varène (1980: 223–​9). 69 See the comments by Landwehr (1982: 19–​21).

125

126

126

The Quest for Solidity

was required to shape the joining surfaces with care, making them interlock with one another. Then, he had to build a support into the join in the form of a pin, rod, or dowel of metal. Lastly, he must choose the right adhesive to secure the join.70 In one of her seminal articles on the ancient technique of making joins, Amanda Claridge speculated on whether monolithic statues with struts would survive intact any longer than their non-​monolithic counterparts composed of joined pieces. She believes that, although the odds may be in favour of joining and using metals, the state of the statuary that has survived and the number of ancient repairs indicate that the two techniques were more or less equivalent in terms of durability.71 In general, the choice probably depended on the availability of a suitable block. However, a sculptor unsure of his ability or uncertain about the quality of his block would perhaps opt for joining. Although struts increase a statue’s durability, their size may pose technical difficulties that surpass this advantage. Often the dimensions of struts exceed by far the static requirements for stability. This is the case of the Polyclitan Diadoumenus in the Metropolitan Museum of Art. As I have already pointed out, the statue originally included an extensive system of supports. A shorter strut joined the athlete’s calves. A long diagonal shaft was located on each side between the elbow and the hip. Giving a regular profile to these bars, so detached from the body, must have required very careful work. The hefty vertical struts in the upper section of the figure, however, were even more precarious. One of them links the left hand to the shoulder. Another strut connected the right shoulder to the thin ribbon that runs horizontally from the athlete’s head to his hand. Not only was the latter strut hardly necessary to sustain such a light element, but its carving must have posed a major threat to that whole section of the sculpture. Struts like the ones attached to this statue do not aid stability.72 In this particular case, joining the arms to the torso could have represented an easier solution in terms of both time and safety. An even more expedient method would be to carve the single ribbon separately, to be inserted in sockets at both ends, one on the bow and one in the hand. This measure actually seems to have been adopted for other versions of the athlete binding the 70 For ancient cements and adhesives see Claridge (1990: 136, 153–​4) and Burnett Grossman (2003: 4–​5). Unfortunately, as Claridge points out, there are few cases in which both elements of a join are preserved. We are ill-​informed about ancient cements and adhesives, the remains of which have rarely been tested. As a consequence, their strength is in most cases inferred without reliable scientific evidence. 71 See Claridge (1988: esp. 140; 1990: esp. 137). 72 In fact, they were ultimately ineffective in safeguarding the statue’s arms and the ribbon. See also Ridgway (1966: 35) and Geominy (1999b: 57 Appendix IV).

127

Stability

taenia around his head.73 A join in this position remains virtually invisible while guaranteeing maximum stability to the fragile detail and allowing its easy substitution in case of breakage. In reviewing this discussion of the technology of marble carving and the ancient trade in marble and statues, we can conclude that struts can and do serve a utilitarian purpose in marble sculpting. They protect the statue from the blows of the chisel. Once the statue is in place, they offer additional security to the composition. These concerns do not apply exclusively to the copies of bronze prototypes. However, struts that are particularly long, appended to a slender form, or placed so as to create a large undercut contradict these principles. The process of shaping long, faceted rods threatened the stability of the statue. Evidence points to a complex and heterogeneous picture rather than a definite set of rules. Sculptors seem to have negotiated case by case the advantages of struts, according to the quality of the marble block and the availability of alternative solutions such as joining or a different form of support (e.g. a shorter strut, attached to a less slender point of the statue). The challenges that large and carefully finished struts presented to their carvers require, in my opinion, an explanation that goes beyond practical considerations alone.

73 E.g. the late Hellenistic head in Rome, Museum of Ancient Sculpture G. Barracco (inv. 107) on which see H. von Steuben (in Helbig4, vol. II, 644 no. 1893), Zanker (1974: 12–​13 and Pl. 14.1, 3) and Kreikenbom (1990: 198 no. V.39 and Pls. 312b, 313a), and the one in Eichenzell (Fulda), Schloß Fasanerie, listed in Kreikenbom (1990: 200 no. V.47 and Pl. 328b). Similar considerations apply to the two version of the figure at the British Museum, the Farnese (inv. 1864,1021.4) and the Vaison Diadoumenus (inv. 1870,0712.1), on which see Smith (1892–​ 1904: vol. I, 266–​8 nos. 500–​1), Zanker (1974: Pls. 13.3, 14.4) and Kreikenbom (1990: 189 nos. V.5–​6 and Pls. 261–​4). Scholars have suggested that holes and sockets may indicate that a band of bronze was attached to the head, an explanation that seems plausible when the surface of the fillet is uneven and sunken, like in the Vaison Diadoumenus. A degree of uncertainty is raised by cases in which modern restorations included new ends for the ribbon, such as the head in Dresden (Albertinum, inv. 71), on which see Kreikenbom (1990: 200 no. V.46 and Pl. 327a) and C. Vorster (in Knoll, Vorster, and Woelk 2011: vol. II, 689–​93 no. 160).

127

128

6

Statuesque Statues

The versions of the Dancing Faun type from the so-​called ‘Invitation to the Dance’ group, especially the statue at Palazzo Corsini, indicate that supports could be used as a means to highlight the figure’s movement. In particular, idiosyncratic tree trunk supports sometimes assist figures rushing forward or pulling back in a pose of bewilderment and surprise.1 The case of the Tyrant Slayers, and in particular the statue of Aristogeiton, is illuminating. All of the replicas of the body include a tree stump that follows the line of the lower left leg and runs, more or less, parallel to the right leg extended behind (Fig. 36).2 Although carved in different epochs, they employ a similar strategy to integrate their auxiliary supports into the composition. The Tyrant Slayers group is based on an early classical bronze prototype displayed in the Athenian agora, which is mentioned by several literary sources and represented on a small set of Attic vases and reliefs.3 The fifth-​ and fourth-​century BC two-​dimensional depictions of the original and the later Roman copies agree on both the group’s general iconography and the characters’ attributes. Harmodios and Aristogeiton each brandish a sword. In addition, the latter carries a chlamys draped over his left arm. On vases and reliefs Aristogeiton’s cape is variously represented as either a short stole hanging loose from his arm or a rich drapery that E.g. the statue of Marsyas from the Esquiline at the Museo Gregoriano Profano, inv. 9974, on which see W. Fuchs (in Helbig4, vol. I, 764–​7 no. 1065) and Vorster (1993a: 21–​4 no. 3, 203–​5 Figs. 16–​20). Cf. the Hypnos at the Prado Museum in Madrid (inv. 89–​E), on which see Schröder (2004: 233–​8 no. 144). 2 On the Farnese Aristogeiton (Naples, National Archaeological Museum, inv. 6010) see C. Capaldi (in Gasparri 2009: 180–​4 nos. 84–​5) with earlier bibliography, as well as Brunnsåker (1955: 47–​53 no. A1). The Capitoline statue, found near the Church of S. Omobono (inv. 2404, Centrale Montemartini, no. II.58) is discussed at length by Germini (2008: 27–​42, 189–​91 no. 30). For the torso at the Boboli Gardens see Brunnsåker (1955: 53–​ 5 no. A2, Pls. 4–​5). 3 See S. Schmidt (2009), Brunnsåker (1955: Pls. 4–​5) and Azoulay (2014: 106 Fig. 9, 109–​10 Figs. 10–​12, 137–​8 Figs. 13a–​b). On the two groups created by Antenor (after the downfall of Hippias) and by Kritias and Nesiotes (in 477/​6 BC to replace the former) see Anguissola (2012: 68, 153–​4) with earlier literature. 1

128

129

Statuesque Statues

reaches down to his left thigh.4 Theoretically, the Roman sculptors could have arranged the chlamys with comparable freedom. Neither solution would affect the figure’s recognisability. A  longer cape, however, would require no supports. Instead, both the replicas that preserve some traces of the arrangement of the cape rely on a strut attached to the short cloth that visually extends the line of the lower left leg. They differ in that the statue now in Naples uses a thin branch departing from the trunk to brace the cloth, whereas the Capitoline copy shows the outline of a round strut on the upper thigh (Fig. 37). The statues of Aristogeiton indicate that, aside from their practical purposes, both figural and non-​figural supports were also used to supplement the figure and enhance its movement. In this chapter, I abandon the perspective of functional utility and discuss the role of struts in the composition as abstract complements to the body and framing devices. Struts count among several techniques –​including typology, style, labelling, the representation of a statue base, or architectural niche –​that the artist may exploit to establish the artificial nature of his figure. Peter Stewart has discussed at length how these visual tools belong to the vocabulary of the ‘statuesque’ or ‘the statue-​like quality of a statue’, referring to a shared conception of what a statue typically looks like.5 These techniques rely on intuitive strategies of differentiation and help to distinguish the depictions of artworks from the representation of living creatures. The use of well-​established typologies (e.g. ‘archaic’ features for divine images) or explicit labelling calls a statue’s mate­ rial nature to the viewers’ attention. Surrounding the statue with borders and cases, instead, creates a basic architectural framework that highlights an

4 A short chlamys hanging loose is depicted on a Panathenaic amphora at the British Museum (inv. 1866,0415.246, Vase B 605, 403/​2 BC), on which see CVA British Museum (III H.f., Pls. 2.3 and 6), ABV (411 no. 4 and Addenda2, 107), S. Schmidt (2009: 228 Fig. 10), Azoulay (2014: 106 Fig. 9b), as well as on an Attic red-​figure chous in Boston (Museum of Fine Arts, H.L. Pierce Fund, 98.936, ca. 400 BC), pictured in S. Schmidt (2009: 227 Fig. 9) and Azoulay (2014: 109 Fig. 10). Other examples show a long chlamys reaching down to the leg. These are an Attic red-​figure chous at Villa Giulia (inv. 44355, ca. 400 BC), on which S. Schmidt (2009: 233 Fig. 16) and Azoulay (2014: 110 Fig. 11), an Attic red-​figure chous from Spina (Ferrara, Museo Nazionale di Spina, inv. 6406, ca. 400–​390 BC), on which S. Schmidt (2009: 233 Fig. 17) and Azoulay (2014: 110 Fig. 12) and the so-​called Elgin Throne (Malibu, J. Paul Getty Museum, inv. 74.AA.12, ca. 300 BC). For the Elgin Throne see Lapatin and Wight (2010: 22–​3) and Azoulay (2014: 137–​8 Figs. 13a–​b). See also the storming figure on a red-​figure lekythos at Athens (National Archaeological Museum, inv. 1697/​CC 1430) in ARV2 (805 no. 82). 5 On the concept of ‘statuesque’ and the devices that highlight this quality see Stewart (2003: 92–​111).

129

130

Figure 36  Statue of Aristogeiton, Hadrianic–​Antonine period. Marble. H. 183 cm. Naples, National Archaeological Museum, inv. 6010

131

Figure 37  Statue of Aristogeiton from the area of S. Omobono, first century BC. Marble. H. 180.5 cm without plinth. Rome, Musei Capitolini, inv. 2404 /​Centrale Montemartini, no. II.58

132

132

Statuesque Statues

area of special visual emphasis.6 As Georg Simmel argued in his now-​classic exploration of the subject, a frame ‘excludes all that surrounds it, and thus also the viewer as well, from the work of art, and thereby helps to place it at that distance from which alone it is aesthetically enjoyable’. The frame articulates two fundamental qualities of a work of art, ‘distance and unity’ –​the separation from the sphere of immediate life and self-​sufficiency.7 I argue that, especially in the case of sculptural copies, struts contribute to shift the reference from the statue’s living counterpart to the art-​object in itself, the series to which it belongs, and the prototype. In this sense, struts offer a means of visual and material distinction, as they both frame the figure within a system of linear supports and highlight its material nature. By reifying artificiality, struts emphasise the artfulness of the work of art.

Abstract Complements Scholars have already observed that the form and treatment of non-​figural supports are occasionally adapted to the individual composition. Mary Hollinshead perceptively described the ‘smooth curving hornlike strut’ extending from the hip to the hand of the young Eros, Apollo, or Thanatos from the Gardens of Maecenas as an element that ‘complements the quiet, static tone of the statue’.8 These remarks are crucial in the case of decorated struts. Struts with painted or carved motifs raise a fundamental question about their visibility and the viewer’s reception. On the one hand, sculptors may have used decorative patterns to adorn an unwelcome feature, which they could not conceal otherwise. On the other hand, this may just as well be a deliberate strategy to draw attention to the strut. Or again, technically useful struts could be (also) valued as an additional surface that offered the space and place for abstract motifs that met the period’s decorative taste. The corpus of cylindrical or slightly conical struts 6 The role of architectural motifs, such as borders and frames, to differentiate areas of special emphasis in a composition is explored with reference to Roman strigillated sarcophagi by J. Huskinson (2015: 94–​7). On the semantic potential of statue bases see also Ma (2013: 38–​43, 255). Discussions of the rhetoric and semantics of pedestals and frames from the Renaissance to modern art are numerous. For pedestals see Gerstein (2007); for an introduction to the question of frames and boundaries see Schapiro (1969) as well as the collections edited by Duro (1996) and Körner and Mösender (2008). The methods and meanings of framing devices in the Greek and Roman world are explored in an important volume edited by Verity Platt and Michael Squire, The Frame in Classical Art. A Cultural History, forthcoming when this book was being prepared for the press and therefore only marginally acknowledged here. 7 Simmel (1902), cited here in the translation published in 1994 by M. Ritter in Theory, Culture and Society 11: 11–​12. 8 See Hollinshead (2002a: 130, 132, as well as 144 for a strut as an ‘abstract complement’).

133

Abstract Complements

encircled by a spiral groove is indicative of an attention to the individual work of art, which resists simple categorisation. In a number of cases, the spiral motif matches and continues the texture of the adjoining surface.9 This strategy was adopted by the second-​century AD sculptor of a small figure of Heracles at the Boston Museum of Fine Arts, which is only 57 cm tall with its plinth.10 The statuette, found in Rome between the Aventine and the Tiber, is considered to be the copy of a Greek bronze from about 450 BC. Heracles rests with his right hand on the club, joined to the thigh above the knee by means of a short strut, finely shaped in a regular groove pattern. With its exquisitely carved surface, the strut makes a transition between the rugged surface of the demigod’s wooden club and his polished skin. The decorative pattern serves, at once, to assimilate and differentiate. A comparable design that distinguishes the strut from the adjoining flesh was chosen for the struts on two very fine mid-​second-​century AD versions of the wine-​pouring satyr, both from Rome and carved about one century later than their counterpart in Palermo.11 Both include a tapering spiral strut to support the outstretched lower left arm, with a carefully carved surface that marks a striking contrast to the satyr’s smooth skin. The extended 9 The statue of the snake-​god Glycon from Tomi (Costanza, National Museum for History and Archaeology, inv. 2003) includes a thin spiral strut supporting the head, with a pattern similar to that of the coils. See G. Bordenache (1964: 157–​63 and Figs. 1, 6a, 7a; in LIMC, vol. IV, 280 no. 1). The rope-​like strut of a fourth-​century AD seated statuette perhaps from Civita Lavinia in Latium, on which see Y. Marano (in Sena Chiesa 2012: 226 no. 106) and S. A. Dayan (in Giuliano 1979: 325–​7 no. 192), was given a surface akin to the drapery. Also in the statue of a sacrificing woman at the Uffizi (inv. 1914 no. 131) the choice of a rod-​like connector seems intended to soften the passage between two wavy adjoining surfaces, the drapery and the flame. The horizontal, slightly curved, and faceted strut (ca. 32 cm long) from the woman’s thigh to the altar supports a further conical strut with a groove pattern, that extends diagonally to the right wrist (ca. 31 cm long) –​a singular case of struts propping one another. Due to the visibility of these ‘sgraziati puntelli’, G. A. Mansuelli (1958–​61: vol. II, 107–​8 no. 134) described this statue as a ‘replica sciatta’ of a Hellenistic body type. See also Hollinshead (2002a: 132 and note 47) and Bieber (1977: Pl. 149 Fig. 866). 10 Boston, Museum of Fine Arts, inv. 14.733. See Caskey (1925: 133–​5 no. 64) and Comstock and Vermeule (1976: 89–​90 no. 139). Hollinshead (2002a: 120, 129) also comments on the spiral strut. 11 Paris, Louvre, inv. 2333. See Gercke (1968: 9 no. T.5) and J.-​L. Martinez (in Pasquier and Martinez 2007: 274–​6 no. 67). Rome, Palazzo Altemps, inv. 8597. See Gercke (1968: 5–​6 no. St.6), A. A. Amadio (in Giuliano 1992: 194–​9 no. 24), De Angelis D’Ossat (2002: 257; 2011: 179). See also the statue of the type called Apollo of Mantua from the collection of Cardinal Mazarin, now at the Louvre (inv. MA 689; MNB 257). With his left hand, the god held a lyre against his body. Today, all that remains of the instrument is one of the curved raised arms, embellished by a delicate pattern of parallel ridges. The sculptor gave the same treatment to a short strut between the right wrist and hip. In this way, he matched the surfaces of the only two inanimate elements of the composition, while setting them apart from the soft modelling of the nude body. Muthmann (1951: 169 n. 102) dates the statue to the Hadrianic or early Antonine period, Zanker (1974: 61 no. 2) to ca. AD 50, E. Simon (in LIMC, vol. II, 374 no. 39a) to the age of Claudius. See also Papini (2008: 16, 18, and 39 n. 12) against dating to the Claudian or Neronian period.

133

134

134

Statuesque Statues

left arm of the satyr from the Borghese collection, now at the Louvre, was supported by a finely worked spiral groove (Fig. 38). A twisted strut must have once completed also the Ludovisi replica, whose modern restorers misunderstood the spiral stub sprouting from the youth’s left hip and transformed it into an unusual drinking vessel –​a long, tapering rhyton. As I will discuss in the last chapter, the exploitation of the effects of contrast in texture and light reaches its peak with late antique small-​scale statuary, in response to a growing taste for intricate patterns and abstraction. The best-​known example is probably a late antique miniature Venus removing her sandal, found together with other pieces in a seaside villa just east of ancient Alexandria.12 With its soft and abstract modelling, the sculpture bears the stylis­tic traits of the period at the turn of the fourth and fifth centuries AD. A thick twisted strut, with deep ridges, links the bodies of the goddess and Cupid. The highly polished surface of the strut alternates between light and shade with each twist of the spiral, thus increasing the impression of precious artificiality.

Framing Devices As Peter Stewart has compellingly argued, the depiction of bases is the most effective device for an artist to characterise, in various media, the representation of a statue as opposed to living figures. Plinths are decorative devices that provide the figure with an appropriate framework. Together with other and more elaborate motifs, such as side columns or architectural niches, the depiction of bases clearly alludes to the setting of the statue. Beginning in the second century AD, sculptors increasingly used profiled plinths, carved from the same block of marble as the statue, for figures in the round.13 Sculptors also reached comparable effects by aligning large marble shafts to the limbs or attributes of a statue. While highlighting the pose, struts enclose the body in a network of lines. Like plinths and other framing devices, struts are immediately recognisable as extrinsic elements, in the sense that they are unrelated to the narrative. The potential of struts was fully exploited by the carver of the wounded warrior now at the Metropolitan Museum of Art, thought to represent the Greek hero Protesilaos (Fig. 39).14 The warrior stands with his feet carefully 12 Alexandria, Greek and Roman Museum, inv. 29456. For this little group of exquisite workmanship see Hannestad (1994: 123–​6 and Figs. 80–​2) and Stirling (2005: 108). 13 The standard treatments of profiled plinths are Lippold (1923: 100–​2) and Muthmann (1951: 120–​8). See also the excellent summary with a discussion of selected case studies in Filges (1999: 397–​404), as well as the comments by Schade (1998: 192–​3; 2007: 194). For moulded bases in late antique sculpture see Stirling (2005: 106–​7). 14 New York, Metropolitan Museum of Art, acc. no. 25.116, Hewitt Fund 1925. This sculptural type is discussed by Langlotz (1971) and Kauth (2010: esp. 31–​6 for the question of dating).

135

Figure 38  Statue of Pouring Satyr from Rome, second century AD. Marble. H. 144 with the plinth. Paris, Louvre Museum, inv. 2333

136

Figure 39  Statue of the so-called Protesilaos, possibly combining a head from the Antonine period and an earlier body. Marble. H. 220 cm. New York, Metropolitan Museum of Art, acc. no. 25.116, Hewitt Fund 1925

137

Framing Devices

Figure 40a–​b  Drawn reconstruction of the so-​called Protesilaos. Front and rear views. The reconstructed struts are represented with a dotted line

placed on a sloping surface, perhaps in the act of descending from the ship. Two marble bars support the figure’s alighting pose. A shorter strut linked the right shoulder to the wrist. A hefty bar, longer than the arm itself, runs parallel to the first bar from the right hip to the elbow. Together, the two struts emphasise the figure’s expansive movement, with a spear in the raised right hand (Fig. 40a–​b). The lower strut continues the line of the tree trunk support placed behind and to the left of the figure. Seen from behind, the figural and non-​figural supports accentuate Protesilaos’ step forward. In the front view, instead, the double strut above and beneath the right arm creates a mirror-​like distribution of masses between the lower left and upper right section of the statue. A  solid mass of marble augments the figure on the lower left side. The bars on the upper right side frame a comparable plane surface, including ample void areas. I will argue at a later point in this chapter that the Palermo version of the Pouring Satyr, equipped with a long and conspicuous strut, employs a similar composition strategy to distribute the projecting masses on opposite sides.

137

138

138

Statuesque Statues

The struts attached to the Castel Porziano discus thrower and other replicas of the same type were a conspicuous element of the composition, with a considerable impact on both the front and rear view. The left lower leg, both arms, and the shoulders create a half-​round shape encircling the figure. The strut and the right thigh form a straight line. The curved and linear trajecto­ ries culminate and intersect at the crucial point of the composition, the raised hand with discus. The arched strut stretching from the waist to the underside of the raised right elbow of the so-​called Dresden Artemis has a similar effect.15 This sculptural series descends from a prototype from 360–​350 BC. Not all of the approximately thirty versions of this body type include a raised right arm in the act of pulling an arrow out of the quiver. Those that do, however, do not include any struts supporting the arm, with the exception of the example in Dresden. In his catalogue entry, Wilfred Geominy hinted at possible expressive reasons for this device. In his opinion, the sculptor may have used the strut as a means to stress the copy’s faithfulness to its prototype.16 I will address in the next section the ways in which struts visualise exactitude in the transmission of forms. Independently from this statue’s relationship to a classical prototype, however, the shape of the strut is specifically adapted to the individual context. In the front view, the strut runs parallel to the strap across the goddess’ chest. Viewed from the rear, it loosely echoes the line of her quiver (Fig. 41a–​b). Together, the strut and the right forearm, bent in the act of drawing an arrow out of the quiver, form a shape that recalls that of a bow. Unfortunately, we can only guess how this shape would mirror the attribute on the goddess’ left side, probably the bow itself. The taste for decorative frames in the form of virtuoso strut-​work reaches its apex in the corpus of miniature statuary produced in late antiquity. A statuette produced in a later period, the late antique Diana from the villa at Saint-​Georges-​de-​Montagne outside Bordeaux, is famed by a uniquely complicated arrangement of struts.17 The goddess of the hunt stands with a hind at her feet in front of a tree. The arched struts connecting her right thigh, her windswept drapery, and her raised right arm echo the curve of her bow and continue the line of the branches above as segments of a circle. The circular pattern encloses Diana’s upper body as if in a medallion.18 15 Dresden, Albertinum, inv. Hm 117. See W. Geominy (in Knoll, Vorster, and Woelk 2011: vol. I, 183–​9 no. 15). On the replica series see also Schröder (2004: 379–​82 no. 182). 16 W. Geominy (in Knoll, Vorster, and Woelk 2011: vol. I, 183). 17 Bordeaux, Musée d’Aquitaine, inv. 71.16.1. For comments on the struts see Hollinshead (2002a: 134–​5). The statuette of Diana and the other sculptural finds from St Georges-​de-​ Montagne are discussed by Stirling (1996; 2005: 30–​7). Excellent side and rear views are provided by Stirling (2005: 32–​3 Figs. 5–​6) and Bergmann (1999: Pl. 27.1–​2). 18 Cf. the effect of the flying drapery of the so-​called Selene from Silahtarağa, on which see Bergmann (1999: Pl. 22).

139

Supporting Copies

At the same time, the lacy network of struts emphasises the activity of the figure’s limbs.19 The goddess’ expansive pose is, at once, emphasised and restrained by this network of struts. Painstakingly carved in the round, the struts also speak to the statuette’s superb workmanship. Paraphrasing Jacques Derrida’s comments on Kant’s characterisation of frames, we can observe that in all these cases struts stand out both from the statue and from the context, for ‘the parergonal frame stands out against two grounds but with respect to each of these two grounds, it merges with the other’.20 Struts frame the image by following and highlighting its outline rather than enclosing its field.21 The structural supports are, at once, present and absent –​distinct from the result of mimesis that is the composition yet complementary to its structure and dynamics.

Supporting Copies Scholars in general consider struts as evidence that a given statue is the (Roman) marble copy of a (Greek) bronze original. In particular, variation in the placement of struts indicates that the prototype had no struts at all because it was made of a lighter material such as bronze. ‘Each copyist’, summarised Brunilde S. Ridgway, ‘put the strut where he felt that the statue needed it most and gave it the shape fancied by his imagination since there was no prototype requiring faithful copying’.22 In the same way, the fact that two or more versions of the same composition have struts in the same places signals a shared prototype in marble, which itself included struts. Certainly, replica series based on ancient Greek models form a recognisable 19 The strut between the left thigh and wrist of the second-​century AD statue of the type ‘Versailles Diana’ from Perge (Antalya, Museum, inv. A-​3731), on which see Özgüt (1996: no. 8), similarly highlights the figure’s expansive pose. Another late example, a statuette of Jonah praying in the Cleveland Museum of Art (John L. Severance Fund 1965.240), makes an equally adroit use of the struts to dramatise the figure’s movement. The prophet stands with raised hands, connected to the chest by two tapering struts. Although in themselves small, these struts are very large compared to the size of the miniature statue. A further strut links the thumb and index finger of the right hand. The left hand, today damaged, likely included a similar connector. The position and angle of these small bridges highlight the spread of Jonah’s hands while invoking the Lord. See Wixom (1967: 81; 1979: 409–​10), Kitzinger (1978: 669–​ 73), and, for the struts, Hollinshead (2002a: 135). 20 See Derrida (1987: 61). For Derrida’s discussion of Immanuel Kant’s Critique of Judgement see my Introduction. 21 M. Schapiro (1969: 228) discusses cases where ‘the frame accents the forms of the signs rather than encloses a field on which the signs are set’, conveying the impression that ‘the image comes first and the frame is traced around it’. 22 See Ridgway (1966: 35).

139

140

Figure 41a–​b  So-​called Dresden Artemis, mid-​second century AD. Marble. H. 152 cm with the plinth. Front and rear views. Dresden, Albertinum, inv. Hm 117

141

Figure 41a–​b  (Cont.)

142

142

Statuesque Statues

sample in the corpus of Roman statuary.23 In most cases, it seems that prototypes in bronze were copied in marble and conspicuous struts frequently occur in this category of works. Nonetheless, these premises are not sufficient to consider struts as synonymous with copying; not all statues with external supports depend on a lost Greek prototype. Scholars have provided contrasting explanations for why the Romans apparently preferred to copy bronze works. One obvious reason is that bronze was the favourite medium of the most celebrated masters of the classical period. Some have argued that technical aspects of the sculptural process favoured copying bronze originals instead of marbles; the process of casting the plaster model necessary to produce replicas would damage the polychromy of a marble statue.24 We may argue, however, that this holds equally true in the case of bronzes, since contact casting also harms the delicate patina of the metal. Related to this, one may ask why so many Roman statues that seemingly replicate ancient Greek antecedents are made of marble. In a way, this question is inappropriate because it does not take into account the fact that most ancient bronzes (both Greek and Roman) have been lost to us due to the perishable nature of the material and the practice of recasting. One possible answer lies in the relative costs of materials, which would have been lower in the case of marble. However, the amount of labour necessary to produce a marble statue is significantly greater than that required for bronze, making this estimate unlikely.25 Another possible explanation is, again, that of polychromy, since marble offers a broader range of opportunities in matters of colour and individual finish.26 Others believe that the change of medium is intended to identify a statue as a copy. Implicit in this approach is the idea of a strict semantics of materials, associated to certain categories of works to the point of advertising their relative status –​the idea, in Wilfred Geominy’s eloquent words, that to ancient eyes ‘a work of art required its suitable material’.27 23 For an introduction to the large body of scholarship on the concepts of ‘originality’ and ‘copy’ in Roman art see Anguissola (2012: esp. 44–​57; 2015). 24 So Landwehr (1985: 183); see also Geominy (1999b: 46–8). 25 Andreae (1992: 21) notes that the costs of specialised workmanship and transport must have increased the price of a marble statue substantially. According to the calculations attempted by Duncan-​Jones (1982: 126; also 1962: 112 n. 160; 1965: 197), workmanship accounted for ca. 42–​8 per cent of the overall cost of a bronze statue and ca. 90 per cent of a marble work. On the price of statues in the Roman world see also Bang (1921) and Coarelli (1983). 26 See Andreae (1992: 21). 27 This view is compellingly argued by Geominy (1999b: 48–​9). For considerations on the relative merits of bronze and marble and the latter’s use to bestow the quality of an agalma on a statue, see Tuchelt (1979: 68–​90 and esp. 71–​4).

143

Supporting Copies

In general, a series of replicas includes figural and non-​figural supports of a different shape and size, as well as in a different position. The degree of variation is particularly evident in cases where the pieces are displayed together. For instance, the two replicas of the Three Graces found in the Baths of Cyrene that I have already mentioned were carved in the same period and exhibited together, flanking the entrance to the frigidarium. Nonetheless, they are made of a different type of stone, have a different size, and a different system of struts. Both technical and stylistic arguments indicate that the pieces belong to two different workshops. Other examples, however, demonstrate that these remarks have no general validity. Sometimes, statues that are otherwise stylistically different adopt similar strategies regarding the type and placement of supports. The two copies of the so-​called Dresden Youth type found at the Albanum Domitiani at Castel Gandolfo were probably carved in the same period and displayed together, each with a comparable system of struts.28 Although both date to the first century AD, they present significant stylistic differences, one of the two showing harder modelling, with sharper lines for the ribcage and the groin. The discrepancies in treatment have prompted the suggestion that the pieces must be the work of different carvers.29 These artists apparently shared the same approach to supports, appending to their figures a tall tree stump behind the right leg, a strut to fasten the lowered arm to the right side, and a bulkier strut attached to the left thigh. Although a number of torsos of the Dresden Youth type exist, mostly produced in the course of the second century AD, the use of struts of a size comparable to those of the Castel Gandolfo statues is extremely rare.30 The pair was found together with another statue derived from a prototype that scholars ascribe to the school of Polyclitus, a version of the Westmacott Ephebe. This piece, dating to the late first century BC, likewise shows traces of a particularly 28 Castel Gandolfo, Villa Barberini, Antiquarium, inv. 36407 and 36408. See Liverani (1989: 63–​4 nos. 24–​5) and Berger (1992: 306–​9 Figs. 473–​6). On the find location and associated pieces see Neudecker (1988: 142–​3), Liverani (1989: 63; 1993: 123–​5) and Newby (2005: 99–​105). 29 So Liverani (1989: 64; 1993: 123). See also Newby (2005: 101, 104). 30 The eponymous statue in Dresden (Albertinum, inv. Hm 88), dating to the first quarter of the first century AD in the age of Augustus or Tiberius, is one of the rare replicas that may have included similar struts. On the right thigh there is a trace of a strut that probably connected to the arm or wrist, while a larger area on the left thigh attests to an exceptionally voluminous strut that supported the extended right arm. See C. Vorster (in Knoll, Vorster, and Woelk 2011: vol. II, 704 and Figs. 164.1–​2). Traces of struts on both sides are visible in an early imperial replica in Basel (Antikenmuseum, Sammlung Ludwig, inv. S630) for which see A. Linfert (in Beck, Bol, and Bückling 1990: 595–​6 no. 117) and Berger (1990: 139–​54 no. 229; 1992: 312–​13 Figs. 482–​3), as well as in a Flavian torso at the Museo Nazionale Romano (inv. 52400), discussed by D. Candilio (in Giuliano 1995: 31–​2 no. S43).

143

144

144

Statuesque Statues

large strut on the boy’s head, which is uncommon within its replica series.31 Perhaps the three statues, although produced by different workshops and in different periods, were chosen for display together due to their congruence in size and subject, as well as, perhaps, for their technical similarities. These examples indicate that context played a significant role in the choice of the stylistic features of statuary, and that struts counted as an important stylistic element. Another replica series after a bronze prototype made by Polyclitus, the Diadoumenus, corroborates these considerations. The almost thirty extant versions of the full figure or the torso speak to the great variability in the use of both figural and non-​figural supports. Sometimes, the tree trunk support has a rather simple shape, with a plain decoration of parallel ridges.32 In several other statues, the carvers rendered the characteristic bark and branches of a palm tree.33 The earliest replica of the Diadoumenus, found in Delos, is equipped with a distinctive support that resembles an olive tree. A cloth and a quiver hang from the stump, where they were placed by the athlete in order to release his arms and fully display his body. The use of struts also varies considerably. Bridges often connect the calves to each other or fasten the figure to the tree trunk support. In the upper body, the range of choices increases. In one instance, a torso from Perge, the coarse mass of a neck strut can still be observed on its back.34 Elsewhere, the sculptor relied on the careful arrangement of the limbs alone to secure his work’s stability.35 None of the replicas, however, comes even close to the statue now in New York in terms of the number and size of the struts attached to the athlete’s arms. Both the strategies adopted by the carver of this statue –​connectors from 31 Castel Gandolfo, Villa Barberini, Antiquarium, inv. 36420. See Liverani (1988–​9: 105, 107; 1989: 55–​9 no. 21 and Fig. 21.5; 1993: 120–​2 and Figs. 31–​3, 36). On the series see Linfert (1993: 163–​5) and Rausa (1994: 193–​6). The statue in Cambridge (Fitzwilliam Museum, inv. GR3.1948), dated to the early imperial period by Budde and Nicholls (1964: 22 no. 42 and Pls. 12–​13), has a strut in the same position. See also Berger (1992: 323 Figs. 504–​5) and Linfert (1993: 180 Fig. 35, 184 Fig. 48, 188 Fig. 61). A mid-​first century AD head in Dresden (Albertinum, inv. Hm 85) shows a rough patch on the upper part, which may indicate the point of contact with the raised hand. See C. Vorster (in Knoll, Vorster, and Woelk 2011: vol. II, 701–​3 no. 163). 32 E.g. the torsos at Kassel (Antikensammlung, 7, inv. Sk 11) and Geneva (private collection, previously in London, Lansdowne House). See Kreikenbom (1990: 197 nos. V.34–​5 and Pls. 301–​5). 33 E.g. statues in New York (Metropolitan Museum of Art, inv. 25.78.56), Tripoli (Museum, inv. 22), London (British Museum, inv. 501), Rome (Musei Capitolini, Palazzo Nuovo, inv. 1878). See Kreikenbom (1990: Pls. 250, 258, 263, 266). 34 See Kreikenbom (1990: 192 no. V.17) for this statue now at Antalya, Museum, inv. A 3064. 35 Like the statues at Tripoli (Museum, inv. 22) and at the British Museum (inv. 500 and 501). See Kreikenbom (1990: Pls. 258, 262–​4).

145

Supporting Copies

hips to arms and struts linking the wrist or the fillet to the shoulder –​are exceptional. The former is only observed in a torso in Basel, dating to the age of Claudius, which bears traces of struts on both hips that possibly reached the underside of the athlete’s arms.36 The latter type of struts occurs in a second-​century AD copy today at the Prado Museum in Madrid and, perhaps, in a heavily restored piece at Villa Albani.37 Both the replica series of the Dresden Youth and the Diadoumenus derive from prototypes in bronze created within the same artistic milieu around the 420s BC. The shape and placement of struts likely reflect the static requirements of the individual piece, which are related to the quality and features of the marble block. These remarks, however, cannot explain why a certain number of pieces include such elaborate strut-​work, unnecessary to the figure’s stability. This question is all the more relevant since, for the most part, sculptures with struts that exceed the static requirements of marble are indeed replicas of bronze statues and, more precisely, replicas of naked athletes carved in the mid to late first century AD. It is therefore worth considering in further detail Wilfred Geominy’s suggestion that the primary purpose of particularly large and functionally redundant struts is to highlight the status of a work as a copy. In a slightly different articulation of the same argument, Elizabeth Bartman maintains that struts may ‘have been visually synonymous with the copy’ and functioned as ‘visual cues’ ensuring that a copy ‘could be appreciated as such even by those viewers who were not connoisseurs able to recognize its pedigree’.38 Geominy’s more radical position is that sculptors occasionally ampli­fied the auxiliary devices required by marble in order to emphasise the change of medium and the nature of their work. Excessive struts would then advertise the commitment to reproduce the pose of an ancient Greek model faithfully, notwithstanding the technical challenges posed by the heavier material. The different medium and the struts, together, turn copies into a distinctive category of artwork. Clearly, this line of thought and its implications challenge the current tendency to seeing ‘true copies’ –​intended, manufactured, and displayed to establish a link to a given Greek masterpiece –​as a marginal phenomenon of Roman art. According to the prevailing view, this class of items was 36 Antikenmuseum inv. BS 225. See Kreikenbom (1990: 191 no. V.10 and Pls. 268b–​70). 37 In the case of the Madrid statue (inv. 88-​E), modern restorations prevent any remarks about the arrangement of the right arm. See Kreikenbom (1990: 188 no. V.3 and Pls. 253–​4) and Schröder (2004: 67–​73 no. 104). For the Diadoumenus now at the Villa Albani see Kreikenbom (1990: 190 no. V.9 and Pl. 268a). 38 See Geominy (1999b: 49) and Bartman (1992: 39). A brief discussion of this point is in Anguissola (2013: 5, 16–​17).

145

146

146

Statuesque Statues

produced and traded principally to become a conventional part of larger sculptural settings in public and private buildings. As such, ‘true copies’ of ancient Greek masterpieces must have enjoyed far less recognition than ‘free emulations’ based on the styles of the ancient Greek masters. Certainly, the idea of ‘Roman copies after Greek originals’, as posed by nineteenth-​ and early twentieth-​century art historians, revolves around a simplistic view of the relationship between Greek and Roman cultures and fails to account for the retrospection that permeated Roman art at every level. The Romans’ appropriation of Greek art operated within a living tradition and an evolving visual code. Nonetheless, if the capacity to vary on a given repertoire was a much-​esteemed quality for rhetoricians, poets, and artists alike, reproducing an illustrious prototype exactly was also valued by informed viewers. As I have argued elsewhere, within Roman visual culture both ‘true copies’ and creative imitations held a central role in collective imagery, everyday practice, and art criticism.39 By accepting that struts could also advertise the status of being a copy, we can make sense of a fundamental dichotomy about this material. On the one hand, large struts emphasise a figure’s movement and, in consequence, the sculptor’s adherence to the model, which was not adapted to the heavier and less ductile medium. On the other hand, the figure’s movement appears frozen, surrounded by intricate strut-​work. In this sense, oversized struts may proclaim the artificial nature of an art-​object, which ceases to refer to its living counterpart and begins to refer to itself, the replica of a man-​made image. This discussion of struts can be associated with the explanations provided by some scholars for sculptors’ decisions to retain in a finished work the so-​called ‘points’, the marks made on a block of stone in order to transfer the measures of a model or bozzetto.40 These marks appear as small dome-​shaped protruberances with a depression in the centre for the point of the compass. They were occasionally left on pieces painstakingly carved to the tiniest detail and in most cases appear on the figure’s head. Points of this type are preserved in some of the high-​quality versions of classical types that I  have already introduced here. A striking example is the eponymous Dresden Youth. The statue, which still shows the patches left by the modern restorers who carved away the bulky struts on both sides of the boy, is the work of an exceptionally accomplished 39 See Anguissola (2012: 54, 71–​2, 74–​5, 125, 173; 2014b: esp. 138–​46), as well as the comments by Stewart (2003: esp. 233–​5), Hallett (2005: esp. 435), and Kousser (2008: esp. 150). 40 See Fittschen (1977: 81), Geominy (1999b: 45), and Zanker (2002: 100).

147

The Individual Copy

sculptor. On top of the head, among the locks treated in calligraphic detail, the artist left two round measuring points as a testament to his diligence.41 The Lancellotti discus thrower, which according to historical sources still preserved, at the time of its discovery, the stub of a long strut on the right side, includes two prominent round points in the hair, just above the forehead.42 The sculptor of the exquisite Ludovisi replica of a Pouring Satyr, originally equipped with a delicate spiral strut, chose the same location for two small points placed on symmetrical locks above the fillet.43 Points and struts share the nature of technical devices. Both visualise fundamental steps and procedures of stone carving, reminding the viewer of the challenges inherent to the material. They differ in that measuring points are virtually unnecessary once the statue is finished and their removal presents no challenges. Struts, instead, are in most cases important for the stability of a statue. Additionally, both giving a strut a faceted shape and cutting it away puts the adjoining limbs at risk. Furthermore, if ­measuring points can be detected only at a close distance, struts are almost always visible and occasionally even larger than other parts of the figure. The fundamental question that both devices raise, if we consider them to be a testament to faithful copying, is that of the actual exactitude. In general, the pieces with unusually large struts or measuring marks left in place are not closer to their lost prototype than the other replicas –​as far as an exercise in Kopienkritik can tell. If this explanation holds true, the purpose might rather be to create an impression of faithfulness rather than to document the actual degree of faithfulness.44

The Individual Copy The method for making copies entailed, first, the creation of a plaster model of the original and, secondly, the transfer of points marked on 41 See Pfanner (1989: 240 no. Dresden.1) and Knoll, Vorster, and Woelk (2011: vol. II, 709 Fig. 164.8). 42 See Pfanner (1989: 248 no. 12). 43 Two more points are on the back of the head. According to M. Pfanner (1989: 249 no. 16) the varying distance between the marks on the front and back of the head attests to the use of measuring devices of different sizes. 44 Referring to the Scylla group at Sperlonga, W. Geominy (1999b: 50–​1) suggests that large struts can occasionally be used to give the impression that a certain artwork derives from a celebrated masterpiece, regardless of whether this is actually the case. A similar argument applies to miniatures with bulky and non-​essential struts, which may hint at some association with large-​scale statuary. See Anguissola (2013: 17) and Hollinshead (2002a: 136 n. 56).

147

148

148

Statuesque Statues

it to the block of stone used for the statue.45 Due to the lack of written accounts, uncertainties remain about how the measuring points were taken. In modern times, sculptors relied on instruments, such as the so-​ called pointing machine, which guarantee a high level of accuracy. There is no evidence, however, that the same technology was available during the Greek and Roman periods.46 Obviously, the more points that are taken, the more accurate a copy can be.47 Since, however, most points were eventually carved away, it is ultimately impossible to determine what rules Roman artists followed with regards to the number and placement of their reference marks. Besides, at least the secondary points could simply be marked with charcoal on the marble surface, leaving no traces of the process. Moreover, the fact that our evidence for plaster models is essentially limited to the Baiae finds prevents any insights about the average degree of completeness, faithfulness, and reliability of the guidelines available to the Roman copyists. If we return to the question of the struts in light of the ancient technologies for stone sculpture, two aspects become apparent. In the first place, supports are the only elements that a sculptor cannot derive from the plaster model, unless the prototype is also a statue with supports. This would not have presented any difficulties in making short connectors between two vertical masses of any type. In order to accommodate large struts that intersected various planes of the marble block, however, the grid that was prepared for transferring the measures to the stone would have to be adapted accordingly. The nature of the medium increases the challenge. A  block of marble has a major plane along which it can be 45 The process of making casts is described by Landwehr (1982: 16–​18; 1985: 12–​25), while Pfanner (1989: esp. 187–​92) addresses the technique of taking and transferring measuring points. See also Richter (1955: 105–​11; 1962), Strong and Claridge (1976: 203), and Rockwell (1993: 118–​22). 46 In its basic form, the pointing machine consists of an inverted ‘T’ equipped with moveable sockets, into which adjustable arms are fitted. See Richter (1955: 106; 1962: 53). Weitmann (2012) presents the fragments of a bronze tool found in the Terrace House 2 of Ephesus interpreted as an early form of pantograph, possibly used to create sculptural copies. I. Kovalleck (in Krinzinger 2010: 324, 328–​9 no. A–​B 35), who dates the tool to the third quarter of the third century AD, thinks instead that it is a type of surveying instrument akin to a groma. 47 In his 1802 handbook for practitioners, Francesco Carradori describes the challenge of accuracy in taking the points: ‘At all the measurements taken, make a small depression with a rounded chisel and a very small hole that is marked with a pencil. This should always remain visible until the work is almost finished. Although this is a simple operation, it nonetheless requires great care not to make any errors in measurement, the consequence of which could be of irremediable damage’ (tr. M. Kalevi Auvinen, 2002: 37).

149

The Individual Copy

easily cut, which the sculptor exploits for the main axis of his composition. The secondary planes, used for all projections outside the core, intersect according to largely unpredictable patterns.48 The difficulty in predicting the static behaviour of a block may be the cause of the slight variation in the position and orientation of the massive struts on the left side of the Castel Gandolfo statues of the Dresden Youth type, which scholars have interpreted as evidence corroborating the presence of two different carvers or an indicator of chronology.49 In light of these remarks, it becomes even more doubtful whether an unaccomplished and anxious sculptor, uncertain of his medium and of his own skill, would undertake the hazard of carving unguided, daring undercuts. A proficient craftsman, instead, could find in the interplay of structural devices a strategy to hint at his own ability in retaining the forms of the prototype while in fact producing creative modifications that affect the overall composition of the statue as well as its stability. Even if we agree that struts sometimes functioned as a conspicuous testament to consummate stone carving, we are ultimately unable to determine whether all viewers understood the implications of this strategy and what their responses might have been. In order to investigate in closer detail at least the intended effect of a support, we should consider the work not only as a whole, but also the work in comparison to similar pieces. The series of the Pouring Satyr, which according to scholars reflects a bronze original created by Praxiteles, provides a perfect case in point. Archaeologists have identified some fifteen statues and torsos, mostly found in Rome or its environs, which belong to the series.50 The statue represents an adolescent satyr, his body arranged in a sinuous curve, who is pouring wine from a jug held in his raised right hand into a vessel he is holding below in his left hand. The replicas differ in a number of iconographic details. For instance, only few of them feature the satyr’s small tail, which was omitted by most copyists.51 The 48 This point is raised with regards to struts also by M. Hollinshed (2002a: 122). 49 See Liverani (1988–​9: 105; 1989: 64; 1993: 123), who also observes the different size of the tree trunk support. 50 Including fragments and heads, the number of the replicas rises to around thirty items. Lists are provided by Gercke (1968: 1–​16), Corso (2004: 285 n. 478), Martinez (2007b: 259). See also Weege (1929). 51 These are a fragmentary statue at the Walters Art Gallery (inv. 23.22) and torsos at the Museo Nazionale Romano (inv. 554) and in the Galleria dei Candelabri (inv. 2426). For the statue in Baltimore see Weege (1929: esp. 3–​11), Gercke (1968: 1–​2 no. St. 1), Martinez (2007b: 252–​3

149

150

150

Statuesque Statues

oinochoe that the satyr holds in his raised right hand is preserved only in the Berlin replica (Fig. 42). This statue is also the only one that still includes the attribute in the left hand. The lowered forearm rests on the high palm trunk support and the satyr holds a drinking horn (rhyton).52 The only other example that preserves the left hand is the statue now in Palermo. This, however, could not possibly have held a rhyton. The forearm is perfectly horizontal and the hand, with the flat palm turned upwards, must have supported a separately worked bowl. Variations are not limited to the position of the left arm and the attribute, but also involve statics and the mechanisms of support. Occasionally, stability was achieved by joining separate pieces and different parts of the figure, most notably the extended left forearm, which could be attached to the main block of the torso.53 However, almost all of the replicas included a vertical support on the satyr’s left side. The size and shape of the trunk, as well as its distance from and connection to the body, vary considerably. Different sculptors opted for smooth pillars that follow the figure’s gentle bend, knobbly tree trunks, palm tree trunks, or more or less elaborate combinations of attributes.54 The second-​century AD replica at the Louvre, thought to have come from Rome and to have previously belonged to the Borghese collection, has a smooth, undifferentiated tree trunk that runs very close to the leg and is partly covered by an animal skin. This detail, which has been interpreted as either a fawn’s skin (nebris) or a leopard’s and Figs. 164a–​b); for the torso at the Museo Nazionale Romano see Gercke (1968: 10–​11 no. T. 8), H. von Steuben (in Helbig4, vol. III, 140–​1 no. 2229), J. Papadopoulos (in Giuliano 1979: 89–​90 no. 70); for the torso in the Vatican museums see Lippold (1956: 111–​12 no. 12 and Pl. 53 Fig. 12), W. Fuchs (in Helbig4, vol. I, 405 no. 510), Gercke (1968: 11 no. T. 9), Martinez (2007b: 252 and Fig. 163). 52 Antikensammlung, inv. Sk 257. See Gercke (1968: 7–​8 no. T. 2), M. Kunze (in Knittlmayer and Heilmeyer 1998: 179 no. 104). The statues from Castel Gandolfo in Dresden (inv. Hm 100, 102), on which see Gercke (1968: 2–​3 nos. St. 2–​3) and C. Vorster (in Knoll, Vorster, and Voelk 2011: vol. II, 863–​76 nos. 207–​8), preserve only the handle of the jug. According to Weege (1929: 15) and Gercke (1968: 4–​5 no. St. 5, 18), the right arm with the jug of the Palermo statue is authentic; J.-​L. Martinez (in Pasquier and Martinez 2007: 272–​3 no. 66) is doubtful. The attribute and position of the left hand are discussed by Weege (1929: 14, 29) and Martinez (2007b: 249–​52); see also Daehner (2005: 241). 53 In the early imperial statue today in Baltimore, round holes were drilled to attach the satyr’s tail and the left forearm. 54 Muthmann (1951: 29, 33, 73, 115, 116, 119, 122) discusses the tree trunk supports of selected items in this series. The pillar flanking the extensively restored satyr now at Petworth House (inv. 6) bears the signature of the copyist, Apollonios, which some scholars date to the late first century BC while others consider it a modern forgery. Hutton (1914–​ 16: 163–​5) and Raeder (2000: 55–​7 no. 8) believe that the inscription is authentic, while Michaelis (1882: 600 no. 6) and Löwy (1885: 346 no. 517) are doubtful. See also Gercke (1968: 10–​11 no. T.6).

151

Figure 42  Statue of a Pouring Satyr, late Hadrianic–​early Antonine period. Marble. H. 156 cm. The head is a plaster cast of the Dresden replica. Berlin, Antikensammlung, inv. Sk 257

152

152

Statuesque Statues

skin (pardalis), is decisive in vesting the figure with an identity. An ­exquisite Antonine replica at Palazzo Altemps, that also preserves two small ­measuring points in the hair, exploits in full the descriptive potential of the support, adding an entire set of attributes –​an animal skin, a set of pan-​ pipes, and a curved crook.55 These two copies, both found in Rome and possibly carved in Luna marble, feature one or more attributes hanging on the tree trunk and an elaborate solution for the stability of the left forearm.56 Both versions include a tapering spiral strut to join the outstretched limb to the satyr’s side. Other versions of the satyr include struts of a different shape or none at all, relying, like the Berlin satyr, on a taller tree stump to support the arm. Although it is often impossible to reconstruct the shape of the struts based on the remaining stubs, their location appears fairly consistently on the upper left thigh or hip.57 The Pouring Satyr in Palermo, however, surpasses all other replicas in the creative use of its structural armature. Unlike any of the other examples, this fine first-​century AD statue has a tree stump on the satyr’s right side. The trunk, at 87  cm high, reaches the satyr’s waist and surpasses in size the supports of most pieces in the series. Nowhere else is this support so detached from the body, to which it is linked by means of two stone bridges at the hip and a point immediately below the knee. One further feature, which I have mentioned in previous chapters, singles out this statue from the rest of the series. The Palermo satyr includes a strut of exceptional size between the left hand and thigh (Fig. 43a–​c). While regularly attracting derogatory remarks –​as ‘disfiguring’ and ‘unseemly’ –​this strut has also prompted considerations on the iconog­ raphy of the prototype. Scholars have thought that such a bulky support betrays the sculptor’s wish to be true to his model, which therefore featured an open hand carrying a shallow vessel. Unlike the drinking vessel of the Berlin copy, which is considered a copyist’s addition, a bowl or cup held in the flat left hand could not be made to connect with the 55 On this statue’s support see Muthmann (1951: 73, 122) and A. A. Amadio (in Giuliano 1992: esp. 199). 56 The type of marble is determined, apparently solely on the basis of autoptic examination, by A. A. Amadio (in Giuliano 1992: 194) and J.-​L. Martinez (in Pasquier and Martinez 2007: 274). 57 Each of the four copies from Castel Gandolfo included one such strut, but this is only preserved in the Getty statue. A torso at the Museo Nazionale Romano (inv. 554) also shows the remains of a strut on the upper left thigh. The torso in the Galleria dei Candelabri (inv. 2426) differs from the rest of the series in that a thick strut was placed towards the front thigh.

153



Figure 43a–​c  Detail drawing of the strut under the left hand of the Pouring Satyr in Palermo. The strut is 39 cm long

154

154

Statuesque Statues

tree trunk below or with the figure’s side. Therefore, a strut would be needed to manage the additional weight on the hand and wrist.58 This reading conflicts with other observations. The same position could be achieved by placing a high tree trunk under the forearm, in the way that the sculptor of the Berlin replica did. Besides, the question of why the tree stump was placed on the opposite side remains unanswered. We may explain this variation by assuming that the statue belonged to a mirror-​image display together with another lost replica, which included the customary tree trunk on the left side. Indeed, pendants were very common in Roman sculptural displays and we know of a number of pieces that reverse a given type.59 Nonetheless, there is no evidence that pairing two or more mirror copies of the same type as symmetrical pendants was a common practice.60 The exhibition context of the four identical pouring satyrs from Castel Gandolfo corroborates this impression. Furthermore, in no case is the mirror reversal of a widespread type limited to the supports. A closer look at the placement of supports suggests an alternative ­explanation. The angle of the strut runs almost parallel to the figure’s lower left leg, the upper right leg (the lower right leg being drawn slightly outwards and behind), and the tree trunk on the opposite side. The anatomical partitions and the supports also create a main line that cuts the figure into two halves. The sculptor has carefully arranged the flat left hand along the line created by the short strut between the left forearm and hip, the hips themselves, the bridge between the right hip and the stump, and the upper knob on the stump. This line runs parallel to both the one made by the knees and to the lower edge of the strut, and it counterbalances the bend of the torso and the shoulders. Seen from the left-​hand side, the strut echoes the angle of the lower right leg pulled back, thus creating a symmetrical projection from the core. The smooth tree trunk, which is already bulkier than usual, seems to have been relocated on the right side to give space to the prominent strut. Together, the supports frame the figure and enhance the youth’s off-​balance pose. To this end, the supports are clearly distinguished from the body. The 58 The question is summarised by Martinez (2007b: 249–​52 and 272–​3 no. 66). 59 On mirror reversals see Lippold (1923: 166–​8), Lauter (1967: 123, 126), Vermeule (1977: 27–​ 35), Robertson (1981: 111, 125, 166), as well as Bartman (1992: passim). 60 Excellent discussions of pendants in Roman sculptural displays are those by Bartman (1988), Slavazzi (2002), and Pafumi (2007). See also the comments by Lippold (1923: 163–​5) and Koortbojian (2002: 196–​200).

155

The Individual Copy

tree stump stands at an unusual distance from the body. The strut on the opposite side is given an elaborate and artificial shape, detached from the left thigh in a sharp angle. The sculptor apparently responded to a true technical challenge –​to support an outstretched arm –​with a design that also pursued an expressive objective. In doing this, he succeeded in distinguishing Praxiteles’ invention and his own additions, the latter assisting, framing and highlighting the former. The choice in the matter of supports must have reinforced the impression that this particular version reproduced the pose of the o ­ riginal in its tiniest detail.61 To us, the gesture of the left hand, with the flat palm turned upwards, certainly recalls the famous parallel of the so-​called Marathon Boy, a bronze work of the late classical period associated with the circle of Praxiteles.62 A further observation adds to the statue’s complexity. We do not know what sort of vessel rested on the Palermo satyr’s hand; we can even ignore the question of whether the statue was carved with a precise attribute in mind. However, it seems implausible that a bowl made of marble would simply be placed on the palm, without fastening the two surfaces with a pin. The satyr likely held an attribute of a less brittle mate­rial, such as metal, in his extended left hand. So not only is a large stone strut unnecessary to sustain a metal vessel, but its very existence draws attention to the attribute and the different medium. Conspicuous struts individualise a composition in a way that any viewer can grasp easily. They establish a visual link –​by similarity or contrast –​ to other members of the series and trigger the process of comparison.63 61 W. Geominy (in Knoll, Vorster, and Voelk 2011: vol. II, 872) put forward an ingenious idea about the material of the prototype. Based on the fact that in some replicas the foot of the free leg stands flat on the ground, fully attached to the base, he argued that the original was also made of marble. This explanation is rooted in Geominy’s general view of struts (1999b: 50–​1, 57–​8 Appendix V) as deliberate suggestions of imitation of bronze and easily interpretable cues for faithfulness. Implicit in this is the idea that struts were included also in statues that did not depend on bronze prototypes in order to simulate such a relationship. For the stronger arguments that, in the case of the Pouring Satyr, militate in favour of a bronze prototype, see Gercke (1968: 18), J. Papadopoulos (in Giuliano 1979: 89), Corso (2004: 288). 62 Athens, National Archaeological Museum, inv. X15118. See Kaltsas (2002: 242–​3 no. 509) and A. Pasquier (in Pasquier and Martinez 2007: 112–​15 no. 17) with earlier bibliography. Weege (1929: 24 Figs. 25–​7) provides parallel images of the Pouring Satyr in Palermo and the Marathon Boy (left-​hand side). 63 Pioneering work on the polychromy of ancient sculpture has provided precious insight into the methods used by copyists to distinguish their work. The colour pattern of different versions of the same type did not follow the same criteria. Even pieces with an exceptional degree of surface finish could be extensively painted according to individual preference. See Østergaard (2015: esp. 115–​16) and Østergaard, Sargent, and Therkildsen (2014: 60–​1, 65–​6).

155

156

156

Statuesque Statues

Roman sculptors both employed complicated techniques to produce full-​ scale copies and ensured that their work bore the stamp of their own artistic choices. The examples that I have discussed invite us to address originality as a quality that contrasts with, relates to, and distinguishes versions of the same type. Rather than being a straightforward measure of faithfulness, this originality was established through a process of recognising and comparing artworks that was inherent in the Roman ‘ethos of emulation’.

157

7

Value and Ingenuity

Modern viewers rarely acknowledge that the athlete with a body-​scraper in the Vatican Museum, thought to replicate Lysippus’ bronze Apoxyomenos, originally included a diagonal strut from the right thigh to the wrist (Fig. 44). Longer than the athlete’s arm, this bar must have been one of the very first features of the statue that caught the eye of its viewers.1 At the time when this marble was first displayed in Rome, around the mid-​first century AD, the memory of the bronze created by Lysippus centuries before was still vivid. According to Pliny the Elder, the original Apoxyomenos was in Rome at the time of Tiberius. It had been the subject of a public scandal when the emperor removed it from the Baths of Agrippa. Following a general outcry, he was forced to return the statue to its place at the baths.2 The fame and accessibility of the original must have rendered the copy immediately recognisable. The two main differences, the material and the large strut, would also have been immediately noticed. The sculptor of the Vatican Apoxyomenos could have avoided the strut if he had chosen. He might have carved the outstretched right arm ­separately and attached it to the body. In so doing, he would have closely followed the bronze model, which had no struts. Besides, this was not the only type of athlete known by Roman sculptors and patrons to use a body-​scraper. Several replicas in both bronze and marble exist of another late fourth-​ century BC athlete se destringens, the type known as the Ephesus Scraper. The lowered arms, with the hands and strigil held together in front of the left hip, required neither piecing nor large supports to stand safely.3 In spite of these considerations, those who made and purchased the Vatican Apoxyomenos preferred a copy of this specific opus nobile, which would draw attention to its technical challenges. Working the large undercut between the torso and the arm, the strut must have presented a challenge 1 On the strut, see Geominy (1999b: 59 Appendix VII), Hollinshead (2002a: 146), Anguissola (2013: 8, 11, 14). There was also a shorter strut from the diaphragm to the left wrist. 2 Natural History 34.62. On the display context of Lysippus’ original in Rome see Bravi (2012: 137–​9). 3 For the athlete of the Ephesus type see Saladino (2006) and Daehner and Lapatin (2015: 272–​7 nos. 40–​4).

157

158

Figure 44  Statue of a man using the body-​scraper (Apoxyomenos), mid-​first century AD. Marble. H. 205 cm. Vatican Museums, Museo Pio-​Clementino, Gabinetto dell’Apoxyomenos, inv. 1185

159

Struts and Joins

to the sculptor who risked compromising the statue’s stability; creating this strut would have required many strokes of the tool close to points of potential weakness. In this chapter, I  argue that statues such as the Vatican Apoxyomenos implicitly encouraged viewers to consider the stability of such prominently supported works and to question whether in fact the presence of a strut posed further difficulty to the carving of the statue’s limbs. In addition, such struts visualised the impressive size of the marble block required for a monolithic composition, advertising both the sculptor’s proficiency and the owner’s financial means.

Struts and Joins Around the same decades when the bronze statue made by Lysippus was on display at the Baths of Agrippa, Roman sculptors created different versions of this work that approached the figure’s arrangement in remarkably individual ways.4 At the time when the Vatican Apoxyomenos was erected in Rome, the method of assembling a composition from different blocks was still more common than creating a monolithic sculpture out of a single block. By attaching a separately carved arm, the sculptor of the Vatican Apoxyomenos would have complied with current workshop practices. Instead, he employed a strategy that his contemporaries may have perceived as a technical innovation –​or, at any rate, as a departure from the standard. There seems to be no general rule as to the use of struts rather than separate carving. Draped statues provide excellent conditions for the fixing of separate pieces, which can emerge from a sleeve or fold of clothing in such a way that the join remains invisible. Nonetheless, statues such as the Dresden Artemis indicate that sculptors did not necessarily favour this option over the alternative of struts. At the same time, a number of nude figures in bold, expansive poses use joins for the outstretched limbs.5 4 Due to the poorly preserved surface, it is unclear whether the mirror-​image version of the same type, today at the National Museum in Rome (inv. 2002865), included struts or relied on joins. The torso, possibly carved in Pentelic marble like the Vatican Apoxyomenos, has been dated to around the age of Tiberius. See J. Papadopoulos (in Giuliano 1979: 335–​7 no. 199), Moser von Filseck (1988: 40–​2), M. Sapelli (in Moreno 1995: 202 no. 4.29.5). According to H. Lauter (1967: 123, 126, 128), the mirror version was displayed as pendant for a replica similar to the Vatican Apoxyomenos. For the sculptural type of the Vatican Apoxyomenos see also Weber (1996). 5 E.g. the statue of Hypnos at the Prado Museum in Madrid (inv. 89-​E), which dates to the Hadrianic age and is thought to be derived from a prototype from the third quarter of the second century BC. See Schröder (2004: 233–​8 no. 144).

159

160

160

Value and Ingenuity

In general, Roman sculptors joined the individual pieces using a socket-​ and-​tenon system or metal dowels. Piecing was extensively used and could involve everything from joining small patches, making a statue in two halves, or adding projecting limbs, heads, segments of drapery, and attributes. At the top end of the market this method allowed sculptors to create complex compositions that combined, for example, white and coloured stones.6 At the lower end of the market it provided a less expensive alternative to monolithic figures. Although joining in marble is neither an easy nor, probably, a very ­durable option, several considerations militate in its favour.7 First, this technique ensures a lower overall cost by significantly reducing the necessary size and quality of the block. A cheaper grade of marble can be chosen for the core if overall stability is not endangered by projecting limbs.8 Small scraps of marble left over from other projects, as well as blocks unsuitable for a whole figure, may be used for the attached elements. Second, piecing reduces time and labour. It enables easy adaptation of stock bodies and allows for quicker methods of carving. Several sculptors could work simultaneously on separate portions. In large sculptures the points that required careful finishing could be made easily accessible and small additions could be conveniently carved on a bench. Third, a statue composed of several pieces presents definite advantages once in place. Smaller attached parts can be easily repaired, replaced, and even substituted on occasions. However, the alternatives of piecing and of carving monolithic figures assisted by struts are not mutually exclusive. Occasionally, the two methods are combined in the same statue, as is the case with the colossal Venus from the Baths of Faustina at Miletus (Fig. 45). The sculptor carved the goddess’ right arm, raised at the level of the head, in one piece with the body, adding two almost parallel shafts from the right hip to the elbow and from the 6 The fragmentary statue of Daedalus from Philadelphia in Jordan, carved around ad 200 after a Hellenistic prototype, provides an example of the use of piecing as an independent artistic medium (Amman, Museum, inv. J922). The figure was made of five pieces and relied on skilfully concealed joins and iron rods, as shown by Claridge (1990: 150–​3, 161–​2 nn. 72–​5). This particular technique has prompted D. M. Brinkerhoff (1970: 45) to consider the statue an example of long-​distance trade, shipped in pieces for assembly at its destination. A parallel from around the same period is the virtuoso piecing of the sculptures from Silahtarağa, now in the Archaeological Museum of Istanbul, on which see de Chaisemartin and Örgen (1983; 1984). For a modern parallel of piecing as a sculptural tour de force, see E. Bartman’s considerations (2003) on the Ince Diana recreated by Carlo Albacini. 7 See e.g. Merker (1973: 9), Waywell (1978: 63), Claridge (1988: 139–​40). 8 Examples where the sculptors used an inferior grade for the bulk of the torso and made only the attached head, arms, or lower legs in high-​quality stone are provided by Claridge (1988: 144, 148).

161

Figure 45  Headless statue of Venus from the Baths of Faustina at Miletus, second century AD. Marble. H. 135 cm. Istanbul, Archaeological Museum, inv. 2004

162

Figure 46a–​b  Statue of Dionysus with a panther from the nymphaeum at Punta Epitaffio, first half of the first century AD. Marble. H. 140 with the base. Front and left-hand side view. Baiae, Museo Archeologico dei Campi Flegrei, inv. 222739

163

Figure 46a–​b (Cont.)

164

164

Value and Ingenuity

biceps to the wrist. In contrast, the left forearm was worked separately and fastened under the folds of her rich garment. In general, it is difficult to distinguish between sections that were produced as separate joins and ancient restorations that may have replaced a damaged limb or attribute. The careful placement of a join, the lack of signs of reworking, as well as the congruity between the assembling methods and the apparent shape of the block, are all elements that point to an original design.9 Nonetheless, it is also true that joins could be made with a view to easier restoration. In a few cases, however, the placement of joins allows a glimpse into the technical challenges posed by the available block of marble. The statue of Dionysus with a panther, from the Claudian nymphaeum of Punta Epitaffio at Baiae, presents a singularly complex combination of struts and piecing (Fig. 46a–​b).10 Three flat, rectilinear bars connect the left elbow to the side and the figure to its figural support –​a pillar covered on the top and front with a richly draped cloth. In addition to this, a number of miniature struts are attached to the most delicate points –​between the fingers and between the penis and scrotum. The figure also relied on skilled piece-​ work. Although the right arm was supported by a tall pillar-​like support, and the left arm by a strut at the elbow, both the right hand and the left forearm were carved separately. Additionally, the sculptor used an inserted slab for the upper back. This choice likely depended on the shape of the block: wide but flat.11 Clearly, the Roman marble carvers were conversant with a variety of techniques, among which they could choose according to the size, quality, and features of the material. On a more general level, these comments indicate the necessity of considering all sculptural works in three dimensions. The general scholarly neglect of the thickness of a statue limits the ability to study the relationship between the availability of stone blocks and the shape and treatment of works.

9 E.g. the headless statue of Venus at the Centrale Montemartini (inv. 2139) bears the traces of a join for the left forearm. However, the roughly picked surface that extends to the fold of the drapery may point to an ancient repair. See Mustilli (1939: 89–​90 no. 11) and H. von Steuben (in Helbig4, vol. II, 504–​5 no. 1725). 10 Baiae, Museo Archeologico dei Campi Flegrei, inv. 222739. See Andreae (1983: 60–​1 and Figs. 142, 149–​50, 214; 1991: 244, 258–​9) and C. Valeri (in Miniero and Zevi 2008: 160). 11 Parallels for inserted slabs on the upper back of a statue are two versions of the Dresden Youth type in Basel (Antikenmuseum, Sammlung Ludwig, inv. S630) and in the Palazzo dei Conservatori (inv. 944). For the former see Linfert (1993: 173 Fig. 12), for the latter see H. van Steuben (in Helbig4, vol. II, 329 no. 1514), C. Maderna-​Lauter (in Beck, Bol, and Bückling 1990: 638–​9 no. 168) and Linfert (1993: 163 no. 2).

165

Conspicuous Consumption

Conspicuous Consumption Although sculptors began piecing together marble statues as early as the sixth century BC, this did not become the dominant method until the second and first centuries BC and the first century AD, when the still limited supplies of statuary marble had to meet the demands of a growing market. Over time various options were developed for both gravity-​assisted and gravity-​defying joins. By the beginning of the second century AD, however, the size and quality of the available blocks improved substantially and joining became less common.12 As a consequence, a wider range of figure types became available.13 Expansive figures with irregular outlines, requiring an increased amount of stone and labour, replaced the tight poses characteristic of the first century BC and first century AD, which had reflected the confines of narrower blocks. The process through which the trade in statuary marble achieved substantial momentum was gradual, progressing at a different pace in each region. Nonetheless, we can conclude that around the time when workshops began producing heavily propped monolithic statues, marble was still a relatively expensive commodity. Unfortunately, we know very little of the financial terms on which most Roman statuary was produced. Both the stone and the labour contributed to the cost of a statue in varying proportions according to the time and place. Statues boasting bold projections, carved in a continuous block with the core, certainly required a great deal of extra marble. An exceptional composition such as the Vatican Apoxyomenos would call for almost twice the amount of marble needed for a simple vertical figure without any outstretched limbs (Fig. 47a–​d). The strut makes the consumption of marble conspicuous, visualising the size of the block and the absence of joins even from a distance.14 At the same time, such an arrangement substantially increases the cost of labour. The task of carving out such a large undercut and shaping the thin stone frame demands the meticulous work of experienced hands. In assisting large monolithic compositions, struts ‘boast of waste and skill in the amount of marble removed and in the careful carving required’.15 12 Claridge (1988: 146–​8) presents the history of piecing in Roman statuary by focusing on two forms that ‘gained widespread currency in the first century AD’, semi-​draped heroic nudes and cuirassed statues. Although the data for other categories of statuary are less readily accessible, Claridge argues (1988: 149) that the incidence of joins provides an entirely compatible picture. For an overview on the supply of marble in the late republican period and in the first two centuries of the empire see also Pensabene (1998). 13 See Claridge (1988: 149–​52). 14 See Geominy (1999b: 59 Appendix VII) and Anguissola (2013: 13). 15 Trimble (2011: 77).

165

166



Figure 47a–​d  Drawn reconstruction of the Vatican Apoxyomenos. The reconstructed strut is represented with a dotted line. Front, rear, left-​hand side, and right-​hand side view

167

Conspicuous Consumption

If these considerations hold true at a general level, the picture becomes more complicated in the case of sculptural copies. Recent studies indicate that the production of high-​quality full-​scale copies in good-​quality marble was an expensive and time-​consuming process, not necessarily less costly than a statue executed without the constraints of reproducing a standard type.16 Since the plaster models were created by casting each individual portion of a statue separately, the reproduction of a draped figure must have posed immensely higher technical challenges than that of a nude athlete with large plain surfaces that could be cast in a single piece.17 The more complex the figure, with deep undercuts or overlapping garments, the longer it takes to copy using the triangulation method as a larger number of points must be taken into account. A parallel for such skilful exploitation of the material, its perceived qualities, and economic value is the use of coloured stones to carve full-​ size statues, a widespread feature of Roman sculpture since the early imperial period.18 In particular, statues made of exotic basanite seem to encapsulate, with their dark, reflective surface, multiple layers of meaning and contradiction.19 The greenish colour of the stone obviously recalls the patina of bronze or, more precisely, of old bronze. Also, the calligraphic quality of metalwork is replicated with extraordinary accuracy. At the same time however, working this hard, unforgiving stone is far more difficult than casting bronze. The sculptor is confronted with a hard and yet easily fractured material, of which only very limited supplies were available to a few affluent patrons. Whilst emulating the visual effects of metal, the stone advertises its own costliness and technical challenges. In a similar way, struts advertise both faithful reproduction, material worth, and technical skills. 16 Landwehr (2010: 37), reports the results of an experiment conducted with the assistance of the sculptor and restorer Silvano Bertolin in order to calculate the labour required to produce a copy of the Sciarra Amazon. According to their estimate, making a full plaster cast of this statue requires about 195 form pieces for overall 400 man-​hours. A further 100 or so hours are necessary to work over the partial casts. Based on that plaster model, an experienced sculptor should work approximately 2,200 hours to carve a true-​to-​scale marble copy. Instead, free hand carving of a statue of the size and shape of the Sciarra Amazon would take some 1,400 hours, about half of the time required by the copy. 17 See Landwehr (1982: 16–​17). Of course, plaster casts could be used over and over again to create new statues, increasing the return on the initial investment. 18 See Belli Pasqua (1995) and for a summary Gasparri (1994: 269 with bibliography at p. 278). 19 See for instance the comments by J. Daehner (in Daehner and Lapatin 2015: 280–​1 no. 44) about the statue of the Ephesus Apoxyomenos type from Castel Gandolfo (Villa Barberini, Antiquarium, inv. 36405).

167

168

168

Value and Ingenuity

Ex uno lapide It seems no coincidence that in roughly the same years that saw the creation of many overpropped statues, our primary Latin source on the history of ancient art, Pliny the Elder, repeatedly praised artists who were able to create complex compositions ex uno lapide or ex eodem lapide, that is, from a single block of stone.20 Implicit at several points in Pliny’s account of marble sculpture is the idea that this medium offered fewer possibilities for artists to achieve fame than those allowed by painting or bronze statuary.21 This seems chiefly to depend on the urban environment of which the statues were a part. In consequence of their sheer quantity (multitudo operum) in a busy city, statues were easily dismissed from the viewer’s mind and their authors’ names forgotten.22 This passage in the Natural History, which provides the context and circumstances for the viewing of (marble) sculpture, suggests awareness of statuary as an urban phenomenon, which entailed shared practices, responses, and challenges.23 By Pliny’s time, statuary was an expected feature of every city in the empire and, above all, of Rome. Statues adorned most public spaces –​baths, fountains, theatres, sanctuaries –​as well as the villas and townhouses of the wealthy. Pliny’s account of the accumulation and overcrowding of images creates the premise for a section on the highest achievements of marble sculpture –​ in particular, works displayed in Rome and, among those, the sensational examples of monolithic sculpture. In his discourse, technical virtuosity that overcomes the difficulty of the material is a quality capable of distinguishing a single work from the multitude of sculptures that surround it.24 This quality is essentially a consequence of its having been carved out of a single block. This means, according to Pliny, that the Laocoon has to be considered 20 Natural History 36.34, 36.36, 36.37, 36.41. On the meaning of ex uno lapide in Pliny’s encyclopedia and the history of this literary topos see Settis (1999: 79–​81). 21 Pliny’s Index auctorum suggests that he was familiar with a number of technical treatises de pictura (‘On Painting’) and de toreutice (‘On Bronze Statuary’), but lacked comparable references for marble. As observed by Settis (1999: 41–​2, 44), this probably encouraged him to develop new strategies for the presentation and selection of artworks. 22 Natural History 36.27. In a passage from book 34 (34.35) Pliny also notes the proliferation of bronze statues, albeit in different terms. In this case, he is not referring to setting or viewing practices, but rather offers a disclaimer for the impossibility of treating the subject in full. 23 Stewart (2003: 118–​56) addresses the question of whether Romans perceived statues as a collectivity of objects or a ‘population’. This attitude is clear in a much later source from the early sixth century AD, Cassiodorus’ Formula to the urban prefect on the establishment of an architect for the city of Rome (Variae, 7.15). In this heartfelt plea for the preservation of Rome’s sculptural heritage, statues are described as ‘a population … almost equal to that created by nature’. 24 On the concept of ingenium see Rouveret (1989: esp. 342–​5 = 2014) and Robert (1995: 301).

169

Ex uno lapide

a paramount example of marble sculpture (marmoris gloria) and ­superior to any other work in bronze or painting (opus omnibus et picturae et s­ tatuariae artis praeferendum).25 Of all four statues said to be carved from a single block of stone, one –​a team of four horses with a chariot driven by Apollo and Diana –​had been placed by Augustus in a prominent public spot on the Palatine. The other three belonged to some of the most famous collections of late republican and early imperial Rome. Asinius Pollio, who achieved the consulship in 40 BC and in consequence of his victory over the Parthini in Illyria in the following year was awarded a triumph, amassed in his monumenta a sizeable collection of artworks, including a group ex eodem lapide of Dirce bound to the Bull by Zethus and Amphion.26 Varro, a contemporary of Pollio and himself a keen collector, owned a monolithic group of winged Cupids playing with a lioness. The Laocoon was displayed in an even more prestigious context in Pliny’s era. This work was the prized possession of future emperor Titus. The author of each of these masterpieces ex uno lapide is known; the virtuosity of their execution has ensured the long-​lasting memory of their sculptors: respectively, Lysias for the c­ hariot of Apollo and Diana; Apollonius and Tauriscus for the group of Dirce; Arcesilaus for the lioness and Cupids; and the Rhodian trio of Hagesander, Polydorus, and Athenodorus for the Laocoon.27 Most of the marble statues that I have discussed in these c­ hapters –​such as the Vatican Apoxyomenos, the wine-​pouring Satyr in Palermo, the New York Diadoumenus, or the Castel Porziano Discobolus –​were far less complicated in pose than the groups mentioned by Pliny. Furthermore, as copies in a world of copies, they were constantly at risk of passing unnoticed among the wealth of similar statues derived from the same prototypes. In their unique and prominent supports, these works could demonstrate their (relative) costliness, as well as the technical skills of their makers. Carved from blocks of marble larger than was strictly necessary, on a more modest scale they emulated the prestige of the excellent compositions dedicated by 25 Natural History 36.37. The statement that the Laocoon is an opus praeferendum to other works in different media is discussed by Andreae (1987: 11–​14) and Settis (1999: 9, 45). 26 For Asinius Pollio’s collection, see Becatti (1956), Isager (1991: 163–​7), and Bravi (2012: 82–​93). 27 The question of authorship occupies a prominent place in the Natural History (see the explicit comments in 36.27–​9). In general, primacy is given to the attribution of a work to a known artist and therefore some justification has to be provided for the choice of mentioning a piece whose artist is unknown. For the artists mentioned by Pliny in relationship to works ex uno lapide or ex eodem lapide see DNO, vol. IV, 645–​7 s.v. Tauriskos, Apollonios no. 1 (3432); vol. V, 147–​8 s.v. Arkesilaos no. 4 (3742), 425 s.v. Lysias no. 1 (4031), 496–​8 s.v. Athanodoros no. 2 (4110).

169

170

170

Value and Ingenuity

the emperors or displayed in the homes and gardens of the most prominent collectors. The taste for monolithic compositions described in literary sources and attested by the large number of overpropped statues, coincides with a crucial architectural innovation of the early empire, monolithic columns.28 Both sculptures and columns carved out of a single block shared the complicated –​and strictly speaking unnecessary –​additional problems and efforts involved in their production and transportation. This parallel is implicit in Pliny’s account of Rome’s cityscape, which famously include a list of obelisks.29 The monumental size of monolithic obelisks transferred to Rome from Egypt, as well as the obvious difficulties posed by their transportation and erection, contributed to the visualisation of both the engineer­ing and political triumphs of the empire. The identification of the group of Dirce mentioned by Pliny with the so-​called ‘Toro Farnese’ unearthed in 1546 in the Baths of Caracalla, is highly controversial. Notwithstanding its huge dimensions, the colossal group, now in Naples, was perhaps carved in a single block of white marble, like the work of Apollonius and Tauriscus recorded by Pliny.30 The only other of Pliny’s works ex uno lapide which, according to the historians of ancient art, may have survived to our time –​the Laocoon –​is, in contrast, certainly composed of several blocks, as was promptly observed by Michelangelo who upon its discovery in 1506 nevertheless pronounced it a ‘singular miracle of art’. The piecing was so adroitly concealed that only ‘experts in the art’ would notice.31 If we accept that the Laocoon, now at the Vatican Museums, is the same statue that so impressed Pliny, we should question whether he was aware of the statue’s multiple blocks or if he had been deceived by the skilful piece-​work. The literary tradition of the label ex uno lapide may help to clarify this point. As a well-​established topos of 28 See Fant (2008: 130–​1). In the early first century AD, Strabo (Geography 12.8.14) comments on the costly extravagance (polyteleia) of the great monolithic pillars (kiones … monolithoi megaloi) that, in his time, were being transported from the quarries at Docimium to Rome. 29 Natural History 36.69–​74 (the obelisk placed by Augustus in the Circus Maximus is said to be eiusdem lapidi at 36.71). On the obelisks in Rome in the age of Augustus see Swetnam-​Burland (2015: 68–​71, 90–​104). Parker (2014) addresses the relationship between monumentality and engineering challenges. 30 Naples, National Archaeological Museum, inv. 6002. For the technical features of this group see Prisco (1991: esp. 61). For different answers to the questions of chronology and identity see Andreae (1996), Kunze (1998: esp. 36–​8), La Rocca (1998: esp. 258, 269), F. Rausa (in Gasparri 2010: 20–​5 no. 2). A summary is provided by A. Bravi (2012: 89). 31 The remarks on the Laocoon by Michelangelo and his companion during the visit, the sculptor Gian Cristoforo Romano, are mentioned in a letter by Cesare Trivulzio on 1 June 1506, for which see Maffei (1999: 108).

171

A ‘Paragone’ of the Arts

the learned discourse on architecture and statuary, it seems to have been a conventional form of praise rather than an observation based on reality.32 Perhaps in Pliny’s conscious misuse of the label we can find the literary parallel for the workshops’ documented practice of combining highly visible struts and joined limbs in the same figure. When the posture and size of the figure and the available block of stone required that at least one of the projecting limbs be carved separately, appending a large strut to the figure could simulate the conspicuous consumption and coveted status of ex uno lapide. This hypothesis is all the more provocative if we believe that the same sculptors carved the Vatican Laocoon and the Scylla from Sperlonga. Obviously, a multi-​figure group such as the battle between Ulysses’ crew and a giant sea-​monster could not be extracted from a single block of marble. Nonetheless, the sculptors struggled to limit the number of blocks used for the task, advertising this technical choice by the inclusion of many large struts.

A ‘Paragone’ of the Arts The inscription on Ulysses’ ship at Sperlonga records three artists carrying the very names mentioned by Pliny (without patronymics and in a different order) as the authors of the Laocoon –​the Rhodians Athenodoros (son of Agesandros), Agesandros (son of Paionios), and Polydoros (son of Polydoros). The first question this signature raises is, of course, whether the sculptors are identical or if we are perhaps dealing with several generations of the same artistic family.33 Related to this is the question of the date of the Sperlonga sculptural display. Scholars have proposed contrasting chronologies that range from the mid to late Hellenistic period to the ages of Augustus, Tiberius, the Julio-​Claudian, or the Flavian emperors.34 Recent investigations have demonstrated that the fine-​grained, almost 32 Settis (1999: 49–​50) comments on the equivalence, in ancient art criticism, of being and seeming made ex uno lapide. See also Lavin (1998) for the response to this topos by Renaissance sculptors, whose passion for monumental scale and the integrity of the block became a veritable ethical imperative. 33 For the inscription, see Conticello (1974: 32–​3 and Pl. 33b). The literature on the identification of these sculptors is discussed by E. Rice (1986: 233–​50) and S. Settis (1999: 27–​40). The prosopographical analysis presented by Rice and Settis addresses the identity of the sculptors of the Laocoon not only with those at work at Sperlonga, but also with individuals mentioned in inscribed documentary sources, thus providing a highly plausible chronological framework. 34 The extensive bibliography is summarised by Bruno, Attanasio, and Prochaska (2015: 376). Mid or late Hellenistic: Lauter (1969: esp. 173), Säflund (1972: 36–​7, 70–​1), Coarelli (1973: 114–​15, 118, mid-​second century BC for the group of Polyphemus and ca. 80–​40 BC for the

171

172

172

Value and Ingenuity

uniformly white marble used at Sperlonga comes from the Phrygian quarries of Docimium, whose substantial exploitation began around the last two decades of the first century BC. The close analytical similarity strongly suggests that the marble used for all of the Sperlonga statues belonged to a large stock intended for a single sculptural project, executed by coordinated teams or a single atelier.35 Although carved in the same marble and as part of the same display, the Blinding of Polyphemus and the Scylla differ in their general composition, which responds in each case to their specific location in the grotto. The former includes numerous individual figures, while the latter is an intricate composition, probably made from only two massive blocks of marble: one for the ship of Ulysses, the other for the marine monster and its tentacles.36 Both, however, share the conspicuous use of struts that support the expressive postures of the characters that enact two famous episodes from the adventures of Ulysses. In the Polyphemus group an array of thick quadrilateral shafts support the legs and arms of the sleeping Cyclops and of Ulysses’ companions (Fig. 48). In the Scylla group, struts assist and emphasise the desperate struggle of the seamen entangled in the monster’s tentacles and wolf-​like heads (Fig. 49). Several scholars maintain that, also in light of the many struts, the two groups are high-​quality reproductions of lost Hellenistic bronze prototypes.37 An alternative explanation is that their bold designs were inspired by two-​dimensional sources such as paintings, reliefs, or mosaics.38 From this perspective, the struts would allow (and visualise) the translation of a

35 36

37

38

Scylla), Conticello (1974: 44–​9), Strocka (1999: 313). Late republican, early Augustan: von Blanckenhagen (1969: 273–​5), Kunze (1996: 204–​22), Settis (1999: 53–​4), Pollitt (2000: 100–​1). Augustan or Tiberian: Stewart (1977b: 88). Tiberian: Andreae (1987: 7; 1994: 138–​9 = 1995: 152–​3; 2003: 52, 59), Conticello (1996: 285–​6). Julio-​Claudian: Weis (2000: 138). Flavian: Hampe (1972: 58–​63). Bruno, Attanasio, and Prochaska (2015: 383–​9). On the exploitation of the Docimium quarries see Pensabene (2010). On the blocks used for the Scylla and the recumbent Cyclops see Bruno, Attanasio, and Prochaska (2015: 377, 390). For the group of Scylla see also Moreno (1994: vol. II, 616), Conticello (1996: 287), Waywell (1996: 114), Andreae (2003: 51). For the question of originality see Säflund (1972: 36–​7, 77), Stewart (1977b: 88), Andreae (1974: 104; 1987: 7; 1994: 148–​9 = 1995: 160–​2; 1999: 188–​215; 2003: 57, 59), Andreae and Conticello (1987: 24–​7), Moreno (1994: vol. II, 613–​14), Conticello (1996: 287). See also the comments by P. Liverani (2006: 34–​5). So Hollinshead (2002a: 148–​50). That the group of Scylla could derive from a two-​dimensional model was first suggested by P. von Blanckenhagen (1973: 459; 1976: 103). He writes that ‘making an exact copy of so intricate a group seems to me more difficult than designing a new one based on different prototypes’. Himmelmann (1995: 19–​21, 23, 28–​30, 33–​4, 40–​2) explores the transfer of two-​dimensional motifs into three-​dimensional works for large mythological groups.

173

Figure 48  Statue of the so-​called ‘third companion’ of Ulysses, from the Polyphemus group of Sperlonga. Marble. Sperlonga, National Archaeological Museum

174

Figure 49  Group of the Scylla from Sperlonga, detail of a sailor seized by the monster. Marble. H. max preserved 186 cm. Sperlonga, Archaeological Museum

175

A ‘Paragone’ of the Arts

two-​dimensional model into a freestanding marble statue.39 Both accounts agree that the groups derive from prototypes in a different medium. However, this question warrants a more nuanced approach. The relationship between the different sculptural materials or between two-​and three-​ dimensional media should not necessarily be addressed in straightforward terms of derivation. A different, more subtle, and culturally more relevant sort of relationship among the figural arts may be at play here. The relative merit of different materials for sculpture is a topic that surfaces occasionally in the Greek and Latin literature.40 It has been said that Pliny’s account of art collections in Rome between the late republic and his own time seems to reflect an artistic taste that prefers marble to bronze.41 Most marble works presented in the Natural History belong to collections established in the early Augustan period, which consisted almost exclusively of marbles. The late republic and early empire represent the culmination, in Rome, of a long process of assimilation of marble as a material for sculpture and architecture. Pliny himself composed his encyclopedia during a time of swift improvements in the exploitation of marble quarries around the Mediterranean, as well as an exponential increase in the quantity and quality of stone available for ambitious projects. The sculptors’ interest in exploring the limits of their material, by reproducing or emulating designs associated with bronze, is one product of this period of technological and artistic innovation. The comparison between different media was also the object of occasional discussion. Implicit correlations between the potential of sculpture and painting occur as early as the fifth century BC in Greek thought and literature.42 The parallel becomes explicit in the treatises on rhetoric composed in the first centuries BC and AD, famously in a number of passages from Cicero’s Brutus and Quintilian’s Institutes of Oratory. There, key moments and figures in the evolution of the two media are juxtaposed 39 A similar explanation has been suggested by Becatti (1937: 49–​53) for the groups of the Three Graces, whose varying supports may indicate a two-​dimensional prototype. 40 See Tuchelt (1979: 71–​4), Lahusen (1999: 38–​9, 41 n. 11), Settis (1999: 41–​2). For the anecdotes (Cicero, On Divination 1.13.23 and 2.21.48; Pliny the Elder, Natural History 36.14) about statues being already contained in the block of marble, waiting to be extracted from it, see Corso (2004: 280). On the role of sculpture and painting in the epigrammatic tradition (third century BC until ca. 50 AD) see Prioux (2006). 41 See Andreae (1987: 14) and Isager (1991: 174–​8). 42 E.g. the famous exchange between Socrates and two artists, the painter Parrhasius and the sculptor Cleiton (Xenophon, Memorabilia 3.10.1–​8), on which see Pollitt (1974: 30–​1). On the traditional comparison between poetry and the visual arts (the shortcomings of sculpture are famously examined in Dio Chrysostom’s Olympic Discourse 12.70–​1) see Schade (2007: 168–​71).

175

176

176

Value and Ingenuity

in a sequence that illustrates similar developments in the art of making speeches.43 However, it is at the time of the Second Sophistic that this particular form of comparison gains real momentum. In a passage from his Apologia (A Discourse on Magic), written after the middle of the second century AD, Apuleius compares painting and stone sculpture with the vivid reflection of nature provided by mirrors.44 Whereas the latter succeed in imitating every change of expression, all figural arts lack motion (motus). In addition to this, they fall short of reality in aspects that are peculiar to each medium  –​colour (color) in the case of stone sculpture and solidity (rigor) in painting. Philostratus the Elder devotes the preface of his Images to an examination of the mimetic potential of statues and paintings.45 Understandably, in a work about painting, the narrator argues for the pre-​ eminence of ‘imitation by the use of colours’ (zographia) over the ‘forms of plastic art’ (plastike, i.e. ‘modelling, and imitation in bronze, and the work of those who carve Lygdian or Parian marble, and ivory carving, and … the art of gem-​cutting’). Thanks to its colours, as well as to the play of lights and shadows, paint ‘permits the observer to recognise the look, now of a man who is mad, now of a man who is sorrowing or rejoicing’. Painters can convey a large palette of emotions in ‘the varying nature of bright eyes’, characterise their figures by means of different tones for the hair and garments, and set them in the landscape. As far as the depiction of emotions and the environment is concerned, sculpture is deemed to be much less effective.46 Whereas painting allows for the inclusion of a larger variety of positions and interactions, sculpture remains limited by its own materiality. This preoccupation with artistic media and materials was not exclusive to literary sources. Documented cases of two-​and three-​dimensional depictions of the same subject, in the same iconography, displayed in close proximity suggest a keen interest in the comparison of different media and the awareness of their competing qualities.47 The dialogue between sculpture 43 Cicero, Brutus 18.70; Quintilian, Institutes of Oratory 2.13.8–​13, 12.10.1–​9. After mentioning sculpture, Cicero argues that similis in pictura ratio est, while Quintilian moves from painting to the similis in statuis differentia. 44 Apologia 14. On this passage see Slater (1998: esp. 41–​4). 45 Images, preface 1–​2 (tr. A. Fairbanks, Loeb Classical Library). See Abbondanza (2001: 121–​33), Schade (2007: 173–​4). 46 S. Muth (2007) explains the diminishing interest in sculpture in the domestic spaces of the late antiquity as a consequence of the perceived inadequacy of this medium in expressing complex narrative structures. 47 See Spyropoulos (2001: 129–​32) and von den Hoff (2004: 123) for the marble group of Achilles and Penthesilea set up in the villa of Herodes Atticus in Luku along with a mosaic depicting the same scene. Cf. Anguissola (2012: 140–​2) for the case of the mosaic versions of the Dresden symplegma in the House of the Boat of Psyche at Antioch.

177

A ‘Paragone’ of the Arts

and the decoration of floors and walls not only counted among the criteria for the spatial arrangement, but was essential to the understanding of each individual element. Of course, it is inappropriate to project on ancient sources the concept of paragone that informed the development of artistic theory in fifteenth and sixteenth-​century Italy.48 It would be equally inaccurate to extend to other periods considerations triggered by the particular intellectual climate of the Second Sophistic. Nonetheless, earlier sources also seem to reflect, albeit in a less articulate way, some form of ongoing debate about the individual characteristics of materials and media for the purpose of mimesis. This is particularly clear in the case of the Sperlonga display. Even in antiquity, the two largest groups in the grotto of Sperlonga seem to have triggered debates about the properties of sculpture in relation to other forms of art. The furniture of the cave included a marble plaque with a ten-​line hexameter inscription, composed by a certain Faustinus.49 Although we cannot be sure about its placement, there is little doubt that it was designed for viewing alongside the sculptural groups. The epigram evokes the sculptures in the grotto and in particular the immensum opus of ‘the blinded eye of the semi-​beast who was weighed down with sleep and wine alike, the grottoes and restless waters, the Cyclopean rocks, the savagery of Scylla and the ship’s prow broken in the surging waters’. The programme of the grotto is explicitly presented as the visual equivalent of an epic poem, vivid to the point that it rises even above the work (carmen) of Virgil. Because the inscription was added to the display long after the installation of the marble groups themselves, we cannot consider it as a response to the original concept of the sculptural ensemble. Nonetheless, the cave’s function as the stage for elegant dining rituals indicates that the display at Sperlonga was likely assembled in order to facilitate erudite conversations about mythology and art. Sometimes, artworks appear to incorporate certain of the characteristics and properties of adjacent media. For example, in contrast with the pale surface of the stone, the vivid touches of colour on the eyes and lips of the helmsman from the Scylla group must have added a dramatic, pictorial dimension to the figure.50 In this respect, the idea proposed by Wilfred 48 On the concept of paragone in Quattrocento and Cinquecento art theory see Mendelsohn (1982) and Hessler (2014). 49 The ‘Faustinus inscription’ has been discussed extensively by M. Squire (2007; 2009: 202–​38, both with extensive bibliography), whose translation I adopt here. 50 P. Moreno (1994: vol. II, 613–​14) considers the Scylla group to be a marble copy of a bronze original, which was itself inspired by Hellenistic painting: the bronze group competed with its

177

178

178

Value and Ingenuity

Geominy in a brief discussion of this figure is worth closer consideration.51 He suggests that ancient Roman viewers would immediately recognise the struts as technical aids to the stability of marble copies derived from bronze originals. Perhaps, we can take a further step in this direction. As immediate references to the qualities of materials, struts could provide a convenient visual tool to articulate a less specific type of relationship –​the comparison between materials. In cases such as the group of Scylla, expansive poses and an intricate design were part of a project aimed at demonstrating the artists’ commitment to achieving images that lay beyond the capabilities of stone. Alongside other methods, such as surface finish and colour, struts could broadcast the artist’s struggle with the limits of stone and highlight the emotional range of his works. Like other technical and stylistic devices such as the use of polychromy, struts allow the transgression of material limitations. By doing so, they call attention to those material constraints and to the field of operation of the artist. The medium of marble is, at once, surpassed and present.

source of inspiration in expressivity and intricacy; the Roman copyists put enormous effort into the task of imitating their direct bronze antecedent while incorporating some elements of the painting, such as polychromy, in their work. Detail images of the helmsman’s head are provided by Moreno (1994: vol. II, 604 Fig. 750, 629 Fig. 768), Andreae and Parisi Presicce (1996: 307 Fig. 40), Andreae (1999: 213). 51 See Geominy (1999b: 50–​1).

179

8

Carving and Tradition

The slightly over-​life-​size statue of a drunken Bacchus which, in the words of Vasari, combines ‘the slenderness (sveltezza) of a young man and the fleshiness (carnosità) and roundness (tondezza) of a woman’, was the young Michelangelo’s first commission in Rome (Fig.  50).1 Carved in a single block of stone, the god is depicted in a slightly reeling pose, a goblet of wine in his right hand. The god is apparently unaware of a little faun who gambols behind him, covertly munching on the grapes that fall from his hand. Although he carefully polished his figures, the sculptor could not conceal imperfections such as the dark vein that runs down the backs of Bacchus and the faun, or the stains on the god’s cheeks. The difficulty in purchasing flawless blocks of marble represented a source of concern and disappointment for Michelangelo at the beginning of his activity in Rome. In a letter written to his father in 1497, immediately after completing the Bacchus, he voices his frustration at the money he has just ‘thrown away’ in a faulty stone that had to be replaced.2 Soon after this incident, in order to avoid further setbacks Michelangelo made the first of several journeys to Carrara where he would select his material himself, preferably from the extraction site known as Polvaccio. Whenever personal travel was impossible, he would entrust the initial stages of the work to close associates, a small team of specialised quarrymen and transporters.3 Although we lack such detailed evidence for earlier periods, there is no doubt that preoccupations about the material and its quality were not exclusive to Renaissance artists. Every composition choice and technical feature of a statue, including extrinsic elements such as supports and plinths, requires a contextual account that includes the needs of a specific commission within the established practices of a period and local traditions. In these final pages, I sum up observations of the previous chapters,

Florence, Museo Nazionale del Bargello, inv. S.10. On history and reception see Barocchi (1982: esp. 7). 2 See Stone and Stone (1962: 16–​17), Hirst (1985: esp. 154), Scigliano (2005: 54–​6). 3 See Scigliano (2005: 74–​7 and passim). On the exploitation of the Carrara quarries in the fourteenth and fifteenth centuries see Klapisch-​Zuber (1969). 1

179

180

Figure 50 Michelangelo, Bacchus, 1496–​7. Marble. H. 209 cm with the base. Florence, Museo Nazionale del Bargello, inv. S.10

181

Workshops

in an attempt at pinpointing the actual contribution of struts to the study of workshop’s traditions and practices.

Workshops A passage in an early fourth century AD dialogue, which was traditionally included in the corpus of Lucian, corroborates the impression that the quality of stone was considered responsible, no less than the artist himself, for the success of a work.4 While describing a visit to Aphrodite’s sanctuary at Cnidus and the emotions aroused in him by Praxiteles’ famous cult image, the narrator Lycinus recollects noticing a stain on the immaculate surface of the statue.5 This mark prompts further comments on the challenges posed by a damaged block of stone, to the point that ‘many potential masterpieces of beauty are thwarted by bad luck’. The exceptional skills of Praxiteles are manifest in that he was able to underplay a ‘natural defect’ of the sort that would commonly jeopardise the whole project. Both in antiquity and in later periods sculptural workshops could, to a degree, select their material and adjust the organisation of the workshop according to the needs of a particular commission. In the case of large and ambitious projects, sculptors sometimes chose to rely on a single, dependable source for stone. Carved at some point between the final decades of the first century BC and the beginning of the following century, the magnificent statuary marble used for all of the groups at Sperlonga came from the same excavation point in the Docimium quarries. Recent analyses of the statuary from the nymphaeum that Herodes Atticus, a Roman senator and Greek aristocrat with strong ties to the imperial family, built at Olympia around AD 150, provide comparable evidence. The whole sculptural programme, which included statues of Zeus alongside the portraits of members of the imperial family and of Herodes’ family, was executed in the same stock of high-​quality Pentelic marble. The pieces displayed in the niches of the two-​ storey semi-​circular façade also share the peculiarity of having been carved out of blocks of a size that was above average even for the period.6 4 Pseudo-​Lucian, Amores 15. 5 Lycinus’ comments are meant to introduce a tale of agalmatophilia, about a young man said to have been so in love with the image that he stole into the sanctuary at night and tried to copulate with the statue, leaving an inappropriate stain on Aphrodite’s thigh. 6 On the marble used for the sculptures in the nymphaeum and the size of the blocks see Kane et al. (1999: 322–​3). Studies of the relationship between the shape of sculptures and the availability of stone have been attempted by S. Kane and S. C. Carrier with regard to different geographical environments such as Cyrene (1988) and the Eastern provinces (1992).

181

182

182

Carving and Tradition

Within the complex framework of the Roman carving traditions, the evidence of both figural and non-​figural supports has been the object of only cursory, piecemeal accounts. In a number of cases, the similarity of strut-​work has been considered as evidence that the pieces were part of a coherent programme or were produced in a single sculptural workshop. Understandably, the divergence in the form and location of struts has provoked the opposite conclusion.7 The problem with this approach is particularly evident in the case of statues belonging to the same type, displayed as pendants, whose supports have sometimes received contrasting explanations. The two statues of the Dresden Youth type from Castel Gandolfo, for instance, have been considered the products of different artistic personalities on account of the different treatment of the torsos and the slight variation in the size of the tree trunk support and the placement of the struts. However, the group of four Pouring Satyrs, also found in Castel Gandolfo, have been ascribed to the same hand in consideration of their similar tree trunk supports, although each presents distinctive touches in the treatment of the hair and in the shape and placement of the stump.8 The questions posed by almost identical pendants are epitomised in the famous case of the two under-​life-​size figures of Pan found by Gavin Hamilton in the villa at Monte Cagnolo outside Lanuvium, today both at the British Museum.9 Although typologically close, supplemented by matching tree trunk supports and both signed in Greek characters by a sculptor named M. Cossutius Cerdo, the two diverge in material and treatment. It is likely that one was made in Proconnesian marble from the Agean during the first century BC, while the other was carved in Luna marble during the age of Augustus –​not just as a version of a widespread type, but rather as a pendant to the former.10 As such, the later statue imitates its intended companion also in features that are extrinsic 7 Based also on the different strut-​work, C. Schneider (1999: 23–​4, 26, 43) assigns the statue of Venus and the group of the Muses from the Baths of Faustina at Miletus to different workshops. 8 On the differences in the tree trunk supports, see W. Geominy (in Knoll, Vorster, and Voelk 2011: vol. II, 870, 872). 9 London, British Museum, inv. 1805,0703.29 (late republican) and 1805,0703.28 (Augustan). See Smith (1892–​1904: vol. III, 61–​2 nos. 1666–​7), Rubensohn (1935: 56–​8), Arnold (1969: 247–​8 nos. 2–​3), Ridgway (2002: 205–​6 n. 10). For general information on this replica series see Arnold (1969: 247–​52) and Vorster (1993b: 204–​8). For the find context in the villa at Monte Cagnolo see Vermeule (1977: 68–​70) and Neudecker (1988: 161–​4), for the tree trunk supports see Weinstock (2012: 252–​3 nos. 61–​2). The struts on the upper body must also have been similar, although the arms are modern additions; see Vorster (1993b: 195). 10 Löwy (1885: 263–​4 no. 376a–​b). The signatures diverge in both the ductus and content; only the earlier one informs us of the status of the artist as a freedman. See also Arnold (1969: 248), Vorster (1993b: 202), Kreikenbom (2013: 72).

183

Workshops

to the body type and iconography such as the structural armature and the inscription. The difficulties in the assessment of struts in order to identify specific artistic personalities become clear if we return to the marble sculptures recovered from the Antikythera shipwreck. These constitute the earliest known set of marbles that make extensive use of large supports. Understandably, scholars have highlighted the number, size, and shape of the struts as one of the main technical elements that point to a single source for the Antikythera sample.11 Struts, however, counted among several solutions employed to ensure the stability of marble statues. Pieces that rely on different devices –​struts, joins, or a combination of both –​may share other meaningful similarities in subject, iconography, and style. Although the statues from the cargo were produced in the same material, period, and probably place, we ignore what their intended context was and whether any of them were ever meant to be displayed together. The number, size, placement, and shape of struts cannot provide, alone, any positive attribution to individual ateliers –​let alone to individual carvers. Nonetheless, if the display context is known, technical details such as the struts may add to the body of evidence used to confirm attributions based on style and treatment.12 For instance, in the case of the sculptural programme at the Claudian nymphaeum of Punta Epitaffio, an analysis of the design, style, and stability measures of the individual works indicate three sets of statues, in all likelihood executed by different carvers.13 Two of the figures, carved in very large blocks of marble, must have belonged to a group offering wine to Polyphemus: Ulysses with the cup and his wineskin-​bearing companion (Fig. 51). Both figures stretch their legs in what must be a dramatic movement of approach to the Cyclops. A  robust support resembling a tree trunk, placed under the advanced thigh, supports the pose, while more struts connect the extremities at various points –​at the knee or elbow and under the hand with the cup.14 All these bars have a round profile and their surface finish matches the bark of the tree stump. A number of smaller bridges link less prominent elements. The statue of Dionysus with a panther, which I have discussed in Chapter 7, presents a number of both similarities and differences from 11 12 13 14

E.g. Hollinshead (2002a: 138); see also the comments in P. C. Bol (1972: esp. 93). See Hollinshead (2002a: 238) and Anguissola (2013: 12–​13). See Andreae (1983: esp. 62; 1991: 245) and M. Hollinshead (2002a: 138). The figure of Ulysses is 175 cm high (without head) and 120 cm deep, the wineskin-​bearing companion is 169 cm high for a depth of 110 cm. See C. Valeri (in Miniero and Zevi 2008: 156). B. Andreae (1991: esp. 242, 263–​4) argues that the struts of the Baiae statues indicate that they reproduce prototypes in bronze.

183

184

Figure 51  Statue of the wineskin-​bearing companion of Ulysses from the nymphaeum at Punta Epitaffio, first half of the first century AD. Marble. H. 169 cm. Left-​hand view. Baiae, Museo Archeologico dei Campi Flegrei, inv. 222737

185

Workshops

the Homeric group. They all employ many struts between the torso and arms, as well as between the legs and the figural supports. All include small connectors for the finest details in the composition. A shared decorative taste is reflected in details such as the cup in Ulysses’ hands and the vine branches wrapped around the panther’s neck. The statue of Dionysus, however, is not only smaller, but is the result of joining together the separate parts. Furthermore, its connectors are of a different shape –​ faceted, rectilinear bars –​from those of Ulysses and his companion. The extensive piecing reflects the limits of the available block’s dimensions. If the statues for the nymphaeum were all made according to a coherent programme, the sculptor probably chose a flat block of stone for the Dionysus because it was intended to be placed in a niche.15 Furthermore, the sculptural programme of the nymphaeum includes several other pieces whose composition seems to follow distinct and individual criteria, although they all depend heavily on piecing and joined extremities.16 The statue of Dionysus with a panther was displayed in close proximity to another image of the same god, this one with an ivy crown, both placed in niches along the same wall.17 These two are similar in subject, pose, and size. Their assembling technique is also comparable; each is composed of several pieces carved individually and joined to the core. However, as is especially apparent in the treatment of the head as well as in many anatomical details, the two pieces are the work of different artistic personalities. The sculptors employed in furnishing the nymphaeum may well have belonged to the same larger team or worked in close contact, but nonetheless relied on different techniques suited to the individual modus operandi and the demands of each piece. We should not forget that the character of the specific block of stone is likely to affect the placement of struts and to cause some alterations in the artist’s plan. The provenance and quality of marble also count among the variables that recommend caution in viewing struts as technical indicators for origin or authorship. In order to illustrate this point, we may return to the statuary displayed in the nymphaeum of Herodes Atticus at Olympia and examine some of its peculiarities. 15 For a reconstruction of the sculptural display in the nymphaeum see Andreae (1991: esp. 243–​ 4 Figs. 3–​5). 16 For a portrait statue of Antonia Minor (inv. 222738) and the statue of a young girl (inv. 222740) see Andreae (1983: 54–​8 nos. 6, 8 with conservation reports by C. Piccoli at pp. 79, 80 nos. 2, 4; 1991: 245–​50, 255–​6, 259–​61) and C. Valeri (in Miniero and Zevi 2008: 159, 161). 17 Baiae, Museo Archeologico dei Campi Flegrei, inv. 222741. See Andreae (1983: 61–​2 no. 11 with a conservation report by C. Piccoli at p. 78 no. 1; 1991: 244–​5) and C. Valeri (in Miniero and Zevi 2008: 162).

185

186

186

Carving and Tradition

Figure 52  Torso of Zeus, from the Nymphaeum of Herodes Atticus at Olympia, mid-​ second century AD. Marble. H. max preserved 167 cm. Olympia, Archaeological Museum, inv. Λ 170

The statues and fragments associated with the nymphaeum of Herodes Atticus provide exceptional evidence for a large collection whose setting, date, and material are known. The monument consisted of a semi-​circular façade framing an apsidal basin. In Renate Bol’s reconstruction, which I follow here in the identification of individual figures, twelve portraits of the imperial house were placed in the niches of the lower storey, while portraits of Herodes Atticus and his family occupied the upper level. On both levels, an image of Zeus occupied the central niche (Fig. 52).18 The statues from the nymphaeum fall into five groups: naked or semi-​naked images of Zeus, cuirassed imperial portraits, portraits of male members of Herodes’ family wearing a toga or himation, draped portraits of members of both the imperial and Herodes’ family, and children’s portraits. The works that make

18 On the nymphaeum, its building history, and sculptural decoration see R. Bol (1984); the main questions about the sculptural display are summarised by Trimble (2011: 237–​44).

187

Workshops

up each class of statues display a variety of techniques. These are compact compositions carved out of a single block of stone without the aid of struts, monolithic statues with struts, figures assembled out of two or more separate pieces, and works that combine struts and joins.19 Some of the statues from the nymphaeum also include neck struts, associated with areas of lesser finish on the back. Significantly, these occur on both monolithic compositions and others with an inset head.20 The marbles set up in the nymphaeum differ both in the use of static implements or assembling techniques and in style. This is particularly evident in the case of figures with the same body type, such as the three versions of the Large Herculaneum Woman. These were presumably very close in size but were produced by different methods, followed different criteria in the use of struts, and were treated differently in the drapery and other details.21 The resulting impression is that most of the statues for the nymphaeum belong to a single project, executed during a limited time span (approximately AD 149–​53). It is likely that a single workshop was entrusted with such a large and ambitious commission, but a number of artists must have worked on the individual figures. Each sculptor or team made stylistic and technical choices according to their own sensibility, the subject and iconography, and the block of marble available.22 As the versions of the Large Herculaneum Woman indicate, technical choices about struts and joins do not necessarily match parallel divergences in style.

In the statues of Antoninus Pius (head: Olympia, Archaeological Museum, inv. Λ 165), Hadrian (inv. Λ 148), and Marcus Aurelius (inv. Λ 150) the portrait head was carved as a separate piece. The statue of Marcus Aurelius includes a variety of stability techniques; the head and the raised left arm were joined to the core, a stone bar linked the legs and two other struts connected the right arm to the side. See R. Bol (1984: 151–​9 nos. 28–​30 and Pls. 15–​21). 20 Examples of statues that combine joined heads with struts at the nape are the Large Herculaneum Woman with a portrait of Faustina the Elder (inv. Λ 155) and the Small Herculaneum Woman with a portrait of Athenais, Herodes’ daughter (inv. Λ 159). See R. Bol (1984: 173–​5 no. 37, 180–​2 no. 43 and Pls. 35–​7, 48–​50), Trimble (2011: 365 no. 5). 21 The three statues portray Faustina the Elder, Sabina (Berlin, Antikensammlung, inv. Sk 1398), and Herodes’ wife Regilla (Olympia, Archaeological Museum, inv. Λ 156). The first two combine a stock body and an inset head (with a neck strut in the case of Faustina the Elder) while the latter is carved from a single block of marble. The statues of Regilla and Faustina the Elder share a similar pattern for the drapery, although the treatment of the individual folds is more delicate in the former. In the portrait of Sabina, the arrangement of the drapery follows a slightly different pattern and includes some individual details such as a tiny strut linking the lowest edge of the mantle to the calf. See R. Bol (1984: 171–​5 nos. 36–​8) and Trimble (2011: 365–​7 nos. 5–​7), both with comments on the stylistic differences within the set. 22 See R. Bol (1984: esp. 20–​1) and Hollinshead (2002a: 139). 19

187

188

188

Carving and Tradition

Contexts The nymphaeum of Herodes Atticus is one of the very few instances where neck struts occur outside Asia Minor. Most scholars believe that these protrusions of unworked stone behind the neck were implements for transportation, left in place to protect one of the most delicate points of a statue during travel. This explanation also has been put forward for the sculptures associated with the nymphaeum, which are thought to have been produced elsewhere and then transferred to Olympia.23 The assumption that neck struts had an exclusively practical function is, however, not supported by material evidence. Although neck struts would certainly improve a sculpture’s chance of surviving transport and lifting intact, this idea contrasts with other considerations. First, neck struts often occur on miniature sculptures, which did not require any such precautions. Secondly, one wonders why supports behind the neck also were left on full-​ size veiled statues that were already reinforced by a thick screen of sculpted fabric as well as on female heads whose hair was combed into large buns, strategically located between the nape and neck. The posthumous honorific portrait of Faustina the Elder from the nymphaeum of Herodes Atticus, with its rough boss on the back of the neck, provides a good case in point. This sculptural type offered the possibility to create a cushion of drapery between the shoulders and nape, reinforcing that sensitive spot without the aid of a neck strut. Furthermore, Faustina’s head was inset and therefore did not require additional support.24 Regional concentration is indisputable. Neck struts are relatively common in statuary found in (or known to come from) Asia Minor, specifically western Anatolia. A handful of examples come from the North African sites of Carthage, Cyrene, Leptis Magna, Utica, and Thuburbo Majus. Neck struts are almost unknown elsewhere.25 In consequence, scholars tend to rely on neck struts as evidence for an eastern origin. The reliability of nape struts as indicators of origin has been, at times, stretched to its limits, when statues of unknown origin are ascribed to Anatolian workshops solely on the basis 23 See R. Bol (1984: 21), who believes that the statues for the nymphaeum were carved in Athens. On the neck struts as implements for transport, see also Linfert (1979: 781) and Ridgway (1984: 88). 24 The portrait statue of Regilla of the Large Herculaneum Woman type shows that drapery could be arranged behind the neck so as to support a head carved in a continuous block of stone: see the images in R. Bol (1984: Pls. 32–​3). 25 See Inan and Rosenbaum (1966: 10), Inan and Alföldi-​Rosenbaum (1979: 3). See Braemer (1990: 189–​95) for North Africa, Friedland (2012: 24–​5) for Greece, Stirling (2005: 118), and Inan and Alföldi-​Rosenbaum (1979: 3) for Rome.

189

Contexts

of their strut.26 Alone, the presence of a neck strut is not sufficient to pinpoint the finishing location and the place of display of a statue. A number of factors come into play when considering the origins of a piece, including the source of the marble, the geographic origin or education of the sculptor, and the actual location of the manufacture.27 Unfortunately, petrographic analysis fails to provide unambiguous data about the geography of neck struts. The sample of sculpture that has been investigated to this purpose is still very limited. Therefore, it has so far happened very rarely that isotopic analysis and the study of stylistic features have converged in confirming that the marble was both quarried in the Turkish peninsula and worked by a sculptor trained in western Anatolia.28 When a carver had no other means to reinforce the neck, a rough mass of stone extending from the shoulders would certainly improve the statue’s chance of surviving transport intact. However, the frequent redundancy of this measure and the regional concentration suggest that the feature, probably introduced at first as a practical precaution, had become in time more of a ‘visual convention’. It is likely that sculptors active in western Anatolia, or trained in that tradition, regarded neck struts as a ‘routine part of human sculptural representation’.29 An alternative explanation is that reinforcements for the nape may reflect the division of labour within the workshop or between the quarry and the workshop. The craftsmen in charge of roughing out the figure may have developed a custom of leaving a rough boss in place, to be removed at a later point if necessary.30 In cases such as the statues from the nymphaeum of Herodes Atticus that were displayed 26 In their updated catalogue of portrait sculpture from Turkey, J. Inan and E. Alföldi-​ Rosenbaum (1979: 3, 106–​7 no. 55, 336 no. 336) determine on the basis of the neck strut the origin of two pieces: the Faustina at the Getty Museum and a mid-​third century AD male head in Princeton (Art Museum, acc. no. 1970–​1) on which see also M. Fuchs (in Padgett 2001: 77–​9 no. 19). Trimble (2011: 444 no. 173) argues against the use of this criterion alone to determine the origin of a statue. 27 See the comments by Friedland (2003a: 338; 2012: 83–​9 nos. 4–​5, 134–​7 no. 24) and Stirling (2005: 108, 117–​19). 28 The statue of a young satyr from the villa of the Anicii along the Via Latina now in Boston (Museum of Fine Arts, acc. no. 1974.127) exemplifies this set of questions. Notwithstanding the location of the find in the environs of Rome, the presence of a neck strut suggests that it may be the work of craftsmen trained in the Anatolian tradition. This conjecture has been supported by isotope analyses (conducted, according to the museum’s website, by Harvard Lab No. HI094), which could circumscribe the origin of the stone to a small number of quarries in the eastern Mediterranean. See Comstock and Vermeule (1976: 113–​14 no. 172) and Vorster (1993a: 174 no. 8). 29 For this explanation see Hollinshead (2002a: 132–​3, 140–​1). 30 Leaving a neck strut in place until the final stages of the work is still a common workshop practice, as I witnessed during a visit at the Cervietti workshop in Pietrasanta in June 2016 (organised by Sonia Maffei as part of the conference Le voci del marmo).

189

190

190

Carving and Tradition

against the wall in a niche, the removal of the neck strut was perhaps considered unnecessary. A further possibility is that a rough knob behind a figure’s neck could also be helpful in the effort to lift the marble statue into and out of its crate or to place it in its final location without having to touch the hair, neck, and other parts of the body. In general, neck struts are more frequent in the large-​scale statuary of the Hadrianic and Antonine periods and in late antique mythological statuettes. Spiral struts also share this chronological concentration. The same distribution pattern, therefore, applies to features that stand at the opposite ends of the range of possibilities in matters of surface finish and elegance –​ on the one hand, structural implements that most visibly embody the idea of unfinished work and, on the other hand, supports that have an obvious decorative purpose. Like the rough bosses behind the neck of many statues found in western Anatolia, struts embellished with a spiral groove also betray definite geographical associations. Spiral struts are concentrated in second century AD statuary found in or around Rome. The questions that this cluster poses are very similar to those raised by the sample of statues with neck reinforcements. In most cases, based on autoptic examination, scholars have identified the material of these works as Luna marble. This seems to point to workshops in Italy carving local marble. Recent work on the identification of marble, however, warrants the utmost caution with respect to such broader narratives. In fact, it has been demonstrated that only laboratory analyses provide reliable indications about the provenance of white statuary marble. Scholars have suggested two possible sources for the spiral strut design. This form may be derived from tree trunk supports such as those of the Doryphorus from Pompeii, carved with parallel horizontal ridges that loosely imitate the surface of bark. An alternative source of inspiration could have been provided by spirally fluted columns such as those seen in the Vesuvian area as early as the first century AD.31 In other words, the spiral motif was likely first developed in the early imperial period for figural supports or architectural elements and later adapted for ornamental struts. In my opinion, the question of derivation is much less relevant than 31 These hypotheses are put forward in Hollinshead (2002a: 130), who suggests tree trunk supports such as the one of the Doryphorus from Pompeii or fluted columns similar to those discovered at Stabiae as possible models. On the fluted columns at Stabiae see Benson (1959: 255). For the treatment of tree trunk supports with parallel ridges, see also Weinstock (2012: 135–​65, 269–​70). On the fluting decoration of Roman sarcophagi and its origin see Huskinson (2015: esp. 76–​80).

191

Size

the reason why this design proved so successful. The emergence of spiral struts, I believe, is expression of the taste for elaborate decorative solutions typical of Roman sculpture in the Antonine age. In that period, sophisticated spiral patterns seem to have been particularly fashionable for both struts and elements such as garlands, handles, and other figural attributes.32 The occurrence of spiral patterns for supporting or connecting devices in a small set of sarcophagi produced in Rome reinforces the impression that they belonged to the decorative language of mid-​second century AD sculpture.33 In later periods, struts treated with a spiral motif appear mainly on statuettes found outside Italy, made of stone quarried in the Eastern Mediterranean.

Size Spiral struts occur both on large-​and small-​scale statuary. One of the best-​ known examples of this treatment is a statuette in the Boston Museum of Fine Arts, which is a miniature version of a mid-​fifth-​century BC weary Heracles type. This statuette, only 57 cm tall including its plinth, incorporates a short, twisted strut to bridge the gap between the hero’s thigh and his club (Fig. 53).34 In the realm of small-​scale statuary, it is particularly evident that decorated struts were employed for both finely crafted and poorly executed statues.35 In general, it seems that the assemblage techniques and stability measures adopted for large-​scale statuary were equally common in smaller pieces and 32 E.g. the spiralling design on the arms of the lyre of a seated Hermes found in Mérida (Augusta Emerita), dated to the mid-​second century AD according to an inscription on the turtle shell (Archaeological Museum, inv. 580), on which see García y Bellido (1949: 84–​6 no. 66). See also the comments by E. E. Schmidt (1968: 92). 33 As in freestanding statuary, spiral struts sometimes occur on sarcophagi from later periods, e.g. the Badminton Sarcophagus with the Triumph of Dionysus (New York, Metropolitan Museum of Art, J. Pulitzer Bequest, 1955, no. 55.11.5), on which see McCann (1978: 94–​106 no. 17). The marble has been identified as Parian lychnites by Van Keuren et al. (2015). See also the sarcophagus with the Triumph of Dionysus at the Boston Museum of Fine Arts (acc. no. 1972.650) in marble from Proconnesus, where a small spiral strut connects the last dancer to the right to the satyr supporting Heracles. See Comstock and Vermeule (1976: 153 no. 244). 34 L. D. Caskey (1925: 133–​5 no. 64) considers the profiled plinth, the drill marks in the hair and the strut as indicators of a date in the second century AD. The museum’s website reports the results of analyses carried out by the University of South Florida, which demonstrated that the piece is made of marble from Göktepe (near Aphrodisias). 35 The grooved rod between the head and sceptre of a rather crude statuette of Zeus in the Side Museum (inv. 469, H. 22 cm) is astonishing in two respects: first because of its excessive size in relationship to the figure and second because it is the only part of the composition that seems to indicate some decorative ambition. See Inan (1975: 170–​1 no. 95 and Pl. 78.2–​3).

191

192

Figure 53  Statuette of Heracles, from Rome. Marble. H. 57 cm with the plinth. Boston, Museum of Fine Arts, inv. 14.733

193

Size

even miniatures around or below 50  cm. All the most widespread types of struts also occur on these statuettes –​bars between the calves, bridges that link the main figure to an attribute, an ancillary figure or a support, long shafts attached to outstretched arms, short connectors for details of the body or drapery, neck struts, and spiral grooves.36 Elizabeth Bartman’s study of the miniature versions of three well-​known series –​the Resting Satyr, the so-​called Lateran Poseidon, and the Heracles type known as Epitrapezios (‘on the Table’)  –​has demonstrated that the criteria followed by the carvers of larger replicas correspond to those employed for small-​scale statuary. All three series provide comparable evidence, including items that display an array of both large and tiny struts to sustain limbs and projections of very small size.37 In particular, as I have already discussed, the full-​size replicas of the Resting Satyr, whose average height is around 170 cm, are quintessential in visualising the range of variations in the relationship between the figure and its support that were available to the Roman copyists. The lateral projection, caused by the satyr’s sharp thrust of the hip, is sometimes reduced and the tall support is closer to the body. In other examples, the distance between the two is enhanced and a large void space divides the main vertical masses. Smaller replicas reflect the same degree of individuality in the arrangement of the composition and, just like their larger counterparts, often include thick struts to connect the body and the support.38 The method of piecing, by no means restricted to large or colossal compositions, was employed for some small-​ scale items in this series, which occasionally combined struts and joins.39 36 E.g. the examples illustrated in Filges (1999 and Pls. 33–​9). A fine mid-​second century AD statuette of the Hunting Diana from Caesarea Mauretaniae (Cherchell, inv. S 7) is exemplary for the extensive use of struts in a composition of about 60 cm in size. See Landwehr (1993: 40 no. 26 and Pls. 34–​5). 37 In the case of the Heracles Epitrapezios type, contrasting solutions are employed in the marble miniature at the Cleveland Museum of Art (inv. 55.50), assembled from different pieces, and in the limestone statuette found in the ruins of Niniveh (London, British Museum, inv. GR 1881.7–​1.1 no. Sc. 1726), with a bulky strut from the right knee to hand. See Bartman (1992: 173 no. 3, 181 no. 15). In the series of the Lateran Poseidon, the sole items of monumental size –​the name piece found at Portus (Museo Gregoriano Profano, inv. 10315) and a fragmentary mirror-​image version in Istanbul –​included a quadrangular bar connecting the raised arm to the hip. Miniatures often rely on similar devices. See for instance the statuettes from Eleusis (Museum, inv. 5087, also with a tiny strut connecting the penis to the drapery), Herakleion (Museum, inv. 250), and Carthage (Tunis, Bardo National Museum, inv. C 949). See Bartman (1992: 132–​4 nos. 10 and 13, 143 no. 29). 38 A statuette at the Louvre (inv. N III 976, no. Ma 665, preserved H. 52 cm) has a strut ca. every 12 cm. See Gercke (1968: 44 no. Stte. 11), Bartman (1992: 59, 97–​8 no. 16), J.-​L. Martinez (in Pasquier and Martinez 2007: 264–5 no. 62). 39 The method of piecing is used for miniatures in Athens (National Archaeological Museum, inv. 3198) and Burgos (Archaeological Museum), on which see Gercke (1968: 41–​2 nos. Stte.

193

194

194

Carving and Tradition

Struts in small figures pose further problems of interpretation because they often appear non-​essential and oversized in relationship to the size and weight of their compositions, whose transport and display did not require such precautions. According to Bartman, these struts function as ‘simple visual clues’ that advertised the work as a copy.40 Although this account may well be true in the case of replica series, the sample of overpropped miniatures is so large that a broader explanation is required. As in larger works, struts on small-​size statues and miniatures may advertise the sculptor’s skill in producing elaborate compositions or in perfecting details. Additionally, struts could be used as a more generalised allusion to full-​scale statuary, as they were widespread and recognisable features of freestanding marble statuary, a conspicuous intimation of the statuesque.

Developments The role of struts on miniature figures is particularly significant in the sculptural production from the mid-​third century to the late fifth century AD. In the past thirty years, scholars have gained an increasingly detailed understanding of the final stages of the so-​called ‘mythological’ or ‘ideal’ sculpture. They have highlighted some general trends such as a declining interest in three-​dimensional representation and organic form, all traits that are generally summarised as a growing movement towards abstraction.41 Many examples of late antique mythological sculpture are elaborate small-​scale compositions that include lacy strut-​work, subsidiary figures, and elements of landscape. In these cases, struts often become dominant elements, creating the effect of lattice-​work in marble. Although the level of finish varies from piece to piece, interest in strut-​work appears to be a standard element for this category of sculpture.42 Together with large struts, ostensibly oversized in relationship to the actual static needs, features such as bridges in the drilling of the hair, interdigital struts, and miniature undercutting display the sculptors’ capacity for virtuoso carving. 1, 3) and Bartman (1992: 88–​90 nos. 2–​3). A particularly fine mirror replica from Terracina (Archaeological Museum, inv. 25), which may date to the Claudian period, not only included inset elements (the head and penis), but also relied on a strut to join the body to its figural support. See Gercke (1968: 45–​6 no. Stte. 15), Broccoli (1982: 35–​6 no. 30), Bartman (1992: 100–​1 no. 20). 40 See Bartman (1992: 39) and Hollinshead (2002a: 120–​2). 41 For the production, diffusion, and display of late antique mythological statuary see Hannestad (1994; 2007). 42 See Stirling (2005: 71–​3, 101, 107–​8).

195

Developments

The main stylistic characteristics of late antique mythological statuettes are epitomised in the late fourth-​century Hunting Diana from Saint-​Georges-​de-​ Montagne, discovered at the site of a large villa in southwest Gaul. The goddess, her animal companion, and the tree behind her are all connected by a web of struts that emphasise Diana’s stride forward. A second sculptural group was found in the debris of this sumptuous villa together with the Diana. This sculpture depicts two cupids and a triton attending Venus, who wrings out her hair and gazes into a mirror that was once included as a metal attachment.43 Both the size and the moulding of the base match those of the companion statuette of Diana; countless struts connect even the tiniest details. The taste for struts that link tiny limbs or appendages is particularly evident in many well-​known pieces of late antique statuary.44 In an early fifth-​century AD marble group of Ganymedes and the Eagle found in Carthage, bulky struts connect the vertical masses, whereas tiny connectors emphasise the daring, miniature undercuts between the fingers and the eagle’s feathers.45 The tree behind the figures creates a further intricate network of branches. Both the statuettes from Saint-​Georges-​de-​Montagne and the Ganymedes are thought to originate in the same artistic milieu of Asia Minor, which seems to have been particularly vital in the period and crucial for the production and distribution of late mythological statuettes.46 These compositions, flat cabinet pieces created for a frontal point of view and probably intended for display in a niche or against a wall, share several stylistic features such as polished surfaces and numerous struts. Recent investigations suggest that, not only the exceptional Diana from Saint-​ Georges-​de-​Montagne, but also other mythological statuettes from late villas in Gaul were either imports from the Eastern Mediterranean or the work of Asian sculptors. Although it is impossible to pinpoint a specific city of 43 Paris, Louvre, acc. no. MND 2063 (Ma 3537). For the context see Stirling (2005: 30–​7). 44 See for instance a group of Satyr, Maenad, and Eros reported to have come from Aphrodisias and now in the Museum of Fine Arts in Boston (acc. no. 62.1a–​b), dated to the third or fourth century. The marble is thought to come from either the Greek islands or Asia Minor, perhaps from Göktepe. The figures, which are highly polished on the front, display a striking abundance of struts. The satyr includes an octagonal prism between the lower edge of the pedum and his right thigh with another strut a few centimetres above, between the right wrist and hip, as well as a neck strut. Two thick connectors run between the bodies of the dancing partners. Another strut probably joined the satyr’s raised right foot and the maenad’s garment. The statuette of the infant Eros was equipped with struts on the right hip and thigh, on the left calf, and on the left wrist, in addition to the neck strut. See Vermeule (1964), Comstock and Vermeule (1976: 128–​9 no. 197), Gazda (1981: 163–​6). 45 Carthage, Paleo-​Christian Museum. The group is 49 cm tall. See Gazda (1981: esp. 139–​40 Figs. 16–​18). 46 This artistic environment is reconstructed by Gazda (1981). See also Bonfante and Carter (1987: 251–​5).

195

196

196

Carving and Tradition

origin for the craftsmen who carved these pieces, stylistic features such as the extensive drilling and sensuous treatment of the facial features and the neck struts are useful indications of the source of the sculpting technique.47 These remarks apply to many other sets of late antique small-​scale sculptures, including a group of four statuettes now in Dresden, where they arrived in 1728 as part of the Chigi collection. The four pieces, probably found in Rome or its environs, all share the same size, around 90 cm, and stand on similar plinths, which are profiled only on the front. Three of them represent gods (Demeter, Apollo, and Diana) and are carved in Luna marble. The fourth, made of Parian marble, depicts a cuirassed figure whose head is now broken.48 Notwithstanding the difference in material, all four share such stylistic similarities that they can be considered to be the work of one and the same workshop that was active in Rome during the late fourth century AD. The three figures of gods closely recall the corpus of mythological statuettes found in North Africa, southwest Gaul, and Constantinople. In particular, the statuette of Diana, which loosely repeats the Rospigliosi-​L ateran type, seems to follow the same composition criteria as the roughly contemporary piece from Saint-​Georges-​de-​Montagne (Fig. 54).49 The goddess is shown hunting, with a dog and a hind at her feet and a tree spreading behind her. Her windswept drapery flares in a curve around her upper body. The stubs that remain on the right elbow and on the right thigh indicate that arched struts connected the limbs and the drapery on that side, completing the circular pattern around the figure. The proliferation of struts in late antique mythological sculptures may hint at some more or less specific association with large-​scale statuary. However, the ubiquity, number, and size of the struts, as well as their careful shaping and treatment, suggest that they were first and foremost one of the results of a widespread taste for elaborate compositions. This taste is not exclusive to small-​scale three-​dimensional figures. In a way, intricate strut-​work blurs the lines between the carving and appearance of three-​ dimensional sculpture and forms of deep-​cut relief-​work, not necessarily and not exclusively in stone. A feat of dazzling virtuosity like the Lycurgus Cup, dated to the fourth century AD, is a spectacular expression of this 47 See Stirling (2005: 10, 117–​19, 136; 1997a; 1997b). 48 Dresden, Albertiunum, inv. Hm 250 (Apollo), Hm 265 (Demeter), Hm 270 (Diana) Hm 341 (cuirassed statue). For discussion see C. Vorster (in Schröder 2009: 324–​36 nos. 61–​4; in Knoll, Vorster, and Woelk 2011: vol. I, 604–​24 nos. 138–​41; 2012–​13: 405–​15). 49 Further views are in Knoll, Vorster, and Woelk (2011: vol. I, 612–​13 Figs. 139.2–​5).

197

Developments

Figure 54  Statuette of Diana, late fourth century AD. Marble. H. 71 cm. Dresden, Albertinum, inv. Hm 270

taste.50 The most astonishing feature of this vase made of dichroic glass is that its colour changes depending on whether it is seen in reflected or in transmitted light –​from opaque green to translucent wine-​red. But the cup is also a masterpiece of cutting and grinding. The glass was adroitly cut back in order to create an elaborate frieze in high relief, culminating with King Lycurgus caught in a tangle of grapevines. The fondness for exquisitely intricate patterns has its roots in the decorative developments of Hadrianic and, especially, Antonine sculpture. It is in this era that sophisticated spiral patterns became fashionable, and they began to appear on functionally irrelevant struts. Small stone bridges in the deep drill channels increased the effects of texture and light of the volumi­ nous hairstyles. A work like the figural table support with Dionysus, Pan, and a little Satyr in marble from Docimium at the National Archaeological Museum of Athens is a highlight of this taste for polished surfaces, 50 London, British Museum, no. 1958,1202.1. A new technical study of this cup is found in Lierke (2013: esp. 96–​8). See also Elsner (2013) for the meaning, function, and context of the vase.

197

198

198

Carving and Tradition

chiaroscuro, and deep relief-​work –​with a network of branches, struts, and intertwined attributes connecting the figures while constructing a frame around them.51 The corpus of small-​scale late antique sculpture indicates that the makers and patrons in this period were particularly fond of patterns created by intricate constructions that encased the body within networks of abstract forms. In this respect, late mythological statuettes are much more explicit than their earlier counterparts about the decorative role of struts as complements to the human body, elements of visual complexity, and testaments to their carvers’ ability.

51 Athens, National Archaeological Museum, inv. no. 5706. See Kaltsas (2002: 349 no. 739), as well as the discussions of this piece by Schade (2007: 163–​5 and Pls. 8–​10) and Feuser (2013: 209–​10 no. 33).

199

Conclusion

Among the forms of Japanese traditional theatre, puppet theatre, or Bunraku, stands out for the elaborate criteria that regulate the action on stage.1 Bunraku puppets often display exquisitely carved heads and sumptuous costumes (Fig. 55). Their head, arms, and legs are detachable. Supported by a pole in its torso, the puppet is capable of performing a number of simple movements. In the most sophisticated examples, fingers and thumbs are also moveable and a network of strings allows a wide range of expressions –​opening, closing, and rolling the eyes, raising and lowering eyebrows, opening and closing the mouth. The technology of these figures is so complicated that their manipulation requires a team of three trained puppeteers. Puppeteers move in full view of the audience, clothed entirely in black and wearing gauze-​like hoods over their head. The public is invited to concentrate on the story represented on stage, on the lavish costumes, and on the emphatic gestures of the characters. The theatrical experience relies on the premise that viewers do not let the stage apparatus, although completely visible, interfere with their understanding and enjoyment of the dramatic fiction. At the same time, however, a form of representation like the Bunraku calls explicitly upon the viewers to contemplate –​and appreciate –​the fictional nature of the characters and mechanisms. For the public, it is impossible not to notice the three ‘anonymous’ men in black who operate the puppets and not to wonder at their coordinated movements and their skills in extracting nuanced expressions from the figures. The three men who function as supports to the puppets are, at once, present and absent, visible and invisible. This shift depends on whether the viewers’ attention is focused on the events represented on stage or on technical challenges. Although drawn from a different period, cultural environment, and, especially, from a different field of artistic expression, this example enucleates the same questions at stake when we try to make sense of ancient structural supports. Are the supports of Roman marble statues always mere 1 On Japanese puppetry see J. R. Brandon in Banham (1995: 139–​40, s.v. bunraku) and Foley (2016: 189–​90, with bibliography at p. 200). For the aesthetics and meanings of objects, props, and stage mechanisms in the context of Greek tragedy see the excellent examination in Mueller (2016).

199

200

Figure 55  Utagawa Kunisada I, Scene from a Bunraku theatre performance, 1856. Woodblock print, ink and colour on paper. 35.5 × 24.3 cm. Boston, Museum of Fine Arts, William Sturgis Bigelow Collection, acc. no. 11.30063

201

Looking at Supports

aids to the technical process of carving, measures for stability, or safeguards for transport? Or did these implements also belong to the broader stylistic vocabulary used to articulate messages about the nature of the artwork, its relationship to tradition, and its quality and workmanship? I will now briefly re-​examine the main points raised in the Introduction, in light of the evidence and assessment that I have offered so far.

Looking at Supports Scholars have made only cursory attempts at accounting for the function of struts, wavering between two equally generic explanations. First, they allowed the sculptor to translate a lighter bronze prototype into heavy stone, and second, they were a means to make a statue more secure for transport. Agreement in stressing the necessity of struts for the mechanics of production and trade carries the corollary of dismissing any further aesthetic concerns. Like the human supports of the Bunraku puppets, I argue that the struts included in most Roman marble statues also allow for multiple degrees of acknowledgement and understanding. Their presence in Roman statuary of all sizes indicates that struts were widely accepted as standard features of marble statuary. Yet, the lack of information from literary sources seems to imply that they were simultaneously seen and overlooked. In the rare instances where the description of a statue includes its supports –​such as Apuleius’ presentation of the Diana at the Hunt, surrounded by her dogs, displayed in the (fictional) house of Byrrhaena –​attention focuses solely on their subject and narrative content. What the ancient viewers thought of non-​figural supports or struts remains entirely a matter of speculation. Struts occur in statues of exceptionally high quality as well as those of poorer workmanship, in colossal groups and statuettes, in statues that reproduce famous types as well as common figures in Roman dress –​ togati, cuirassed statues, and draped ­figures –​and images of gods and athletes which cannot be ascribed to any specific Greek models. Although supports and connectors were already widely employed by Greek sculptors from the archaic period onwards, it is not until the Roman imperial age that they become one of the most familiar features of marble sculpture (Fig. 56). Struts appear with increasing frequency from the beginning of the first century BC, reaching a peak, in terms of size and flamboyance, during the following century. This seems to reflect the increasing availability of large blocks of stone, which allowed sculptors to choose from a variety of

201

202

Figure 56  Statue of a sacrificing woman, Severan period. Marble. H. 189 cm. The statue includes a double twisted strut. Florence, Galleria degli Uffizi, inv. 1914 no. 131

203

Looking at Supports

methods for supporting expansive poses in addition to joining individual pieces. By the time of Hadrian, when statuary marble had become a widely accessible commodity, large monolithic compositions that included non-​ figural supports and struts were the standard. While most struts were probably negligible to Roman viewers, a small proportion of them, exceeding the actual static needs of a statue, must have been intended to be acknowledged in their own right and to enhance interest in the composition. Struts that highlighted daring undercuts could advertise their carvers’ facility in working the brittle medium of marble. At the same time, the use of struts implies that the statue was carved out of a large and expensive block of marble instead of having been assembled from multiple pieces. In particular, miniature statues with oversized props, attached to tiny limbs and diminutive appendages, indicate that struts may have been more than technical devices, but rather elements that contributed to the comprehension and appreciation of the work. Viewers would notice the skilful carving of miniature undercuts between the limbs and the struts, questioning whether the statue depended on a full-​scale model with more compelling static needs. This idea implies that an artwork could elicit a variety of responses, perhaps from the same viewer at the same time. Therefore, the suggestion –​sometimes explicit in scholarly literature –​that we should view statues as if they had no struts, fails to account for a fundamental dimension of the artistic experience.2 All struts –​and many figural supports –​are irrelevant to the narrative of a sculpture and play no part in making sense of the mythological tale, identifying its characters, or connecting it to other representations of the same subject in stone or other media. They are, instead, fundamental to an understanding of the technological challenges posed by that particular composition and the solutions devised by the artist. When concentrating on the subject and its treatment, the technical equipment or any extrinsic additions would fade out of sight. For the purpose of gathering contextual information and judging a carver’s proficiency, elements such as struts, joins, and profiled bases all add up to the overall image of an artwork. The question of visibility becomes even more compelling when we consider the surface treatment of struts. The fact that struts often show a degree of finish exceeding the mere practical needs, in the shape of prisms with flat surfaces and sharp angles, indicates that they were integral to the composition as abstract supplements to the human figure. That the carving of edges Andreae and Conticello (1987: 14) suggest that, when confronted with a group like that of Scylla at Sperlonga, ‘man soll sich diese Stützen bei der Betrachtung als nicht existent denken’. 2

203

204

204

Conclusion

and the imposition of a faceted form could raise questions about nature and artificiality is in fact implicit in Pliny the Elder’s treatment of gemstones.3 There, the question of hard stones carved in the form of regular prisms highlights the irresolvable opposition between nature and ars that is central to the ideological structure of Pliny’s Natural History. In the case of struts, forms, finishes, and decorative details, especially groove motifs, were chosen for expressive reasons and adapted to the individual piece. One cannot fail to notice that there is an inherent ambiguity about this material. We can question, in fact, whether the spiral pattern embellishes a strut that is impossible to conceal, or if it is rather a strategy to draw attention to an intentionally visible element. Other technical features of ancient statuary share the same fundamental ambiguity. In his landmark study of drill-​work in Roman sculptural production, Michael Pfanner accounts for the extensive use of this technique during the imperial period as a manifestation of either a high level of craftsmanship or a lack of skill in putting on the finishing touches with other instruments.4 Drill-​work, just like struts, could be used in different contexts and with different purposes by craftsmen of different origin, tradition, and levels of competence. It is to be hoped that the addition of further pieces to the corpus of statues with decorated struts, as well as broader surveys on polychromy in ancient Roman sculpture, will provide more detailed information about the role and impact of carved and painted patterns. A better understanding of struts and their chronological and geographical distribution is also likely to keep pace with the progress of scientific research about the provenance of white marbles. Apparent concentrations derived from the find contexts and other stylistic features could be confirmed or clarified, providing clues as to the origins of the workshops. Yet other conflicts and difficulties are likely to remain unsolved. Inevitably, our attempts to make sense of the ancient visual practices are in some aspects severely limited. In particular, an entire dimension of the display context of ancient sculpture is almost entirely lost to us, that of ephemeral implements and the relationship with the natural world. As I have argued already, for instance, a statue such as the Pouring Satyr from Palermo could well have been equipped with a changeable metal 3 In a passage about rock crystal (37.26), Pliny remarks that this material is extracted in the form of hexagonal prisms, with uneven angles and polished faces, in such a fashion that ‘art is unable to match’ (ut nulla id arte possit aequari). In 37.76 he explains that beryls are shaped into smooth hexagonal prisms by ingenious carvers ‘since their colour … is enhanced by the reflection from the facets’. See also Natural History 37.79 and 37.137. 4 See Pfanner (1988: esp. 672).

205

The Limits of Function

vessel, substituted from time to time. We will never be able to demonstrate whether the original setting included some interaction with the vegetation of a garden (e.g. the strut could have been wrapped by a climbing vine, up to the cup) or other furnishings (e.g. hanging drapes).5 The fact that in the Palermo Satyr this strut is not required by static constraints eliminates the possibility that an interplay between the marble figure and its surroundings would be required to conceal the oversized support. On the contrary, it is possible that the statue was equipped with a strut so as to allow for such an interaction in the first place. While the available data from the corpus of Roman sculpture are at best fragmentary and ambiguous, often lacking information on the original display context, the silence of ancient written sources prevents us from fully reconciling the archaeological evidence with its conceptual universe. Like any other element of a representation, struts also necessarily provoke some sort of response from their viewers  –​a more or less conscious engagement with the subject, underpinned by collective cultural references as well as influenced by personal taste. Scholarship in the field of ancient art has extensively discussed the difficulty of pinning down the meaning of such an elusive concept as that of the ‘Roman viewer’.6 The same object or monument could trigger different responses that were contingent upon factors such as gender, class, ethnicity, and education. Since no model viewer exists, we can only refer in general to a set of broadly defined attitudes and expectations common to the inhabitants of the Roman world between the first century BC and the end of the empire. Thinking in terms of a shared repository of knowledge about what art should be and how one should respond to it reduces the synthetic distinction between artists and patrons, both of whom also belonged to the society of viewers. At the same time, we cannot fail to recognise the inherent artificiality of this construct.

The Limits of Function The considerations of viewing practices and taste that I have offered here should not overshadow the functional nature of supports and struts. Most struts primarily served to stabilise a work in marble. They provided a tool For a repertoire of images of statues in Pompeian paintings, which often depict statuary as part of lavish garden displays, see the fundamental volume by Moormann (1988). That statues could be dressed or equipped with drapery and other items or clothing is a well-​known practice in the sphere of cult images, as summarised by Romano (1988: 130–​3) and Locchi (2006: 91–​4). 6 See the comments by Elsner (1995), Zanker (1997), and Rutledge (2012: 79–​121). 5

205

206

206

Conclusion

to reconcile material and form in an array of compositions, well beyond the boundaries of the epoch that we usually describe as ‘Roman’. The purpose of my investigation is not to negate the essential utility of such supports, but rather to explore whether this holds true in all cases and whether the evolution of widespread workshop practices, such as the use of struts to brace and sustain marble figures, might have influenced the ways in which sculpture was seen and understood. The task of translating a bronze prototype into a heavier and more brittle material such as marble certainly requires special precautions. These may include supports that increase the bulk of the lower half of the figure, structural elements that fasten projections, or joins that connect the outstretched limbs to the core. The fact, however, that the size and position of struts sometimes exceeds any practical needs and that they are not exclusive to sculptural copies after bronze prototypes, indicates that this explanation is insufficient. Other avenues of investigation may account for this body of material. One possibility is to examine in closer detail the stages and techniques for the production of a marble statue, regardless of its relationship to any prototypes. There are, in fact, two moments in the production process of a marble statue when struts seem to represent a definite advantage and a useful tool for a sculptor. First, while a statue is being carved, struts can protect a figure against the strokes of the tools. Secondly, they provide additional security to the composition once in place. These advantages apply equally to the replicas of bronze originals and to compositions initiated in marble. However, a number of statues include struts of a size and design that contradict these remarks. Carefully shaped as faceted rods, which create large undercuts between slender forms, struts sometimes represent a challenge to the sculptor. The percussive strokes of the sculptor’s pick and chisels used to rough out and shape a long strut, attached to isolated limbs, add considerable strain to the figure as a whole and to its outstretched extremities. That struts were required for the transport of marbles over long distances and were intended to be removed later is also only partially true. Most struts do not indicate that a work was left unfinished, but they appear on statues that had been worked to completion, with finely carved details and finished surfaces. Obviously, there would be no point in a sculptor neatly shaping a quadrilateral bar or an arched connector if he expected it to be removed once the sculpture had been set in place. In addition, struts do not appear to ensure that a statue would survive transport intact. Comparable or increased safety could be granted by shipping a statue unfinished or as

207

The Limits of Function

separate pieces, to be assembled upon arrival. The method of carving the outstretched limbs as separate pieces presented other practical and financial advantages besides easier shipping. It reduced, in fact, the time and labour, while also allowing substantial savings in raw material. In sum, there is no conclusive evidence that struts were always necessary –​or even useful –​for the purposes of carving, shipping, and installing a statue. Rather, the impression is that sculptors negotiated case-​by-​case the advantages and shortcomings of their technical options, according to many factors, including the quality of marble, the intended destination and context, as well as their own competencies and their workshop’s tradition. The fact that, in a small number of cases, struts were carefully replaced when the statue was restored and its static infrastructure strengthened, allows further insight in the apparent tension between utility and visual practices.7 For instance, the elaborate ancient restorations of an over-​life-​ size himation figure from the Bouleuterion at Aphrodisias, later inscribed as Demos (the personification of the citizenry), include both limbs and struts.8 This figure, carved in the late second century AD from a single block of marble, was depicted in the act of pouring a libation, with the right arm extended to hold a phiale. Four ancient repairs were carried out, probably at the same time the inscription was added to the plinth during the fourth century AD. The front of the plinth was restored, as well as the right forearm and hand at three different points. The forearm was reattached and securely fastened by means of a dowel that ran all the way through the forearm into the right hand. A small pin-​hole served to reattach the broken right thumb. Surprisingly, the large strut that formerly extended from the upper right thigh to the right wrist was also replaced –​or, more likely, reattached –​using a large iron dowel.9 From a technological point of view, reintegrating the long strut from thigh to wrist is superfluous. Once a dowel The ancient restoration of the Castel Porziano Discobolus seem to have repaired or replaced the raised right arm and the bulky strut from arm to hip. See Rizzo (1907: 4) on the holes for a dowel carved on the arm immediately under the shoulder, as well as on the upper edge of the broken strut. Examples of restored struts partially contradict Hollinshead’s remark that in no instance were struts attached to a statue as separate pieces (2002a: 143). 8 Aphrodisias, Museum, inv. nos. 62-​451 (left foot and plinth), 62-​452 (right foot and plinth), 62-​ 486 (body), 63-​19 (hand). See J. Lenaghan and C. C. Hallett (in Smith 2006: 162–​4 no. 44 and Pl. 34). The right forearm was fixed with a cylindrical dowel (diam. 1.5 cm, D. 11 cm). A pour hole (diam. 1.5 cm) drilled down through the forearm intersects with the end of the dowel hole. The small pin hole on the right hand that served for the reattachment of the thumb has a diameter of 0.5 cm. The hole for the iron dowel that served to reattach the strut has a diameter of 2.5 × 2.5 cm. 9 A rectangular cutting on the back of the right hand was probably intended to receive the other end of the restored strut. A second strut, still intact in its original form, extends from the torso to the right arm slightly below the elbow. 7

207

208

208

Conclusion

had been inserted through the entire length of the forearm, the statue of the Demos did not require any further support. The rationale behind this choice becomes clearer within the framework of Aphrodisian marble sculpture. All of the other statues excavated in the Bouleuterion stage building show traces of extensive repairs, which followed consistent criteria. The broken parts, however minute, were painstakingly reattached (or replaced with new pieces carved ad hoc). Clearly, particular attention was paid to the preservation of the sculptural display in its pristine appearance, with an eye to restoring the statues in all their distinctive features, including their technical implements. In a broader perspective, this example demonstrates that visibility and the ‘distracting’ effects of struts on the sculptural programme were not necessarily a concern for ancient viewers.

Visible Signifiers The reworking carried out on the Demos implies that the contribution of the struts to the appearance and status of a statue was acknowledged and recognised. Clearly, such a contribution would not affect the iconography and narratives that unfold from it. Rather, struts belong to the set of accessory devices –​labels, bases, and frames –​that allow a statue to occupy a certain space within the built environment, thus guiding its viewing and understanding. In restoring a statue’s pristine appearance, struts would be seen as intrinsic elements of the figure. The fact that features such as bases, architectural frames, and supports appear on two-​dimensional representations of statues in reliefs, paintings, and on coins demonstrates that they all refer to a shared conception of what a statue should typically look like. However, while figural supports are not infrequent in two-​dimensional images of statues, only in very few cases does the artist seem to have relied on struts to characterise a statue as such. One such example is perhaps the discus of a mould-​made lamp said to have come from Cumae, assigned to the first half of the first century AD (Fig. 57).10 The scene represents two characters from an unidentified mythological scene, facing each other on a thin base line. A young man, with a cloak or an animal skin on his shoulders and wearing a sword, moves to the right. A woman moves to the left towards him. She wears her drapery loose around her waist and wrapped over her left arm. She leans forward 10 British Museum, reg no. 1856, 12–​26.552.a. For comments on this lamp and the depiction of struts see Bailey (1980: 203 no. Q996 and Pl. 27) and Stewart (2003: 93 n. 73).

209

Figure 57  Discus of a lamp, first half of the first century AD. L. 20.7 cm, W. 9.8 cm. London, British Museum, Reg. 1856,12–​26.552.a

210

210

Conclusion

on a staff or spear, which passes under her right armpit. A  snake coils around the end of the staff. Two straight ridges extend from the body of the female figure. One of them stretches between the staff and the woman’s right knee, and the other links her left shin with the base line. The most reasonable explanation is that these ridges represent bulky, quadrangular struts that underpin the detached and off-​balance parts of a sculpture. In consequence, we may suppose that the craftsman who produced the mould either drew this design from an existing three-​dimensional prototype or wished to establish some connection to large-​scale statuary. The sarcophagus of a child at the Capitoline Museum, previously in the Villa Doria Pamphilj and in the Albani collection, seems more straightforward in the use of struts to construct the image of a statue.11 Dated to the final years of the third or the beginning of the fourth century AD, the sarcophagus chest combines mythological and biblical scenes. The creation of mankind at the hand of Prometheus takes pride of place in the centre of the scene (Fig. 58). Prometheus sits with his creation, a little clay figure, on his lap. Obviously, this is a work in progress or one that has just been concluded. A basket with clumps of clay is placed on the ground behind Prometheus, and he still holds the modelling rod. Drawing back his hand from the diminutive figure, he seems to judge the quality and effect of his craftsmanship. Standing before him, Athena, identified by her customary helmet and aegis, places a butterfly on the head of the clay figure, symbolising her gift of the life-​giving soul. Between the two creators, side by side with the little moulded man, stands another figure of identical size. This is fully frontal, placed on a tall pedestal. Both hands are fastened to the body by struts, carefully carved out of the background. Only the base and the struts distinguish the two small-​scale figures, one of which is a statue (with base and struts), the other a soon-​to-​be man (standing free on his feet and ready to open his arms). From a narrative point of view, of course, their relationship remains ambiguous  –​is the shape of the first human being derived from a pre-​existing iconic image? Visualising ‘artificiality’ appears to be, understandably, one of the main preoccupations in the few Roman sarcophagi with the myth of Prometheus. Often, the lifelessness of the human simulacrum created by Prometheus is conveyed by the figure’s small size or by placing the figure on a plinth, like a statue.12 Elsewhere, the 11 Rome, Capitoline Museum, Sala delle Colombe, inv. 329. See Stuart Jones (1912: 142–​3 no. 13 and Pl. 34), B. Andreae (in Helbig4, vol. II, 109–​11 no. 1257), Sichtermann and Koch (1975: 63–​4 no. 68 and Pls. 165–​7), Koch and Sichtermann (1982: 183–​4 and Fig. 215). On the struts of the model statue, see also Stewart (2003: 93 n. 73). 12 E.g. J.-​R. Gisler in LIMC (vol. VII.1: 533, 545–​6 nos. 1, 101, 104).

newgenrtpdf

211

Figure 58  Sarcophagus chest with the myth of Prometheus, late third or early fourth century AD. L. 117.5 cm, H. 66 cm, D. of side 43.5 cm. Detail. Rome, Capitoline Museums, Sala delle Colombe, inv. 329

212

212

Conclusion

clay figure lies stiff on the ground, still immersed in its clump of modelling material.13 References to both statuary stillness and materiality define the figure as a man-​made image, void of life and unable to move. Like plinths and lumps of raw material, struts were an immediate sign of artificiality and, as such, could be employed in a visual narrative. The conventionality of this language is clear in the Capitoline sarcophagus. As characteristic of marble sculptural technique, struts help to distinguish the two artificial ­figures –​one of marble, forever motionless, the other in clay, awaiting life. Struts allow form to transcend the limits of the medium, expanding the range of designs available to the sculptors in marble. In this sense, struts are a powerful reminder that the artwork is made of a given material, in this case stone. Struts demonstrate a work’s material circumstances and technologies in a much more direct way than is possible in other figural arts. It is precisely because of their fundamental technical function –​as reinforcements to the body  –​that struts can become meaningful decorative additions in their own right. In responding to technical challenges, artists could devise designs that also pursued expressive objectives. The v­ isual potential of struts is occasionally exploited and stretched to its limits, regardless of actual functional utility. Thus, struts were sometimes decorated with carved motifs in order to either assimilate or distinguish them from the adjoining surfaces. Marble shafts and rods could be aligned or juxtaposed to the limbs and attributes, enclosing and framing the figure. In all these cases, struts are integral to the composition and function as abstract complements to the human figure. In addition to this, using sturdy supports that fastened the outstretched limbs to the body in compositions created ex uno lapide required a great deal of extra marble, sometimes –​as in the case of the Vatican Apoxyomenos –​almost twice the amount needed for a more contained human figure. The type and size of supports also provides precious information concerning the Roman practice of imitation and seriality –​the retrospective attitude and eclectic artistic taste that scholars have described as a distinc­ tive ‘ethos of emulation’. The presence and variety of supports among the items of well-​known replica series has generally been considered as evidence of a bronze prototype, which naturally needed no such additions. Surely, scholars have argued, struts would improve a heavy statue’s chances of surviving transport and the passage of time intact. It is just as indisputable that struts would be very useful to any sculptor attempting a reproduction in stone of a figure conceived in a lighter material. Nonetheless, the 13 E.g. J.-​R. Gisler in LIMC (vol. VII.1: 547 no. 110). See also Robert (1919: 447–​9 no. 357).

213

Visible Signifiers

variety of supports within a given series also seems to bear more complex implications. By emphasising emphatic or ‘extravagant’ poses conceived in bronze, struts also emphasise the sculptor’s adherence to his model, which was not adapted to the heavier and more brittle medium of marble. If supports are to be investigated as technical measures that are relevant for broader art-​historical issues, our theoretical framework should allow for a comprehensive account of social agency mediated by visual images. This point will be clarified by one further case study, which has been the object of closer scrutiny at various points in this book, the Pouring Satyr at the Museum Salinas in Palermo (Fig. 59). Unlike any other statue in this replica series, this satyr has a long and finely polished bar of stone connecting the youth’s extended left hand to his thigh. Whereas all other known copies include a figural support in the form of a tree stump or a pillar on the satyr’s left side, this fine first-​century AD statue alone has its support on the right. The smooth tree stump, which is already bulkier than usual, seems to be relocated on the right side to save space for the prominent strut. The angle of this strut runs parallel to the figure’s lower right leg and the tree trunk support. From a tectonic perspective, such a conspicuous strut, larger than the satyr’s arms, is unnecessary. All other copies demonstrate that the same composition could be just as conveniently arranged either without struts at all or with a much shorter bar between arm and body. The high tree trunk on the right side of the satyr, placed at an unusual distance from his body, is equally exceptional. Here, figural and non-​figural supports seem to have been devised as one of the most prominent features and main attractions of the statue. In a statue such as the Palermo satyr, the problems of mimesis and the physical limits of the medium appear to be two intrinsically conflicting elements. We may try to reconcile these features by pinpointing the multiple agencies at work –​patron, sculptor, and audience –​and the types of response that the statue was intended to generate. One approach might be to cast the main actors involved in the making and viewing of the Palermo satyr –​the artist, the artwork, and the audience –​in the conceptual tools developed by Alfred Gell to account for agency in art.14 Gell’s semiotics of art revolves around the basic concept of the index, a sign from which one can make causal inferences ‘about the intention or capabilities of another 14 For the passages quoted see Gell (1998: 13, 27). According to Gell’s construction, which relies on the terminology of Peircean semiotics, art objects act as extensions (indexes) of their makers’ or users’ agency by virtue of being deployed in social relationships. See Layton (2005: 34), Tanner and Osborne (2007), as well as the comments, applied to the field of Graeco-​Roman visual culture, by Pucci (2008), Bussels (2012: 18–​19), Ma (2013: 297–​306).

213

214

Figure 59  Replica of the Pouring Satyr in Palermo. Left-​hand view

215

Visible Signifiers

person’. The Pouring Satyr, in this sense, is a material indication of a complex negotiation of agencies. In this model, three aspects contribute to the formation of the index:  the artist to whom ‘causal responsibility for the existence and characteristics of the index’ is ascribed (i.e. the sculptor who decided to append conspicuous supports to his Pouring Satyr); the recipients or ‘those in relation to whom … indexes are considered to exert agency, or who exert agency via the index’ (i.e. those who commissioned, purchased, displayed, and viewed the marble satyr); and the prototypes, the ‘entities held … to be represented in the index, often by virtue of visual resemblance’ (i.e. the putative original, in this case a bronze satyr pais by Praxiteles). Of course, considerations about the normativity of a prototype or the constraints of genre and tradition increase the degree of ambiguity in the relationship between the artist and the prototype and their mutual agency. When illustrating the Stoic, Aristotelian, and Platonic principles of causality by means of an example drawn from the sphere of art, Seneca employs similar categories.15 The entities that cause a given statue (i.e. the index) to exist as such are its material (materia, e.g. marble vs. bronze), the artist (artifex, opifex), the form (forma, e.g., the facies of a given satyr by Praxiteles), the purpose (faciendi propositum), and the model (exemplar, a term that translates Plato’s concept of idea). A work of art is therefore the result of five factors: that from which (id ex quo, the material), that by which (id a quo, the maker), that in which (id in quo, the form), that because of which (id propter quod, the purpose), that in reference to which (id ad quod, the model). These factors combine to create the statue itself (id quod ex his est, what we may call the index). All these entities variously enact their agency and, in doing so, affect further agents’ actions. The Roman sculptor, influenced by his late classical model, shaped the index, a marble version of a famous Pouring Satyr type. But what induced the artist to modify his prototype the way he did? The Pouring Satyr from Palermo exercises agency with the recipient who is encouraged to make inferences on the statue’s merits, its author’s skills, and the patron’s intentions by means of its individual features. Just like the other details of the statue, its bulky supports inspire the observer’s judgement and imagination. A further element to be considered in this anatomy of our statue’s agency is the position and role of other copies. The individual statue was meant to take its place among a world –​both physical and mental –​of other versions of the same 15 Moral Letters to Lucilius 65.5 (for a commentary to the passage: Seneca. Selected Philosophical Letters, tr. B. Inwood. Oxford: Oxford University Press, 2007: 139–​42). See also Stewart (2003: 238).

215

216

216

Conclusion

composition in a variety of public and private spaces. The need to differentiate and distinguish the single piece from many similar ones may, in fact, have influenced the arrangement and treatment of the body and its additions. In a statue like the Pouring Satyr from Palermo, struts help to channel this intention.

Seeing Supports The reconstruction of agencies and the content associated with technical features such as struts also allows us to tackle the thorny issue of realism or verisimilitude. Unlike the muscles, lineaments, and hair of the pouring boy, rendered in finest detail, the conspicuous supports contrast with the pretension to a resemblance to nature. Set between huge, rectilinear props, the boy’s graceful movement and his convivial act of pouring wine are locked in a frame that runs parallel to the body. This seems to contrast with the paramount interest in naturalism and the faithful reproduction of nature that is found in the Roman written sources about art. Like the ars of the writer and orator, the skills of painters and sculptors were essentially linked to the concept of naturalism.16 The quality of ‘being likely to be natural’ or, in other terms, the impression ‘of being not man-​made’ is far and away the most remarked upon quality in Pliny the Elder and many other authors. This type of criticism is often reflected in the many anecdotes about the tricks or surprising results caused by a piece’s exceptional realism, which succeeds in temporarily suspending the boundaries between animate and inanimate. Our difficulties in understanding the supports of Roman marble statuary depend mainly on the conflict between the expectations about ancient naturalism and the effect of artificiality conveyed by these implements. Even figural supports, that purportedly reproduce the shape and bark of a tree trunk, are often baffling in their artificial juxtaposition to the human body –​recall the regular, linear grooves around the tree trunk support of the famous Doryphorus from Pompeii, which abruptly ends against the The bibliography on the notions of ‘realism’ and ‘naturalism’ in the Hellenistic and Roman art criticism is understandably large. Among the most recent contributions see the work by Rutledge (2012: 93–​102) and the examples presented by Isager (1991: esp. 136–​9). On the terms ‘realism’ and ‘naturalism’ and their implications, see Y. Chevrel, s.v. naturalistisch in ÄGB (vol. IV, 404–​32) and W. Klein, s.v. Realismus/​realistisch (vol. V, 149–​97), as well as R. Bianchi Bandinelli, s.v. realism in EAA (suppl. 1970: 656–​7). See also Hardiman (2012) on the virtue of ‘realism’ in the ancient non-​specialised criticism, as well as the sources collected in Rusnak (2001). 16

217

Seeing Supports

athlete’s thigh. Although made in a different context, in reference to friezes such as that of the Column of Trajan, Peter von Blanckenhagen’s description of Roman narrative art as ‘a curious combination of rich and exact realism in detail and complete lack of realism in composition’ would seem to apply also to the effect of struts, as well as to the many figural supports whose shape is only apparently naturalistic.17 However, if we return to the notion of struts as implements consciously employed to stress the statuesque, i.e. the man-​made nature of a statue, we find that this strategy draws on the same principles as other methods used to stress the artificiality of artworks. As Peter Stewart has argued, together with iconographic elements such as bases or architectural niches, stylistic features could also contribute to characterise statues as such in two-​­dimensional depictions. Elements such as archaism, frontal or profile representation are markers of rigidity and lifelessness that viewers would intuitively perceive as inconsistent with a living creature. Struts can function along the same premises. Large struts both call attention to a figure’s movement and the challenges that it posed to the marble carver and convey the impression of suspended movement, only conceivable in a lifeless art-​ object. Clearly, these remarks bear far deeper implications in the case of the replicas or variants of some ancient and widely known prototypes. Struts demand two types of attention for a work of art. These are defined by Michael Polanyi as ‘focal awareness’ and ‘subsidiary awareness’, and they stand in fundamental opposition to each other.18 The former concentrates on the object of representation in itself, in our case the body that the statue represents –​or, to return to the example that opened this section, the story and characters of a Bunraku play. A ‘subsidiary awareness’ of the technological dimension of the mimesis, however, is constantly retained by the viewer and can be summoned anytime. The struts of Roman marble statues –​ like Bunraku puppeteers –​are not entirely invisible, but only ­temporarily removed from perception. Statues combining a sense of momentous tension and a large strut such as, for instance, some of the versions of the Discobolus, are at once lifelike because of naturalism at work, and highly 17 See von Blanckenhagen (1957: 81). This passage was very appropriately quoted by Andrew Stewart (1977b: 80) to explain the ‘unrealistic’ effect of the sculptural groups at Sperlonga. 18 The notions of ‘focal awareness’ and ‘subsidiary awareness’, introduced by Polanyi in two articles from 1970, are perceptively discussed by L. Giuliani (2013b: esp. 346) with regard to Michelangelo’s statue of Lorenzo de Medici in the Sagrestia Nuova in Florence. Giuliani argues that the squared object under Lorenzo’s elbow is a rough block of marble, a subtle reference to the image’s material and technology. According to Giuliani (2016), the appearance of a similar block as an attribute of Sculpture in Michelangelo’s tomb in Santa Croce indicates that this reference was understood and acknowledged by contemporary viewers.

217

218

218

Conclusion

unrealistic because of the explicit nature of their assemblage as works of art. Although referring specifically to pictorial representations, Ernst Gombrich provided an analysis of the experience by which we are aware of the surface of the picture and of its absent object, arguing that we switch between two moments or modes –​first experiencing one and then the other.19 Pictures, in other words, depict by relying on ambiguities of visual discrimination. Richard Wollheim argued that such switching does not entirely encompass the phenomenology of perception, which rather rests on a ‘twofoldness’ by which we see both the artwork with its material and technical features, and the person or object represented by means of those features.20 The viewer is called to ‘see the object, or what is represented, in the medium of representation’, being simultaneously aware of the features that sustain this perception. While seeing the subject represented, the viewer recognises that the technical and material features of the artwork are not to be attributed to it. The experiences of the scene depicted and the pictorial design are mutually dependent. The experience of an artwork, in other words, does not coincide with the phenomenology of seeing the actual subject, but depends on the distinctive way in which that material is organised. The potential of suspended movement for the purposes of clarifying an image and indicating its artificial nature was skilfully exploited in the learned discourse about theatrical performances. A passage from Plutarch’s Sympotic Questions, where the elements of dance are singled out and discussed, is explicit in indicating ‘frozen’ poses as fundamental tools for dramatic representations, easily understood by the public as typical of a man-​made image.21 Dancing, according to the definition provided by Plutarch, consists of movement (kinesis) and position (schesis), separated by intervals at the culminating points of the movement. In particular, movements are called ‘phrases’ (phora), whereas ‘poses’ (schemata) is the name given to ‘the representational positions to which the movements lead and in which they end, as when dancers compose their bodies in the attitude (schema) of Apollo or Pan or a Bacchant, and then retain that aspect like figures in a picture (graphikos)’. This account suggests that the pantomime’s ‘stops’ had both a representational function, as related to the story-​telling on stage, and a rhythmic function, punctuating the flow of movements in 19 For the famous rabbit/​duck illusion, see Gombrich (1960: esp. 5–​6). 20 Wollheim (1980: 137–​51). An overview of these questions is in Hopkins (1998: 18–​20, 37–​8), Lopes (2005: 28–​45), Irvin (2013: esp. 608), Matravers (2013: esp. 203–​4). On sculpture in particular, see also Hopkins (2003). 21 Plutarch, Sympotic Questions 9.15 = Moralia 747c (tr. F. H. Sandbach, Loeb Classical Library). On this passage, see Webb (2008: 51) and, for the concept of schema, Catoni (2005: 211–​13 = 2008: 192–​3).

219

Seeing Supports

the dance.22 Plutarch’s image of the bodies of dancers arranged in recognisable poses finds an immediate parallel in Lucian’s famous description of Myron’s Discobolus (or rather, of a copy after Myron’s Discobolus) from The Lover of Lies.23 In a widely cited passage from this fictional dialogue, Lucian’s alter-​ego Tychiades guides the reader in the understanding of the statue’s iconography. Tychiades’ presentation is based on the same dramatic succession of movement and pauses to both assess the identity of the figure and praise its mimetic qualities. The figure, known with the title of Discobolous (diskobolos), is a discus thrower (diskeuon) bent over in the pose (schema) of the throw. Tychiades then explains the details of this pose and concludes with its exceptional effect –​the impression that the athlete had been brought to life and is about to ‘spring up all at once with the cast’ (Fig. 60). The text invites two different and successive approaches. These require the viewer to alternately concentrate on the pose itself and on the impression of life that it elicits. The similarities between this description of an artwork and the language of pantomime dance is not surprising if we think about Lucian’s keen interest in this form of art, which he passionately defended in the eccentric treatise known under the title De Saltatione. Although these texts are not directly related to the evidence provided by struts and supports, the attitudes, expectations, and viewing modes that they imply are representative of broader cultural practices. As the only feature in a sculpture where signifier and signified coincide (marble), struts are a powerful reminder of the material and the technology required to extract the figure from the block. Struts are ‘performative traces’ of the carving process that, by diverting attention from the subject and style, convey it through material –​the ‘degree zero’ of a statue.24 In their obvious intima­ tion of materiality, struts remind viewers that, regardless of the success of pictorial mimesis, what they are looking at is a block of stone. And marble, in Roman imagination, is the quintessential symbol of lifelessness –​a 22 A similar account is provided by Libanius in his Oration 64.118. He distinguishes three stages. First, there is the ‘continuity and number of their [i.e. the dancers’] whirling’, second ‘the suddenly rigid position (stasis)’, and third ‘the posture (typos) maintained in this position (stasis)’. The dancers conclude every movement ‘in a motionless position as if they are fixed with glue’ and it is in this position that ‘the picture (eikon) presents itself ’ (tr. Molloy 1996: 175). It seems relevant here to observe that typos in Greek has a large set of meanings, that span from ‘form’, ‘shape’, and ‘model’ to ‘figure worked in relief ’ and ‘statue’, as discussed by Anguissola (2012: 90–​1). 23 Lover of Lies 18. On this passage, see Anguissola (2012: 129–​31, 154). For considerations on the movement of the Discobolus, see Giuliani and Catoni (2016: esp. 13–​15, 32–​35). 24 I borrow the concept of ‘performative traces’ from a paper (‘Periferie, epitomi, residui: strategie dell’attenzione’) delivered by Salvatore Settis at the Kunsthistorisches Institut of Florence, as part of the symposium Detail und Aufmerksamkeit in honour of Alessandro Nova (May 2014).

219

220

Figure 60  Replica of the Discobolus from Castel Porziano. Rear view

221

Seeing Supports

medium that, unlike metal alloys, remains unalterable. Ovid, exceptionally talented in turning visual stimuli into textual images, extensively exploits the referential potential of marble in creating the vivid tableaux in his Metamorphoses. There, whenever dramatic actions are ‘frozen’ in a motionless, speechless paralysis, he invokes an image created in stone. For example, Narcissus entranced by his own image, is spellbound ‘like a statue carved from Parian marble’.25 This book attempts to provide an overview of a very diverse phenomenon. Struts had, of course, a general structural utility. They are found at points of potential weakness and increase a figure’s stability. Nonetheless, in a number of cases their placement and size require a more complex explanation. Struts could also function as abstract complements and framing devices that emphasise the figure’s pose. In this sense, struts number among the techniques that an artist may exploit to establish the artificial nature of his figure. Immediately recognisable as extrinsic elements, irrelevant to the narrative, struts belong to the visual tools that constitute the vocabulary of the statuesque. In addition to this, struts could advertise the value of a statue by pointing to both the size of the block and the artist’s facility in working this challenging medium. Struts play an integral part in our experience of marble sculpture. Yet their subordinate status as functional instruments has resulted in their neglect as an object of scholarly examination. The lack of a single explanation for the set of questions, ideas, and interpretive problems that I have illustrated in these pages should not discourage us from further exploration of this subject. On the contrary, they invite a more developed and nuanced reading of supports as well as other relevant stylistic features when questioning provenance, chronology, authorship, and, most significantly, the production, selection, exhibition, and viewing of sculptures in the Roman world.

25 Ovid, Metamorphoses 3.418–​19. For metamorphoses into marble statues, see also 2.830–​2, 5.177–​235, 7.790–​1, 11.58–​60. The potential of marble to signify artificiality and the absence of life is fully exploited in an episode from book 11 (11.401–​6), about a wolf that has been transformed in a marble statue. Nothing has changed about the appearance of the animal, but for the colour. The white surface of the stone leaves no doubt that the beast has turned into an inanimate object, finally inoffensive. On the semantics of marble in Ovid’s Metamorphoses see also Isager (1991: 177–​8).

221

222

References

222

Abbe, M. B. 2015. ‘Polychromy’, in Friedland, Grunow Sobocinski, and Gazda 2015: 173–​88. Abbondanza, L. 2001. ‘Immagini della phantasia. I quadri di Filostrato maior tra pittura e scultura’, Römische Mitteilungen 108: 111–​34. Adam, J.-​P. and Varène, P. 1980. ‘Une peinture romaine représentant une scène de chantier’, Revue Archéologique: 212–​38. Adam, S. 1966. The Technique of Greek Sculpture in the Archaic and Classical Periods. London:  Thames & Hudson (British School of Archaeology at Athens, suppl. vol. 3). Addison, J. 1705. Remarks on Several Parts of Italy in the Years 1701, 1702, 1703. London: J. Tonson Albizzati, C. 1916. ‘Two Ivory Fragments of a Statue of Athena’, Journal of Hellenic Studies 36: 373–​402. Amelung, W. 1903–​8. Die Sculpturen des Vaticanischen Museums, 2  vols. Berlin: Reimer. Andreae, B. 1974. ‘Die römischen Repliken der mythologischen Skulpturengruppen von Sperlonga’, in B. Conticello and B. Andreae, Die Skulpturen von Sperlonga. Antike Plastik 14: 63–​110.  1982. Odysseus. Archäologie des europäischen Menschenbildes. Frankfurt am Main: Societäts-​Verlag.   1983. ‘Le sculture’, in F. Zevi (ed.), Baia. Il ninfeo imperiale sommerso di Punta Epitaffio. Benevento: Banca Sannitica, 49–​66.  1987. Plinius und der Laokoon. Mainz am Rhein: P. von Zabern.   1991. ‘Il ninfeo di Punta dell’Epitaffio a Baia’, in S. Stucchi and M. Bonanno Aravantinos (eds), Giornate di studio in onore di Achille Adriani. Rome: ‘L’Erma’ di Bretschneider, 237–​65.   1992. ‘Die römischen Kopien in Marmor nach griechischen Meisterwerken in Bronze als Ausdruck der römischen Kultur’, Studi italiani di filologia classica, 3rd series, 10: 21–​31.  1994. Praetorium Speluncae. Tiberius und Ovid in Sperlonga. Mainz am Rhein and Stuttgart:  Akademie der Wissenschaften und Literatur and Steiner (= Praetorium Speluncae. L’antro di Tiberio a Sperlonga ed Ovidio. Soveria Mannelli: Rubbettino, 1995).   (ed.) 1995. Bildkatalog der Skulpturen des Vatikanischen Museums. I.  Museo Chiaramonti, vol. II. Berlin: De Gruyter.

223

References  1996. Der Farnesische Stier. Schicksale eines Meisterwerkes der pergamenischen Bildhauer Apollonios und Tauriskos von Tralleis. Freiburg im Breisgau: Rombach.  1999. Odysseus. Mythos und Erinnerung. Mainz am Rhein: P. von Zabern.  2001. Skulptur des Hellenismus. Munich: Hirmer.   2003. ‘Il gruppo di Scilla di Sperlonga ricomposto’, in G. Fiorentini, M. Caltabiano, and A. Calderone (eds), Archeologia del Mediterraneo. Studi in onore di Ernesto de Miro. Rome: ‘L’Erma’ di Bretschneider, 51–​60. Andreae, B. and Conticello, B. 1987. Skylla und Charybdis. Zur Skylla-​Gruppe von Sperlonga. Mainz am Rhein and Stuttgart: Akademie der Wissenschaften und Literatur and Steiner. Andreae, B. and Parisi Presicce, C. (eds) 1996. Ulisse. Il mito e la memoria. Catalogue of the exhibition. Rome: Progetti Museali Editore. Anguissola, A. 2005. ‘Roman Copies of Myron’s Discobolus’, Journal of Roman Archaeology 18: 317–​35.  2012. ‘Difficillima imitatio’. Immagine e lessico delle copie tra Grecia e Roma. Rome: ‘L’Erma’ di Bretschneider.   2013. ‘Marks of Imitation or Signs of Originality? An Approach to Structural Supports in Roman Marble Statuary’, in A. Bokern, M. Bolder-​ B oos, S. Krmnicek, D. Maschek, and S. Page (eds), TRAC 2012. Proceedings of the XXII Theoretical Roman Archaeology Conference. Oxford:  Oxbow Books, 1–​20.   2014a. ‘Remembering with Greek Masterpieces: Observations on Memory and Roman Copies’, in K. Galinsky (ed.), Memoria Romana. Memory in Rome and Rome in Memory. Ann Arbor, MI: University of Michigan Press, 117–​34.   2014b. ‘The Prestige of Copies. Greek Originals and the Roman Art of Imitation’, in B. Backes and C. von Nicolai (eds), Kulturelle Kohärenz durch Prestige. Munich: Utz, 133–​66.   2015. ‘Idealplastik and the Relationship between Greek and Roman Sculpture’, in Friedland, Grunow Sobocinski, and Gazda 2015: 240–​59. Antonelli, F., Pollini, J. and Cancelliere, S. 2017. ‘A Short Note on the Archaeometric Study of Two Sculptures in the Gabinetto Segreto of Naples National Archeological Museum. The Pan and the She-​Goat Group and the Bikini Venus’, Archaeological and Anthropological Sciences 9/​4: 685–​91. Antonsson, O. 1937. The Praxiteles Marble Group in Olympia. Stockholm:  Victor Pettersons Bokindustri. Arata, F. P. 2005. Opere d’arte dal mare. Testimonianze archeologiche subacquee del trasporto e del commercio marittimo di prodotti artistici. Rome: Libreria dello Stato, Istituto Poligrafico e Zecca dello Stato. Arnold, D. 1969. Die Polykletnachfolge. Berlin:  De Gruyter (= Jahrbuch des Deutschen Archäologischen Instituts. Ergänzungsheft, 25). Avery, C. 1987. Giambologna. The Complete Sculpture. Oxford:  Phaidon and Christie’s.

223

224

224

References Avery, V. 2015. ‘Alessandro Vittoria and the Art of Carving’, in Helmstutler Di Dio 2015: 87–​110. Azoulay, V. 2014. Les  Tyrannicides  d’Athènes:  Vie et mort de deux statues. Paris: Editions du Seuil. Bacchi, A. and Tumidei, S. 1998. Bernini. La scultura in San Pietro. Milan: Federico Motta. Baker, M. 2000. Figured in Marble. The Making and Viewing of Eighteenth-​Century Sculpture. London: V&A. Bang, M. 1921. ‘Preise von Statuen’, in L. Friedlaender (ed.), Darstellungen aus der Sittengeschichte Roms in der Zeit von August bis zum Ausgang der Antonine, vol. IV. Leipzig: S. Hirzel, 312–​25. Banham, M. (ed.) 1995. The Cambridge Guide to Theatre. Cambridge: Cambridge University Press. Barbanera, M. 2006. Original und Kopie. Bedeutungs-​und Wertewandel eines intellektuellen Begriffspaares seit dem 18. Jahrhundert in der Klassischen Archäologie. Stendal:  Winckelmann-​Gesellschaft (= Originale e copia nell’arte antica. Mantua: Tre Lune, 2011)   2008. ‘Original und Kopie. Aufstieg und Niedergang eines intellektuellen Begriffspaares seit dem 18. Jahrhundert in der Klassischen Archäologie’, in K. Junker and A. Stähli (eds), Original und Kopie. Formen und Konzepte der Nachahmung in der antiken Kunst. Wiesbaden: Reichert, 35–​61. Barker, S. 2016. ‘Calculating the Manpower for the Marble Decor of the Villa’, in E. K. Gazda and J. R. Clarke (eds), Leisure and Luxury in the Age of Nero. The Villas of Oplontis near Pompeii. Catalogue of the exhibition. Ann Arbor, MI: Kelsey Museum of Archaeology, 133–​5. Barocchi, P. 1960–​2. Trattati d’arte del Cinquecento. Fra manierismo e controriforma. Bari: Laterza.  1982. Il Bacco di Michelangelo /​Michelangelo’s Bacchus. Florence: SPES. Bartman, E. 1988. ‘Décor et Duplicatio:  Pendants in Roman Sculptural Display’, American Journal of Archaeology 92: 211–​25.  1992. Ancient Sculptural Copies in Miniature. Leiden: Brill.   2003. ‘Piecing as Paragone. Carlo Albacini’s Diana at Ince’, in J. Burnett Grossman, J. Podany and M. True (eds), History of Restoration of Ancient Stone Sculpture. Los Angeles: The J. Paul Getty Museum, 115–​26. Basile, J. 1993. ‘The Capestrano Warrior and Related Monuments of the Seventh to Fifth Centuries BC’, Revue des Archéologues et Historiens d’Art de Louvain 26: 9–​31. Bassett, J. 2008. The Craftsman Revealed. Adriaen de Vries Sculptor in Bronze. Los Angeles: The Getty Conservation Institute. Bailey, D. 1980. A Catalogue of the Lamps in the British Museum, vol. II, Roman Lamps Made in Italy. London: British Museum. Beaven, L. 2004. ‘Bernini’s Last Papal Portrait. The Bust of Pope Clement X Altieri and its Audience’, in D. R. Marshall (ed.), ‘The Italians’ in Australia. Studies in Renaissance and Baroque Art. Florence: Centro Di, 95–​106.

225

References Becatti, G. 1937. ‘Le tre Grazie’, Bullettino della Commissione Archeologica Municipale di Roma 65: 41–​60.   1956. ‘Letture pliniane. Le opere d’arte nei Monumenta Asini Pollionis e negli Horti Serviliani’, in Studi in onore di Aristide Calderini e Roberto Paribeni, vol. III, Studi di Archeologia e di Storia dell’Arte Antica. Milan: Casa Editrice Ceschina, 199–​210. Beck, H., Bol, P. C. and Bückling, M. (eds) 1990. Polyklet. Der Bildhauer der griechischen Klassik. Catalogue of the exhibition. Frankfurt am Main: Liebieghaus. Beck, H. and Bol, P. C. (eds) 1993. Polykletforschungen. Berlin: Gebr. Mann. Belli Pasqua, R. 1995. Sculture di età romana in ‘basalto’. Rome: ‘L’Erma’ di Bretschneider. Bennett, S. K. 1994. ‘The Washington Bronze Dionysos’, unpublished MA thesis. University of Maryland. Benson, J. L. 1959. ‘Spirally Fluted Columns in Greece’, Hesperia 28: 253–​72. Bergemann, J. 1990. Römische Reiterstatuen. Ehrendenkmäler im öffentlichen Bereich. Mainz am Rhein: P. von Zabern. Berger, E. (ed.) 1990. Antike Kunstwerke aus der Sammlung Ludwig. III. Skulpturen. Mainz am Rhein: P. von Zabern.   (ed.) 1992. Der Entwurf des Künstlers. Bildhauerkanon in der Antike und Neuzeit. Catalogue of the exhibition. Basel: Antikenmuseum und Sammlung Ludwig. Bergmann, M. 1999. Chiragan, Aphrodisias, Konstantinopel. Zur mythologischen Skulptur der Spätantike. Wiesbaden: Reichert. Bertoletti, M. 2006. ‘The Engine Room. The Monumental Center’, in M. Bertoletti, M. Cima and E. Talamo (eds), Centrale Montemartini. Musei Capitolini. Milan: Electa, 42–​73. Bieber, M. 1961. The Sculpture of the Hellenistic Age. 2nd edn. New York: Columbia University Press.  1977. Ancient Copies. Contributions to the History of Greek and Roman Art. Albany, NY: New York University Press. Bildwerke des Altertums in Abgüssen aus dem Albertinum zu Dresden. 1953. Dresden: Staatliche Kunstsammlungen. Blinkenberg, C. 1933. Knidia. Beiträge zur Kenntnis der Praxitelischen Aphrodite. Copenhagen: Levin & Munksgaard. Blümel, C. 1927. Griechische Bildhauerarbeit. Berlin: De Gruyter 1927 (=Jahrbuch des Deutschen Archäologischen Instituts. Ergänzungsheft, 11). Blume, C. 2012. ‘The Role of the Stone in the Polychrome Treatment of Hellenistic Sculptures’, in A. Gutiérrez Garcia-​Moreno, P. Lapuente Mercadal and I. Rodà de Llanza (eds), Interdisciplinary Studies on Ancient Stone. Proceedings of the IX Association for the Study of Marbles and Other Stones in Antiquity (ASMOSIA) Conference. Tarragona: Institut Català d’Arqueologia Clàssica, 754–​62.   2014. ‘Bright Pink, Blue and Other Preferences. Polychrome Hellenistic Sculpture’, in J. S. Østergaard and A. M. Nielsen (eds), Transformations. Classical Sculpture in Colour. Catalogue of the exhibition. Copenhagen: Ny Carlsberg Glyptotek, 166–​89.

225

226

226

References  2015. Polychromie Hellenistischer Skulptur. Ausführung, Instandhaltung und Botschaften, 2 vols. Tübingen: Imhof. Boardman, J. 1995. Greek Sculpture. The Late Classical Period. London: Thames & Hudson. Bol, P. C. 1972. Die Skulpturen des Schiffsfundes von Antikythera. Berlin: Gebr. Mann (Athenische Mitteilungen. Beiheft, 2). Bol, R. 1984. Das Statuenprogramm des Herodes-​Atticus-​Nymphäums. Berlin:  De Gruyter (Deutsches Archäologisches Institut. Olympische Forschungen, vol. XV).  1998. Amazones Volneratae. Untersuchungen zu den Ephesischen Amazonenstatuen. Mainz am Rhein: P. von Zabern.   2004. ‘Die ephesischen Amazonenstatuen’, in P. C. Bol (ed.), Die Geschichte der antiken Bildhauerkunst, vol. II, Klassische Plastik. Mainz am Rhein:  P.  von Zabern, 145–​58. Bonfante, L. and Carter, C. 1987. ‘An Absent Herakles and a Hesperid: A Late Antique Marble Group in New York’, American Journal of Archaeology 91: 247–​57. Bordenache, G. 1964. ‘Contributi per una storia dei culti e dell’arte nella Tomi d’età romana’, Studii Clasice 6: 155–​78. Boschung, D. 1989. Die Bildnisse des Caligula. Berlin: Gebr. Mann (=Das römische Herrscherbild, vol. IV). Bourgeois, B. 2003. ‘Secure for Eternity. Assembly Techniques for Large Statuary in the Sixteenth to Nineteenth Century’, in J. Burnett Grossman, J. Podany, and M. True (eds), History of Restoration of Ancient Stone Sculpture. Los Angeles: The J. Paul Getty Museum, 149–​62. Bourgeois, B., Jockey, P. and Karydas, A. 2011. ‘New Researches on Polychrome Hellenistic Sculptures in Delos, III:  The Gilding Processes. Observations and Meanings’, in P. Jockey (ed.), Leukos lithos. Marbres et autres roches de la Méditerranée antique:  études interdisciplinaires. Interdisciplinary Studies on Mediterranean Ancient Marble and Stones. Actes du VIIIe Colloque international de l’Association for the Study of Marble and Other Stones used in Antiquity (ASMOSIA). Aix-​en-​Provence: Karthala, 645–​61. Bouyia, P. 2012. ‘Maritime Commerce and Luxury in the Age of Cicero. The Evidence of the Antikythera Shipwreck’, in Kaltsas, Vlachoianni, and Bouyia 2012: 287–​92. Bradley, M. 2009. ‘The Importance of Colour on Ancient Marble Sculpture’, Art History 32.3: 427–​57. Braemer F. 1990. ‘Les relations commerciales et culturelles de Carthage avec 1’Orient romain à partir de documents sculptés’, in Histoire et Archéologie de l’Afrique du Nord. Actes du IVe colloque international, vol. I, Carthage et son territoire dans l’antiquité. Paris: Edition du CTHS, 175–​98. Braun, E. 1850. ‘L’Apoxyomenos di Lisippo’, Annali dell’Instituto di Corrispondenza Archeologica 22: 223–​51. Bravi, A. 2012. ‘Ornamenta urbis’. Opere d’arte greche negli spazi romani. Bari: Edipuglia.

227

References Brinkmann, V. 2003. Die Polychromie der archaischen und frühklassischen Skulptur. Munich: Biering & Brinkmann (Studien zur antiken Malerei und Farbgebung, vol. V). Brinkmann, V., Koch-​Brinkmann, U. and Piening, H. 2011. ‘Alte Gewänder in neuem Look. Beobachtungen zu den Farben der pompejanischen Artemis’, in M. Kunze (ed.), Die Artemis von Pompeji und die Entdeckung der Farbigkeit griechischer Plastik. Ruhpolding: Franz Philipp Rutzen, 69–​85. Brinkerhoff, D. M. 1970. A Collection of Sculpture in Classical and Early Christian Antioch. Albany, NY: New York University Press. Broccoli, U. 1982. Terracina. Museo e Raccolte Civiche, vol. I, Materiali di provenienza nota. Rome:  Bardi (Cataloghi dei musei locali e delle collezioni del Lazio, vol. IV). Brüggemann, N. 2007. ‘Kontexte und Funktionen von Kouroi im archaischen Griechenland’, in M. Meyer and N. Brüggemann, Kore und Kouros. Weihgaben für die Götter. Vienna: Phoibos, 91–​226. Brunnsåker, S. 1955. The Tyrant-​ Slayers of Kritios and Nesiotes. Lund:  Håkan Ohlssons Boktryckeri. Bruno, M., Attanasio, D. and Prochaska, W. 2015. ‘The Docimium Marble Sculptures of the Grotto of Tiberius at Sperlonga’, American Journal of Archaeology 119: 375–​94. Bruno, M., Lazzarini, L., Pensabene, P., Soligo, M. and Turi, B. 2002. ‘Provenance Studies of the White Marble Blocks and Architectural Elements from Porto and their Contributions to the History of the Marble Trade’, in J. J. Herrmann Jr., N. Herz, and R. Newman (eds), Asmosia V. Interdisciplinary Studies on Ancient Stone. Proceedings of the Fifth International Conference of the Association for the Study of Marbles and Other Stones in Antiquity. London: Archetype, 347–​58. Brunn, H. 1905. ‘Der Hermes des Praxiteles’, in Kleine Schriften, vol. II. Leipzig: Teubner, 379–​98. Budde, L. and Nicholls, R. V. 1964. A Catalogue of the Greek and Roman Sculpture in the Fitzwilliam Museum, Cambridge. Cambridge: Cambridge University Press. Bulard, M. 1906. ‘Aphrodite, Pan et Éros. Groupe en marbre’, Bulletin de Correspondance Hellénique 30: 610–​31. Burford, A. 1969. The Greek Temple Builders at Epidauros. A Social and Economic Study of Building in the Asklepian Sanctuary, during the Fourth and Early Third Centuries BC. Liverpool: Liverpool University Press. Burnett Grossman, J. 2003. Looking at Greek and Roman Sculpture in Stone. A Guide to Terms, Styles, and Techniques. Los Angeles: The J. Paul Getty Museum. Bussagli, M. 1959. ‘Buddha’, in EAA, vol. II, 215–​22. Bussels, S. 2012. The Animated Image. Roman Theory on Naturalism, Vividness and Divine Power. Berlin: Akademie Verlag. Cain, H.-​U. 2009. Aurea Aetas. Die Blütezeit des Leipziger Antikenmuseums zu Beginn des 20. Jahrhunderts. Leipzig: Antikenmuseum der Universität.

227

228

228

References Cancellieri, F. 1806. Dissertazioni epistolari di G.  B. Visconti e Filippo Waquier de la Barthe sopra la statua del Discobolo scoperta nella Villa Palombara. Rome: Antonio Fulgoni. Carpenter, R. 1931. ‘Who Carved the Hermes of Praxiteles?’, American Journal of Archaeology 35: 249–​61. Carradori, F. 1802. Istruzione elementare per gli studiosi della scultura. Florence:  Tipografia della Società Letteraria (Elementary instructions for Students of Sculpture, tr. M.  Kalevi Auvinen. Los Angeles:  The J.  Paul Getty Museum, 2002). Carroll, D. 1987. Paraesthetics: Foucault, Lyotard, Derrida. New York: Methuen. Caskey, L. D. 1925. Catalogue of the Greek and Roman Sculpture, Museum of Fine Arts, Boston. Cambridge, MA: Harvard University Press. Cassieri, N. 2000. Sperlonga. La Grotta di Tiberio e il Museo Archeologico Nazionale. Rome: Istituto Poligrafico e Zecca dello Stato.  2013. Formia. Museo Archeologico Nazionale e monumenti. Rome: Istituto Poligrafico e Zecca dello Stato. Casson, S. 1931. ‘The Hermes of Praxiteles’, American Journal of Archaeology 35: 262–​8. Catoni, M. L. 2005. Schemata. Comunicazione non verbale nella Grecia antica. Pisa: Edizioni della Normale.  2008. La comunicazione non verbale nella Grecia antica. Gli schemata nella danza, nell’arte, nella vita. Turin: Bollati Boringhieri. Chamay, J. and Maier, J.-​L. 1990. Sculptures en pierre du Musée de Genève, vol. I, Art Grec. Mainz am Rhein: P. von Zabern. Cianfarani, V. 1968. ‘Note sul restauro del guerriero di Capestrano’, Rivista dell’Istituto Nazionale di Archeologia e Storia dell’Arte 15: 5–​19. Claridge, A. 1988. ‘Roman Statuary and the Supply of Statuary Marble’, in J. C. Fant (ed.), Ancient Marble Quarrying and Trade. Oxford: BAR, 139–​52.   1990. ‘Ancient Techniques of Making Joins in Marble Statuary’, in M. True and J. Podany (eds), Marble. Art Historical and Scientific Perspectives on Ancient Sculpture. Los Angeles: The J. Paul Getty Museum, 135–​62.   2015. ‘Marble Carving, Techniques, Workshops, and Artisans’, in Friedland, Grunow Sobocinski, and Gazda 2015: 107–​22. Clinton, K. 2005–​ 8. Eleusis. The Inscriptions on Stone. Documents of the Sanctuary of the two Goddesses and Public Documents of the Deme, 2  vols. Athens: Archaeological Society at Athens. Coarelli, F. 1973. ‘Sperlonga e Tiberio’, Dialoghi di Archeologia 7: 97–​122.   1983. ‘Il commercio delle opere d’arte in età tardo-​repubblicana’, Dialoghi di Archeologia, 3rd series, 1: 45–​53. Cole, M. 2007. ‘Bernini Struts’, in J. Sofaer (ed.), Material Identitites. Malden, MA: Blackwell, 55–​66. Coliva, A. 1998. ‘Apollo e Dafne’, in A. Coliva and S. Schütze (eds), Bernini scultore. La nascita del Barocco in Casa Borghese. Rome: De Luca, 252–​75.

229

References   (ed.) 2002. Bernini scultore. La tecnica esecutiva. Rome: De Luca. Comstock, M. B. and Vermeule, C. C. 1976. Sculpture in Stone. The Greek, Roman and Etruscan Collection of the Museum of Fine Arts Boston. Boston: Museum of Fine Arts. Conticello, B. 1974. ‘I gruppi scultorei di soggetto mitologico a Sperlonga’, in B. Conticello and B. Andreae, Die Skulpturen von Sperlonga. Antike Plastik 14: 7–​60.   1996. ‘Il gruppo di Scilla e della nave’, in B. Andreae and C. Parisi Presicce (eds), Ulisse. Il mito e la memoria. Rome: Progetti Museali Editore, 280–​315. Conticello, B., Franchi dall’Orto, L. and Varone, A. (eds) 1990. Rediscovering Pompeii. Catalogue of the exhibition. Rome: ‘L’Erma’ di Bretschneider. Conze, A. 1911–​22. Die attischen Grabreliefs, vol. IV. Berlin: De Gruyter. Corso, A. 1996. ‘The Hermes of Praxiteles’, Numismatica e Antichità Classiche. Quaderni Ticinesi 25: 131–​53.  2004. The Art of Praxiteles, vol. I, The Development of Praxiteles’ Workshop and its Cultural Tradition until the Sculptor’s Acme. Rome:  ‘L’Erma’ di Bretschneider.  2007. The Art of Praxiteles, vol. II, The Mature Years. Rome:  ‘L’Erma’ di Bretschneider.  2010. The Art of Praxiteles, vol. III, The Advanced Maturity of the Sculptor. Rome: ‘L’Erma’ di Bretschneider.  2013. The Art of Praxiteles, vol. IV, The Last Phase of his Activity. Rome: ‘L’Erma’ di Bretschneider. Costabile, F. 1995. ‘Le statue frontonali del Tempio Marasà a Locri Epizefiri’, Römische Mitteilungen 102: 9–​62. Daehner, J. 2005. ‘Grenzen der Nacktheit. Studien zum männlichen Körper in der griechischen Plastik des 5.  und 4.  Jahrhunderts v.  Chr.’, Jahrbuch des Deutschen Archäologischen Instituts 120: 155–​300. Daehner, J. and Lapatin, K. (eds) 2015. Power and Pathos. Bronze Sculpture of the Hellenistic World. Catalogue of the exhibition. Los Angeles:  The J.  Paul Getty Trust. De Angelis D’Ossat, M. (ed.) 2002. Scultura antica in Palazzo Altemps. Milan: Electa. De Angelis D’Ossat, M. and Capodiferro, A. (eds) 2011. Palazzo Altemps. Le collezioni. Milan: Electa. de Chaisemartin, N. and Örgen, E. 1983. ‘Les sculptures de Silahtaraga’, Revue Archéologique: 181–​9. de Chaisemartin, N. and Örgen, E. 1984. Les Documents sculptés de Silahtarağa. Paris: Editions Recherche sur les Civilisations. De Luca, G. 1975. ‘Der Satyr im Palazzo Corsini/​Rom. Eine Replik der Gruppe Aufforderung zum Tanz’, Antike Plastik 15: 73–​81. De Luca Savelli, M. 1981. ‘Le opere del Mochi’, in M. De Luca Savelli (ed.), Francesco Mochi 1580–​1654. Florence: Centro Di.

229

230

230

References Derrida, J. 1987. The Truth in Painting, tr. G. Bennington and I. McLeod. Chicago: Chicago University Press. Dickins, G. 1912. Catalogue of the Acropolis Museum, vol. I, Archaic Sculpture. Cambridge: Cambridge University Press. Di Pasquale, G. 2004. Tecnologia e meccanica. Trasmissione dei saperi tecnici dall’età ellenistica al mondo romano. Florence: Leo S. Olschki. Dohrn, T. 1968. ‘Die Marmor-​Standbilder des Daochos Weihgeschenks in Delphi’, Antike Plastik 8: 33–​53. Dombrowski, D. 1997. Giuliano Finelli:  Bildhauer zwischen Neapel und Rom. Frankfurt am Main: Peter Lang. Dugas, C. 1924. Le sanctuaire d’Aléa Athéna a Tégée au IVe siècle. Paris: Librairie Orientaliste Paul Geuthner. Duncan-​Jones, R. 1962. ‘Costs, Outlays and Summae Honorariae from Roman Africa’, Papers of the British School at Rome 30: 47–​115.   1965. ‘An Epigraphic Survey of Costs in Roman Italy’, Papers of the British School at Rome 33: 189–​306.  1982. The Economy of the Roman Empire. Quantitative Studies. 2nd edn. Cambridge: Cambridge University Press. Duro, P. (ed.) 1996. The Rhetoric of the Frame. Essays on the Boundaries of the Artwork. Cambridge: Cambridge University Press. Eaverly, M. A. 1995. Archaic Greek Equestrian Sculpture. Ann Arbor, MI: University of Michigan Press. Elsner, J. 1995. Art and the Roman Viewer. The Transformation of Art from the Pagan World to Christianity. Cambridge: Cambridge University Press.   2013. ‘The Lycurgus Cup’, in C. Entwistle and L. James (eds), New Light on Old Glass: Recent Research on Byzantine Mosaics and Glass. London: British Museum, 103–​11. Fant, J. C. 2008. ‘Quarrying and Stoneworking’, in J. P. Oleson (ed.), The Oxford Handbook of Engineering and Technology in the Classical World. Oxford: Oxford University Press, 121–​35. Fant, J. C. and Barker, S. 2016. ‘The Cost of Luxury. Procurement and Labor for the Marble Decor of Villa A’, in E. K. Gazda and J. R. Clarke (eds), Leisure and Luxury in the Age of Nero. The Villas of Oplontis near Pompeii. Catalogue of the exhibition. Ann Arbor, MI: Kelsey Museum of Archaeology, 126–​32. Favero, M. 2008. Francesco Mochi. Una carriera di scultore. Trento: UNI Service. Fendt, A. 2012. Archäologie und Restaurierung. Die Skulpturenergänzungen in der Berliner Antikensammlung des 19. Jahrhunderts. Berlin: De Gruyter. Feuser, S. 2013. Monopodia. Figürliche Tischfüsse aus Kleinasien. Istanbul: Yayınları. Filges, A. 1999. ‘Marmorstatuetten aus Kleinasien. Zur Ikonographie, Funktion und Produktion antoninischer, severischer und späterer Idealplastik’, Istanbuler Mitteilungen 49: 377–​430. Fittschen, K. 1977. Katalog der antiken Skulpturen in Schloss Erbach. Berlin: Gebr. Mann.

231

References Fittschen, K. and Zanker, P. 1985. Katalog der römischen Porträts in den Capitolinischen Museen und den anderen kommunalen Sammlungen der Stadt Rom, vol. I, Kaiser-​und Prinzenbildnisse. Mainz am Rhein: P. von Zabern. Foerster, G. 2005. ‘A Modest Aphrodite from Bet Shean’, Israel Museum Studies in Archeology 4: 3–​21. Foley, K. 2016. ‘Puppets in Traditional Asian Theatre’, in S. Liu (ed.), Routledge Handbook of Asian Theatre. London: Routledge, 177–​201. Friedland, E. A. 2003a. ‘Art as Cultural Artifact. Roman Sculpture in the Semitic East’, in D. R. Clark and V. H. Matthews (eds), One Hundred Years of American Archaeology in the Middle East. Proceedings of the American Schools of Oriental Research Centennial Celebration, Washington, DC, April 2000. Boston: American Schools of Oriental Research, 331–​44.   2003b. ‘The Roman Marble Sculptures from the North Hall of the East Baths at Gerasa’, American Journal of Archaeology 107: 413–​48.  2012. The Roman Marble Sculptures from the Sanctuary of Pan at Caesarea Philippi/​Panias (Israel). Boston: American Schools of Oriental Research. Friedland, E. A., Grunow Sobocinski, M. and Gazda, E. K. (eds) 2015. The Oxford Handbook of Roman Sculpture. Oxford: Oxford University Press. Fröhlich, B. 1998. Die statuarischen Darstellungen der hellenistischen Herrscher. Hamburg: Dr. Kovač. Fullerton, M. D. 1990. The Archaistic Style in Roman Statuary. Leiden: Brill. Gallo, D. 1992–​3. ‘Originali greci e copie romane secondo Giovanni Battista ed Ennio Quirino Visconti’, Labyrinthos 21–​4: 215–​51. García y Bellido, A. 1949. Esculturas romanas de España y Portugal. Madrid: Consejo Superior de Investigaciones Científicas. Gasparri, C. 1994. ‘Copie e copisti’, in EAA, 2nd supplement, vol. II, 267–​80.   (ed.) 2009. Le sculture Farnese, vol. I, Le sculture ideali. Naples: Electa.   (ed.) 2010. Le sculture Farnese, vol. III, Le sculture delle Terme di Caracalla. Rilievi e varia. Naples: Electa. Gazda, E. K. 1981. ‘A Marble Group of Ganymede and the Eagle from the Age of Augustine’, in J. H. Humphrey (ed.), Excavations at Carthage 1977 Conducted by the University of Michigan, vol. VI. Ann Arbor, MI:  Kelsey Museum of Archaeology, 125–​81. Gell, A. 1998. Art and Agency. An Anthropological Theory. Oxford: Clarendon Press. Geominy W. 1998. ‘Zum Daochos-​Weihgeschenk’, Klio 80: 369–​402.   1999a. ‘Zur Komposition der Gruppe Die Aufforderung zum Tanz’, in Gedenkschrift für Andreas Linfert. Mainz am Rhein: P. von Zabern, 140–​55. Geominy, W. 1999b. ‘Zwischen Kennerschaft und Cliché. Römische Kopien und die Geschichte ihrer Bewertung’, in G. Vogt-​Spira and B. Rommel (eds), Rezeption und Identität. Die kulturelle Auseinandersetzung Roms mit Griechenland als europäisches Paradigma. Stuttgart: Steiner, 38–​58. Geominy W. 2007. ‘The Daochos Monument at Delphi:  The Style and Setting of a Family Portrait in Historic Dress’, in P. Schultz and R. von den Hoff (eds),

231

232

232

References Early Hellenistic Portraiture. Image, Style, Context. Cambridge: Cambridge University Press, 84–​98. Gercke, P. 1968. Satyrn des Praxiteles, unpublished PhD thesis, University of Hamburg.   (ed.) 1991. Apollon und Athena. Klassische Götterstatuen in Abgüssen und Rekonstruktionen. Catalogue of the exhibition. Kassel:  Staatliche Kunstsammlungen. Gercke, P. and Zimmermann-​Elseify, N. 2007. Antike Skulpturen und neuzeitliche Nachbildungen in Kassel. Bestandkatalog. Mainz am Rhein: P. von Zabern. Germini, B. 2008. Statuen des strengen Stils in Rom. Verwendung und Wertung eines griechischen Stils im römischen Kontext. Rome: ‘L’Erma’ di Bretschneider. Gerstein, A. (ed.) 2007. Display and Displacement. Sculpture and the Pedestal from Renaissance to Post-​Modern. London: Courtauld Institute of Art. Giuliani, L. 2013a. Image and Myth. A History of Pictorial Narration in Greek Art. Chicago: University of Chicago Press (tr. of Bild und Mythos: Geschichte der Bilderzählung in der griechischen Kunst. Munich: Beck, 2003).   2013b. ‘Kästchen oder Quader? Zur Sitzstatue des Lorenzo de’ Medici in der Sagrestia Nuova und zum Problem der Materialität in den Skulpturen Michelangelos’, Mitteilungen des Kunsthistorischen Institutes in Florenz 55: 335–​57.   2016. ‘Michelangelos Quader. Ein Nachtrag’, Mitteilungen des Kunsthistorischen Institutes in Florenz 58: 109–​16. Giuliani, L. and Catoni, M. L. 2016. ‘Myron und die Kunst des Diskuswerfens’, Römische Mitteilungen 122: 13–​43. Giuliano, A. (ed.) 1979. Museo Nazionale Romano. Le Sculture, vol. I.1. Rome: De Luca.   (ed.) 1992. La collezione Boncompagni Ludovisi. Algardi, Bernini e la fortuna dell’antico. Venice: Marsilio.   (ed.) 1995. Museo Nazionale Romano. Le Sculture, vol. I.12, Magazzini. Sculture greche del V secolo/​Parte I. Rome: De Luca. Goldberg, E. L. 1996. ‘Artistic Relations between the Medici and the Spanish Courts 1587–​1621. Part II’, The Burlington Magazine 138.1121: 529–​40. Gombrich, E. H. 1960. Art and Illusion. A  Study in the Psychology of Pictorial Representation. London and New York: Phaidon and Pantheon. Grassinger, D. 1999. Die mythologischen Sarkophage, vol. I. Berlin: Gebr. Mann (Die antiken Sarkophagreliefs, vol. XII.1). Grüßinger, R., Kästner, V. and Scholl, A. (eds) 2011. Pergamon. Panorama der antiken Metropole. Berlin and Petersburg: Staatliche Museen zu Berlin and Imhof. Guattani, G. A. 1784. Monumenti antichi inediti ovvero Notizie sulle antichità e belle arti di Roma per l’anno MDCCLXXXIV. Rome: Pagliarini. Habetzeder, J. 2012. ‘The Impact of Restoration: The Example of the Dancing Satyr in the Uffizi’, Opuscula. Annual of the Swedish Institutes at Athens and Rome 5: 133–​63.

233

References Hafner, G. 1961. Geschichte der Griechischen Kunst. Zurich: Atlantis. Hallett, C. H. 2005. ‘Emulation versus Replication. Redefining Roman Copying’, Journal of Roman Archaeology 18: 419–​35. Hampe, R. 1972. Sperlonga und Virgil. Mainz am Rhein: P. von Zabern. Hannestad, N. 1989. Review of: B. Andreae and B. Conticello, Skylla und Charybdis (Mainz am Rhein, 1987). Gnomon 61: 655–​7.  1994. Tradition in Late Antique Sculpture. Conservation –​Modernization –​ Production. Aarhus: Aarhus University Press.   2007. ‘Late Antique Mythological Sculpture. In Search of a Chronology’, in F. A. Bauer and C. Witschel (eds), Statuen in der Spätantike. Wiesbaden: Reichert, 273–​305. Harder, A. 2012. Callimachus, Aetia. Introduction, Text, Translation, and Commentary, 2 vols. Oxford: Oxford University Press. Hardiman, C. 2012 ‘Popular Aesthetics and Personal Art Appreciation in the Hellenistic Age’, in I. Sluiter and R. M. Rosen (eds), Aesthetic Value in Classical Antiquity. Leiden: Brill, 265–​83. Harvey, I. E. 1989. ‘Derrida, Kant, and the Performance of Parergonality’, in H. J. Silverman (ed.), Derrida and Deconstruction. New York: Routledge, 57–​74. Haskell, F. and Penny, N. 1981. Taste and the Antique. The Lure of Classical Sculpture 1500–​1900. New Haven: Yale University Press. Hauser, F. 1906. ‘Athlet oder Apollon?’ Jahreshefte des Österreichischen Archäologischen Institutes 9: 279–​87. Havelock, C. M. 1995. The Aphrodite of Knidos and her Successors. A Historical Review of the Female Nude in Greek Art. Ann Arbor, MI: University of Michigan Press. Hellenkemper Salies, G. 1994. Das Wrack. Der antike Schiffsfund von Mahdia, 2 vols. Cologne: Rheinland. Helmstutler Di Dio, K. (ed.) 2015. Making and Moving Sculpture in Early Modern Italy. Farnham: Ashgate.   2015. ‘Shipping Sculpture, Shaping Diplomacy. Gifts of Sculpture for Spain’, in Helmstutler Di Dio 2015: 166–​90. Hemingway, S., Milleker, E. J. and Stone, R. 2002. ‘The Early Imperial Bronze Statue of a Boy in the Metropolitan Museum of Art: A Technical and Stylistic Analysis’, in A. Giumlia-​Mair (ed.), I bronzi antichi. Produzione e tecnologia. Montagnac: Monique Mergoil, 200–​7. Hermann Fiore, K. 1997. ‘Apollo e Dafne del Bernini al tempo del Cardinale Scipione Borghese’, in K. Hermann Fiore (ed.), Apollo e Dafne del Bernini nella Galleria Borghese. Cinisello Balsamo: Silvana Editoriale, 71–​109. Hessler, C. J. 2014. Zum Paragone. Malerei, Skulptur und Dichtung in der Rangstreitkultur des Quattrocento. Berlin: Akademie Verlag and De Gruyter. Higginbotham, J. 1997. Piscinae. Artificial Fishponds in Roman Italy. Chapel Hill, NC, and London: University of North Carolina Press. Hiller, S. 1976. ‘Statuenstützen im fünften Jahrhundert v.  Chr’, Antike Kunst 19: 30–​40.

233

234

234

References Himmelmann, N. 1995. Sperlonga. Die homerischen Gruppen und ihre Bildquellen. Opladen: Westdeutscher Verlag. Hirst, M. 1985. ‘Michelangelo, Carrara, and the Marble for the Cardinal’s Pietà’, Burlington Magazine 127.984: 152–​6, 159. Hofter, M. R. 2011. ‘Von Dübeln und Löchern –​Antike Anstückungstechniken an pergamenischen Skulpturen’, in R. Grüßinger, V. Kästner and A. Scholl (eds), Pergamon. Panorama der antiken Metropole. Berlin and Petersburg: Staatliche Museen zu Berlin and Imhof, 334–​7. Holland, L. A. 1956. ‘The Purpose of the Warrior Image from Capestrano’, American Journal of Archaeology 60: 243–​7. Hollinshead, M. B. 1998. ‘Hair Struts in Late Roman Sculpture’, in K. J. Hartswick and M. C. Sturgeon (eds), ΣΤΕΦΑΝΟΣ. Studies in Honor of Brunilde Sismondo Ridgway. Philadelphia:  University of Pennsylvania and University Museum, 119–​30.   2002a. ‘Extending the Reach of Marble: Struts in Greek and Roman Sculpture’, in E. K. Gazda (ed.), The Ancient Art of Emulation. Studies in Artistic Originality and Tradition from the Present to Classical Antiquity. Ann Arbor, MI: University of Michigan Press, 117–​52.   2002b. ‘From Two to Three Dimensions in Unfinished Roman Sculpture’, in J. J. Herrmann Jr., N. Herz and R. Newman (eds), Asmosia V. Interdisciplinary Studies on Ancient Stone. Proceedings of the Fifth International Conference of the Association for the Study of Marbles and Other Stones in Antiquity. London: Archetype, 225–​30.   2015. ‘Awaiting Identity:  Copenhagen’s Diskophoros and its Auxiliary Support’, in P. Pensabene and E. Gasparini (eds), Asmosia X. Proceedings of the Tenth International Conference of ASMOSIA. Rome:  ‘L’Erma’ di Bretschneider, 171–​7. Hölscher, T. 1974. ‘Die Nike der Messenier und Naupaktier in Olympia’, Jahrbuch des Deutschen Archäologischen Instituts 89: 70–​111.  2004. The Language of Images in Roman Art. Cambridge: Cambridge University Press. Hopkins, R. 1998. Picture, Image and Experience. A  Philosophical Inquiry. Cambridge: Cambridge University Press.   2003. ‘Sculpture and Space’, in M. Kieran and D. M. Lopes (eds), Imagination, Philosophy and the Arts. London: Routledge, 274–​92. Hughes, F. 2010. Kant’s Critique of Aesthetic Judgement. A Reader’s Guide. London and New York: Continuum. Hurwit, J. M. 1989. ‘The Kritios Boy: Discovery, Reconstruction, and Date’, American Journal of Archaeology 93: 41–​80. Huskinson, J. 2015. Roman Strigillated Sarcophagi. Art and Social History. Oxford: Oxford University Press. Hutton, C. A. 1914–​16. ‘The Greek Inscriptions at Petworth House’, Annual of the British School at Athens 21: 155–​65.

235

References Ibrahim, L., Scranton, R. and Brill, R. 1976. Kenchreai Eastern Port of Corinth, vol. II, The Panels of Opus Sectile in Glass. Leiden: Brill. Inan, J. 1973. ‘Roman Copies of Some Famous Greek Statues from Side’, Antike Plastik 12: 69–​79.  1975. Roman Sculpture in Side. Ankara: Türk tarih kurumu basımevi.   1993. ‘Der sandalenbindende Hermes’, Antike Plastik 22: 105–​16. Inan, J. and Rosenbaum, E. 1966. Roman and Early Byzantine Portrait Sculpture in Asia Minor. London: Oxford University Press for the British Academy. Inan, J. and Alföldi-​ Rosenbaum, E. 1979. Römische und frühbyzantinische Porträtplastik aus der Türkei. Neue Funde. Mainz am Rhein: P. von Zabern. Irvin, S. 2013. ‘Sculpture’, in B. Gaut and D. M. Lopes (eds), The Routledge Companion to Aesthetics. 3rd edn. London: Routledge, 606–​15. Isager, J. 1991. Pliny on Art and Society. The Elder Pliny’s Chapters on the History of Art. Odense: Odense University Press. Israel, J. 2013. ‘Zum Grabmal von Kallithea  –​Kleinasiatische Bautraditionen im Spätklassischen Athen’, in K. Sporn (ed.), Griechische Grabbezirke klassischer Zeit. Normen und Regionalismen. Munich: Hirmer, 55–​68. Jacquemin, A. and Laroche, D. 2001. ‘Le monument de Daochos, ou le trésor des Thessaliens’, Bulletin de Correspondance Hellénique 125: 305–​32. Jockey, P. 1999. ‘La technique composite à Delos à l’Époque hellénistique’, in Archéomatériaux. Marbres et autres roches. ASMOSIA 4. Bordeaux: Université Michel de Montaigne, 305–​16. Johansen, F. 1994. Catalogue. Roman Portraits I. Copenhagen:  Ny Carlsberg Glyptotek. Jones, E. 2015. ‘The Sculptural Stones of Venice: The Selection, Supply, and Cost of Marble and Stone in the Sixteenth Century’, in Helmstutler Di Dio 2015: 111–​36. Kaltsas, N. 2002. Sculpture in the National Archaeological Museum, Athens. Los Angeles: The J. Paul Getty Museum.   (ed.) 2004. Agon. Catalogue of the exhibition. Athens:  Hellenic Ministry of Culture. Kaltsas, N., Vlachogianni, E. and Bouyia, P. (eds) 2012. The Antikythera Shipwreck. The Ship, the Treasures, the Mechanism. Catalogue of the exhibition. Athens: National Archaeological Museum. Kalveram, K. 1995. Die Antikensammlung des Kardinals Scipione Borghese. Worms am Rhein: Wernersche Verlagsgesellschaft. Kane, S. and Carrier, S. C. 1988. ‘Relationships between Style and Size of Statuary and the Availability of Marble at Cyrene’, in N. Herz and M. Waelkens (eds), Classical Marble. Geochemistry, Technology, Trade. Dordrecht:  Kluwer Academic Publishers, 197–​206.   1992. ‘Relationships between Style and Size of Statuary and the Availability of Marble in the Eastern Roman Empire’, in M. Waelkens, N. Herz and L. Moens (eds), Ancient Stones. Quarrying, Trade and Provenance. Interdisciplinary

235

236

236

References Studies on Stones and Stone Technology in Europe and Near East from the Prehistoric to the Early Christian Period. Leuven: Leuven University Press, 121–​7. Kane, S., Polikreti, K., Maniatis, Y., Herz, N. and Carrier, S. C. 1999. ‘Investigation of the Pentelic Marble Sculptures of the Nymphaeum of Herodes Atticus in Olympia, Greece’, in Archéomatériaux. Marbres et autres roches. ASMOSIA 4. Bordeaux: Université Michel de Montaigne, 317–​23. Kane Trimble, S. 1976–​7. ‘Sculpture Appendix’, Libya Antiqua 13–​14: 313–​30. Kansteiner, S. 2013. Review of: H. Weinstock, Statuenstützen an vorkaiserzeitlichen Marmorstatuen (Hamburg, 2012). Göttinger Forum für Altertumswissenschaft 16: 1047–​51. Kant, I. 2000. Critique of Judgement, ed. P. Guyer, tr. P.  Guyer and E.  Matthews. Cambridge: Cambridge University Press. Karanastassis, P. 1987. ‘Untersuchungen zur kaiserzeitlichen Plastik in Griechenland. II: Kopien, Varianten und Umbildungen nach Athena-​Typen des 5. Jhs. v. Chr.’, Athenische Mitteilungen 102: 323–​428. Kauth, C. 2010. Der sogenannte Protesilaos. Zwei Repliken in New York und London, unpublished MA thesis, Ludwig Maximilian University Munich. Kitzinger, E. 1978. ‘The Cleveland Marbles’, in Atti del IX congresso internazionale di archeologia cristiana, vol. I, I monumenti cristiani precostantiniani. Vatican City: Pontificio Istituto di Archeologia Cristiana, 653–​75. Klapisch-​Zuber, C. 1969. Les maîtres du marbre. Carrare, 1300–​1600. Paris: SEVPEN. Klatt, U. 1995. ‘Römische Klapptische. Drei-​und vierbeinige Stützgestelle aus Bronze und Silber’, Kölner Jahrbuch 28: 349–​573. Knittlmayer, B. and Heilmeyer, W.-​D. (eds) 1998. Die Antikensammlung. Altes Museum. Pergamonmuseum. Mainz am Rhein: von Zabern. Knoll, K. (ed.) 1994. Das Albertinum vor 100 Jahren –​die Skulpturensammlung Georg Treus. Catalogue of the exhibition. Dresden: Staatliche Kunstsammlungen. Knoll, K., Vorster, C. and Woelk, M. (eds) 2011. Skulpturensammlung Staatliche Kunstsammlungen Dresden. Katalog der antiken Bildwerke. II. Idealskulptur der römischen Kaiserzeit, 2 vols. Munich: Hirmer. Koçak, M. 2013. Aphrodite am Pfeiler. Studien zu aufgestützten/​angelehnten weiblichen Figuren der griechischen Marmorplastik. Istanbul: Yayınları. Koch, G. and Sichtermann, H. 1982. Römische Sarkophage. Munich: Beck (Handbuch der Archäologie). Koortbojian, M. 2002. ‘Forms of Attention:  Four Notes on Replication and Variation’, in E. K. Gazda (ed.), The Ancient Art of Emulation. Studies in Artistic Originality and Tradition from the Present to Classical Antiquity. Ann Arbor, MI: University of Michigan Press, 173–​204. Körner, H. 2010. ‘Statuenstützen im Werk Jean-​Antoine Houdons’, in M Bückling and G. Scherf (eds), Jean-​Antoine Houdon. Die Sinnliche Skulptur. Catalogue of the exhibition. Frankfurt: Liebieghaus, 259–​77. Körner, H. and Möseneder, K. (2008). Format und Rahmen. Vom Mittelalter bis zur Neuzeit. Berlin: Reimer.

237

References Kommer, B. R. (ed.) 2000. Adriaen de Vries 1556–​1626. Augsburgs Glanz –​Europas Ruhm. Augsburg: Städtische Kunstsammlungen. Kousser, R. M. 2008. Hellenistic and Roman Ideal Sculpture. The Allure of the Classical. Cambridge: Cambridge University Press. Kranz, P. 1989. ‘Die Asklepiosstatue im Schlosspark von Klein-​Glienicke. Ein neuer Typus und sein kulturgeschichtlicher Hintergrund’, Jahrbuch des Deutschen Archäologischen Instituts 104: 107–​55. Kreeb, M. 1988. Untersuchungen zur figürlichen Ausstattung delischer Privathäuser. Chicago: Ares. Kreikenbom, D. 1990. Bildwerke nach Polyklet. Kopienkritische Untersuchungen zu den männlichen statuarischen Typen nach polykletischen Vorbildern. Diskophoros, Hermes, Doryphoros, Herakles, Diadumenos. Berlin: Gebr. Mann.   2013. ‘Originale und Kopien. Signaturen an späthellenistischen und kaiserzeitlichen Skulpturen’, in N. Hegener and F. Horsthemke (eds), Künstlersignaturen von der Antike bis zur Gegenwart. Petersberg: Imhof, 60–​73. Krinzinger, F. (ed.) 2010. Hanghaus 2 in Ephesos. Die Wohneinheiten 1 und 2.  Baubefund, Ausstattung, Funde. Textband Wohneinheit 1. Vienna: Österreichische Akademie der Wissenschaften. Kunze, C. 1996. ‘Zur Datierung des Laokoon und der Skyllagruppe aus Sperlonga’, Jahrbuch des Deutschen Archäologischen Instituts 111: 139–​223.  1998. Der Farnesische Stier und die Dirkegruppe des Apollonios und Tauriskos. Berlin:  De Gruyter (=Jahrbuch des Deutschen Archäologischen Instituts. Ergänzungsheft, 30).  2002. Zum Greifen nah. Stilphänomene in der hellenistischen Skulptur und ihre inhaltliche Interpretation. Munich: Biering & Brinkmann. Lahusen, G. 1999. ‘Antenore, Plinio e la statuaria romana’, in E. Formigli (ed.), I grandi bronzi antichi. Le fonderie e le tecniche di lavorazione dall’età arcaica al Rinascimento. Siena: Nuova Immagine Editrice, 35–​41. Lahusen, G. and Formigli, E. 2001. Römische Bildnisse aus Bronze. Kunst und Technik. Munich: Hirmer. Lankheit, K. 1962. Florentinische Barockplastik. Die Kunst am Hofe der letzten Medici, 1670–​1743. Munich: Bruckmann. Landwehr, C. 1982. Griechische Meisterwerke in römischen Abgüssen. Der Fund von Baia. Zur Technik antiker Kopisten. Frankfurt am Main: Liebieghaus.  1985. Die Antiken Gipsabgüsse aus Baiae: griechische Bronzestatuen in Abgüssen römischer Zeit. Berlin: Gebr. Mann.  1993. Die Römischen Skulpturen von Caesarea Mauretaniae, vol. I, Idealplastik. Weibliche Figuren benannt. Mainz am Rhein: P. von Zabern.  2000. Die Römischen Skulpturen von Caesarea Mauretaniae, vol. II, Idealplastik. Männliche Figuren. Mainz am Rhein: P. von Zabern.  2006. Die Römischen Skulpturen von Caesarea Mauretaniae, vol. III, Idealplastik. Bacchus und Gefolge  –​Masken  –​Fabelwesen  –​Tiere  –​Bukranien  –​Nicht benennbare Figuren. Mainz am Rhein: P. von Zabern.

237

238

238

References   2010. ‘The Baiae Casts and the Uniqueness of Roman Copies’, in R. Frederiksen and E. Marchand (ed.), Plaster Casts: Making, Collecting and Displaying from Classical Antiquity to the Present. Berlin: De Gruyter, 35–​46. Langl, J. 1887. Griechische Götter und Heldengestalten nach antiken Bildwerken. Vienna: Alfred Hölder K.K. Hof-​und Universitäts-​Buchhändler. Langlotz, E. 1947. Phidiasprobleme. Frankfurt am Main: Klostermann.   1971. ‘Zur Deutung des sogenannten Protesilaos in New  York’, Archäologischer Anzeiger: 427–​42. Lapatin, K. and Wight, K. (eds) 2010. The J. Paul Getty Museum. Handbook of the Antiquities Collection. Los Angeles: Getty Publications. La Rocca, E. 1985. Amazzonomachia. Le sculture frontonali del tempio di Apollo Sosiano. Rome: De Luca.  1987. L’Auriga dell’Esquilino. Rome: ‘L’Erma’ di Bretschneider.   1998. ‘Artisti rodii negli horti romani’, in M. Cima and E. La Rocca (eds), Horti romani. Rome: ‘L’Erma’ di Bretschneider, 203–​74. La Rocca, E. and Parisi Presicce, C. (eds) 2010. Musei Capitolini. Le sculture del Palazzo Nuovo, vol. I. Milan: Electa.   with Lo Monaco, A. (eds) 2012. L’età dell’equilibrio, 98–​180 d.C. Traiano, Adriano, Antonino Pio, Marco Aurelio. Catalogue of the exhibition. Rome:  Mondo Mostre. Lauter, H. 1967. ‘Eine seitenverkehrte Kopie des Apoxyomenos’, Bonner Jahrbücher des Rheinischen Landesmuseums 167: 119–​28.   1969. ‘Zur Datierung der Skulpturen von Sperlonga’, Römische Mitteilungen 76: 162–​73. Lavin, I. 1998. ‘Ex Uno Lapide:  The Renaissance Sculptor’s Tour de Force’, in M. Winner, B. Andreae, and C. Pietrangeli (eds), Il cortile delle statue. Der Statuenhof des Belvedere im Vatikan. Mainz am Rhein: P. von Zabern, 191–​210.  2012. Bernini at Saint Peter’s. The Pilgrimage. London: Tindar Press. Layton, R. H. 2005. ‘Structure and Agency in Art’, in M. Coquet, B. Derlon and M. Jeudy-​Ballini (eds), Les cultures à l’œuvre. Rencontres en art. Paris: Maison des Sciences de l’Homme, 29–​46. Lechat, H. 1903. Au Musée de l’Acropole d’Athènes. Études sur la sculpture en Attique avant la ruine de l’Acropole lors de l’invasion de Xerxès. Lyon and Paris: A. Roy and Libraire A. Fontemoing. León, P. 2008. ‘Nueva replica de la Amazona Sciarra’, in E. La Rocca, P. León, and C. Parisi Presicce (eds), Le due patrie acquisite. Studi di archeologia dedicati a Walter Trillmich. Rome: ‘L’Erma’ di Bretschneider, 243–​53. Lierke, R. 2013. ‘On the Manufacture of Diatreta and Cage Cups from the Pharos Beaker to the Lycurgus Cup’, in C. Entwistle and L. James (eds), New Light on Old Glass: Recent Research on Byzantine Mosaics and Glass. London: British Museum, 89–​102. Linfert, A. 1979. Review of: J. Inan, Roman Sculpture in Side (Ankara, 1975). Bonner Jahrbücher des Rheinischen Landesmuseums 179: 780–​5.

239

References   1993. ‘Aus Anlaß neuer Repliken des Westmacottschen Epheben und des Dresdner Knaben’, in Beck and Bol 1993: 141–​92. Lippold, G. 1923. Kopien und Umbildungen Griechischer Statuen. Munich: Beck.  1936. Die Skulpturen des Vatikanischen Museums, vol. III.1. Berlin: De Gruyter.  1956. Die Skulpturen des Vatikanischen Museums, vol. III.2. Berlin: De Gruyter. Liverani, P. 1988–​9. ‘L’Antiquarium di Villa Barberini a Castel Gandolfo’, Rendiconti della Pontificia Accademia Romana di Archeologia 61: 102–​30.  1989. L’antiquarium di Villa Barberini a Castel Gandolfo. Vatican City: Tipografia Poliglotta Vaticana (Monumenti Musei e Gallerie Pontificie).   1993. ‘Il Doriforo del Braccio Nuovo e l’Efebo tipo Westmacott di Castel Gandolfo. Nota sul restauro e sul contesto’, in Beck and Bol 1993: 117–​40.   2003. ‘Der Augustus von Prima Porta’, in V. Brinkmann and R. Wünsche (eds), Bunte Götter. Die Farbigkeit antiker Skulptur. Munich:  Staatliche Antikensammlungen und Glyptothek, 187–​91 (= 2004. ‘L’Augusto di Prima Porta’, in P. Liverani (ed.), I colori del bianco. Policromia nella scultura antica. Rome: De Luca, 235–​42).   2006. ‘Il Laocoonte in età antica’, in F. Buranelli, P. Liverani and A. Nesselrath (eds), Laocoonte. Alle origini dei Musei Vaticani. Rome:  ‘L’Erma’ di Bretschneider, 23–​40. Locchi, A. 2006. ‘Vestire gli dèi, vestirsi per gli dèi (nella Grecia antica)’, in M. Rivarli, M. Álvarez Sosa and A. Locchi (eds), Contesti antichi. Rome: Edizioni Nuova Cultura, 77–​100. Lopes, D. M. 2005. Sight and Sensibility. Evaluating Pictures. Oxford: Clarendon Press. Löwy, E. 1885. Inschriften griechischer Bildhauer. Leipzig: Teubner.   1905. ‘Athlet oder Apollon?’, Jahreshefte des Österreichischen Archäologischen Institutes 8: 269–​76. Ma, J. 2013. Statues and Cities. Honorific Portraits and Civic Identity in the Hellenistic World. Oxford: Oxford University Press. Maderna-​Lauter, C. 2002. ‘Spätarchaische Plastik’, in P. C. Bol (ed.), Die Geschichte der antiken Bildhauerkunst, vol. I, Frühgriechische Plastik. Mainz am Rhein: P. von Zabern, 223–​69. Maffei, P. A. 1704. Raccolta di Statue antiche e moderne. Rome: Stamperia alla Pace. Maffei, S. 1999. ‘La fama di Laocoonte nei testi del Cinquecento’, in S. Settis (ed.), Laocoonte. Fama e stile. Rome: Donzelli, 85–​230. Mägele, S., Richard, J. and Waelkens, M. 2007. ‘A Late-​Hadrianic Nymphaeum at Sagalassos (Pisidia, Turkey). A  Preliminary Report’, Istanbuler Mitteilungen 57: 469–​504. Manderscheid, H. 1981. Die Skulpturenausstattung der Kaiserzeitlichen Thermenanlagen. Berlin: Gebr. Mann. Mansuelli, G. A. 1958–​61. Galleria degli Uffizi:  le sculture, 2  vols. Rome:  Istituto Poligrafico dello Stato. Marcadé, J. 1969. Au Musée de Délos. Étude sur la sculpture hellénistique en ronde bosse découverte dans l’ile. Paris: De Boccard.

239

240

240

References Marcadé,  J. (ed.) 1996. Sculptures Déliennes. Athens and Paris:  École Française d’Athènes and De Boccard 1996. Marquardt, N. 1995. Pan in der Hellenistischen und Kaiserzeitlichen Plastik. Bonn: Habelt. Martinez, J.-​L. 2007a. ‘L’Apollon Sauroctone’, in Pasquier and Martinez 2007: 202–​15.   2007b. ‘Les Satyres de Praxitèle’, in Pasquier and Martinez 2007: 236–​59. Marvin, M. 2008. The Language  of the  Muses:  The Dialogue between Greek and Roman Sculpture. Los Angeles: The J. Paul Getty Museum. Matravers, D. 2013. ‘Wollheim’, in B. Gaut and D. M. Lopes (eds), The Routledge Companion to Aesthetics. 3rd edn. London: Routledge, 200–​9. Mattusch, C. C. (ed.) 1996. The Fire of Hephaistos. Large Classical Bronzes from North American Collections. Cambridge, MA: Harvard University Art Museums. Maviglia, A. 1913. ‘Gli attributi dei sostegni nella statuaria antica’, Römische Mitteilungen 28: 1–​91. McCann, A. M. 1978. Roman Sarcophagi in the Metropolitan Museum of Art. New York: Metropolitan Museum of Art. McCloskey, M. A. 1987. Kant’s Aesthetics. Basingstoke: Macmillan. Mendel, G. 1912–​14. Musées Impériaux Ottomans. Catalogue des Sculptures grecques, romaines et byzantines, 3 vols. Istanbul: Imperial Museum. Mendelsohn, L. 1982. Paragoni. Benedetto Varchi’s ‘Due Lezzioni’ and Cinquecento Art Theory. Ann Arbor, MI: UMI Research Press. Merker, G. 1973. The Hellenistic Sculpture of Rhodes. Gothenburg: Åström. Michalowski, C. 1932. Exploration archéologique de Délos, vol. XIII, Les portraits hellénistiques et romains. Paris: De Boccard. Michaelis, A. 1882. Ancient Marbles in Great Britain. Cambridge: Cambridge University Press.  1901. Strassburger Antiken. Festgabe für die Archäologische Sektion der XLVI. Versammlung Deutscher Philologen und Schulmänner dargeboten von dem Kunstarchäologischen Institut der Kaiser-​ Wilhelms-​ Universität Strassburg. Strasbourg: Trübner. Miniero, P. and Zevi, F. (eds) 2008. Museo Archeologico dei Campi Flegrei. Catalogo generale. Liternum, Baia, Miseno. Naples: Electa. Molloy, M. E. 1996. Libanius and the Dancers. Hildesheim: Olms-​Weidmann. Moltesen, M. (ed.) 2002. Catalogue. Imperial Rome II. Statues. Copenhagen:  Ny Carlsberg Glyptotek.   (ed.) 2005. Catalogue. Imperial Rome III. Copenhagen: Ny Carlsberg Glyptotek. Montagu, J. 1985. ‘Bernini Sculptures Not by Bernini’, in I. Lavin (ed.), Bernini: New Aspects of his Art and Thought. University Park, PA:  Pennsylvania State University Press, 23–​43.  1989. Roman Baroque Sculpture. The Industry of Art. New Haven:  Yale University Press. Moormann, E. M. 1988. La pittura parietale romana come fonte di conoscenza per la scultura antica. Assen/​Maastricht: Van Gorcum.

241

References Moreno, P. 1994. Scultura ellenistica, 2 vols. Rome: Istituto Poligrafico dello Stato.   (ed.) 1995. Lisippo. L’arte e la fortuna. Catalogue of the exhibition. Rome: Fabbri. Moser von Filseck, K. 1988. Der Apoxyomenos des Lysipp und das Phänomen von Zeit und Raum in der Plastik des 5. und 4. Jhs.v.Chr. Bonn: Habelt. Mueller, M. 2016. Objects as Actors. Props and the Poetics of Performance in Greek Tragedy. Chicago: University of Chicago Press. Müller, H. W. 1975. ‘Zwei Darstellungen des Ganymed aus Ägypten’, in Wandlungen. Studien zur antiken und neueren Kunst. Ernst Homann-​Wiedeking gewidmet. Munich: Stiftland-​Verlag Waldsassen-​Bayern, 235–​42. Murray, A. S. 1885. ‘A Terra-​Cotta Diadumenos’, Journal of Hellenic Studies 6: 243–​7. Mustilli, D. 1939. Il Museo Mussolini. Rome: Libreria dello Stato. Muth, S. 2007. ‘Das Manko der Statuen? Zum Wettstreit der bildlichen Ausstattung im spätantiken Wohnraum’, in F. A. Bauer and C. Witschel (eds), Statuen in der Spätantike. Wiesbaden: Reichert, 341–​55. Muthmann, F. 1927. Hadrianische und Antoninische Statuenstützen. Beiträge zur Geschichte der römischen Kopistentätigkeit. Freiburg im Breisgau:  Freiburger Studentendruckerei.   1931. ‘Fragmente einer Wiederholung des Satyrs mit der Fussklapper’, Athenische Mitteilungen 56: 87–​9.  1951. Statuenstützen und dekoratives Beiwerk an griechischen und römischen Bildwerken. Ein Beitrag zur Geschichte der römischen Kopistentätigkeit. Heidelberg: Carl Winter Universitätsverlag. Neudecker, R. 1988. Die Skulpturen-​Ausstattung römischer Villen in Italien. Mainz am Rhein: P. von Zabern. Newby, Z. 2005. Greek Athletics in the Roman World: Victory and Virtue. Oxford: Oxford University Press. Nielsen, A. M. and Østergaard, J. S. 1997. Catalogue. The Eastern Mediterranean in the Hellenistic Period. Copenhagen: Ny Carlsberg Glyptotek. Niemeier, W.-​D. 2002. Der Kouros vom Heiligen Tor. Überraschende Neufunde archaischer Skulptur im Kerameikos in Athen. Mainz am Rhein: P. von Zabern. Nolte, S. 2006. Steinbruch, Werkstatt, Skulptur. Untersuchungen zu Aufbau und Organisation griechischer Bildhauerwerkstätten. Göttingen: Ruprecht. Østergaard, J. S. (ed.) 1996. Catalogue. Imperial Rome. Copenhagen: Ny Carlsberg Glyptotek.   2008. ‘Emerging Color: Roman Sculptural Polychromy Revived’, in R. Panzanelli, E. D. Schmidt, and K. Lapatin (eds), The Color of Life: Polychromy in Sculpture from Antiquity to the Present. Los Angeles: The J. Paul Getty Museum, 40–​61.   2010. ‘The Sciarra Amazon Investigation: Some Archaeological Comments’, in J. S. Østergaard (ed.), Tracking Colour. The Polychromy of Greek and Roman Sculpture in the Ny Carlsberg Glyptotek, Preliminary Report, vol. II. Copenhagen: Ny Carlsberg Glyptotek, 56–​60 (available for download at www. trackingcolour.com).

241

242

242

References   2015. ‘Identical Roman Copies? Diversification through Color’, in S. Settis, A. Anguissola, and D. Gasparotto (eds), Serial/​Portable Classic. Milan: Fondazione Prada, 113–​17. Østergaard, J. S., Sargent, M. L. and Therkildsen, R. H. 2014. ‘The Polychromy of Roman “Ideal” Marble Sculpture of the 2nd Century ce’, in P. Liverani and U. Santamaria (eds), Diversamente bianco: La policromia della scultura romana. Rome: Quasar, 51–​69. Özgen, E. and Özgen, İ. (eds) 1988. Antalya Museum. Ankara: Ministry of Culture. Özgüt, M. E. 1996. Skulpturen des Museums von Antalya, vol. I, Ausflug in die Mythologie und Geschichte. Antalya: Museum. Padgett, M. (ed.) 2001. Roman Sculpture in the Princeton Art Museum. Princeton: Princeton University Press. Pafumi, S. 2007. ‘Per la ricostruzione degli arredi scultorei del Palazzo dei Cesari sul Palatino’, BABESCH. Annual Papers in Mediterranean Archaeology 82: 207–​25. Papangelē, K. 2002. Eleusina. Ho archaiologikos chōros kai to museio. Athens: EFG Eurobank Ergasias. Papini, M. 2008. L’Apollo di Mantova. Rome: Palombi.   2010. ‘Giunone Cesi’, in E. La Rocca, C. Parisi Presicce and A. Lo Monaco (eds), I giorni di Roma. L’età della conquista. Catalogue of the exhibition. Milan: Skira, 159–​60 no. I.18.   2015. ‘Republican Rome and Italic Art’, in B. E. Borg (ed.), A Companion to Roman Art. Malden, MA: Wiley Blackwell, 95–​113. Paribeni, E. 1959. Catalogo delle sculture di Cirene. Rome: ‘L’Erma’ di Bretschneider (Monografie di Archeologia Libica, vol. V). Parker, G. 2014. ‘Mobile Monumentality. The Case of Obelisks’, in J. F. Osborne (ed.), Approaching Monumentality in Archaeology. Albany, NY:  State University of New York Press, 273–​87. Pasquier, A. 2007a. ‘Praxitèle aujourd’hui? La question des originaux’, in Pasquier and Martinez 2007: 82–​105.   2007b. ‘Les Aphrodites de Praxitèle’, in Pasquier and Martinez 2007: 130–​51. Pasquier, A. and Martinez, J.-​L. (eds) 2007. Praxitèle. Catalogue of the exhibition. Paris: Musée du Louvre. Penny, N. 1993. The Materials of Sculpture. New Haven: Yale University Press. Pensabene, P. 1994. Le vie del marmo. I blocchi di cava di Roma e di Ostia. Il fenomeno del marmo nella Roma antica. Rome:  Ministero per i Beni Culturali e Ambientali, Soprintendenza Archeologica di Ostia.   1998. ‘Il fenomeno del marmo nella Roma tardo-​repubblicana e imperiale’, in P. Pensabene (ed.), Marmi antichi II. Cave e tecnica di lavorazione, provenienze e distribuizone. Rome: ‘L’Erma’ di Bretschneider, 333–​90.   2010. ‘Cave di marmo bianco e pavonazzetto in Frigia. Sulla produzione e sui dati epigrafici’, Marmora 6: 71–​134.  2013. I marmi nella Roma antica. Rome: Carocci.

243

References Pfanner, M. 1988. ‘Vom “Laufenden Bohrer” bis zum “Bohrlosen Stil”. Überlegungen zur Bohrtechnik in der Antike’, Archäologischer Anzeiger: 667–​76.   1989. ‘Über das Herstellen von Porträts: Ein Beitrag zu Rationalisierungsmassnahmen und Produktionsmechanismen von Massenware im späten Hellenismus und in der römischen Kaiserzeit’, Jahrbuch des Deutschen Archäologischen Instituts 104: 157–​257. Picón, C. A. 1995. ‘Polykleitan and Related Sculptures in American Collections: Recent Acquisitions’, in W. G. Moon (ed.), Polykleitos, the Doryphoros, and Tradition. Madison, WI: University of Wisconsin Press, 229–​45. Piwowarczyk, K. 2000. ‘Le dauphin. Attribut des statues romaines’, in K. M. Ciałowicz and J. A. Ostrowski (eds), Les civilisations du bassin méditerranéen. Hommages à Joachim Śliwa. Cracow: Instytut archeologii UJ, 299–​305. Platt, V. 2011. Facing the Gods. Epiphany and Representation in Graeco-​Roman Art, Literature and Religion. Cambridge: Cambridge University Press. Platt, V. and Squire, M. (eds) 2017. The Frame in Classical Art. A Cultural History. Cambridge: Cambridge University Press. Poeschel, S. 2002–​3. ‘Agostino Cornacchinis Reiterstandbild Karl des Großen im Portikus von Sankt Peter in Rom’, in T. Kraus and K. Pabst (eds), Karl der Große und sein Nachleben in Geschichte, Kunst und Literatur. Zeitschrift des Aachener Geschichtsvereins 104–​5: 673–​703. Polanyi, M. 1970. ‘What is a Painting?’, British Journal of Aesthetics 10: 225–​36 (= The American Scholar 39, 1970: 655–​69). Pollini, J. 1987. The Portraiture of Gaius and Lucius Caesar. New  York:  Fordham University Press. Pollitt, J. J. 1974. The Ancient View of Greek Art: Criticism, History, and Terminology. New Haven: Yale University Press.   2000. ‘The Phantom of a Rhodian School of Sculpture’, in N. T. de Grummond and B. S. Ridgway (eds), From Pergamon to Sperlonga. Sculpture and Context. Berkeley-​Los Angeles and London: University of California Press, 92–​110. Pope-​Hennessy, J. 1964. Catalogue of Italian Sculpture in the Victoria and Albert Museum, 3 vols. London: Her Majesty’s Stationery Office. Preisshofen, R. 2002. ‘Der Apollon Sauroktonos des Praxiteles’, Antike Plastik 28: 41–​115. Pressouyre, S. 1984. Nicolas Cordier. Recherches sur la sculpture à Rome autour de 1600. Rome: École Française de Rome. Primavesi, O. 2008. ‘Artemis, her Shrine, and her Smile: Winckelmann’s Discovery of Ancient Greek Polychromy’, in V. Brinkmann, O. Primavesi, and M. Hollein (eds), Circumlitio. The Polychromy of Antique and Mediaeval Sculpture. Frankfurt am Main: Liebieghaus Skulpturensammlung, 24–​77.   2011. ‘Das Lächeln der Artemis. Winckelmanns Entdeckung der Farbigkeit griechischer Skulptur’, in M. Kunze (ed.), Die Artemis von Pompeji und die Entdeckung der Farbigkeit griechischer Plastik. Ruhpolding:  Franz Philipp Rutzen, 17–​67.

243

244

244

References Prioux, E. 2006. ‘Materiae non cedit opus. Matières et sujets dans les épigrammes descriptives (IIIe siècle av. J.-​C. –​50 apr. J.-​C.)’, in A. Rouveret, S. Dubel and V. Naas (eds), Couleurs et matières dans l’Antiquité. Textes, techniques et pratiques. Paris: Éditions Rue d’Ulm, 127–​60. Prisco, G. 1991. ‘Dalle Terme al Museo di Napoli’, in Il Toro Farnese. La ‘montagna di marmo’ tra Roma e Napoli. Naples: Macchiaroli, 47–​68. Proskynitopoulou, R. 2012. ‘L’Éphèbe de Sarrebruck X 26087. Quelques observations techniques sur sa mise en oeuvre, sa restauration et sa présentation au Musée National Archéologique d’Athènes’, in M. Denoyelle, S. Descamps-​ Lequime, B. Mille and S. Verger (eds), Bronzes grecs et romains, recherches récentes. Hommage à Claude Rolley. Paris: Institut National d’Histoire de l’Art. https://​inha.revues.org/​3905. Pucci, G. 2008. ‘Agency, oggetto, immagine. L’antropologia dell’arte di Alfred Gell e l’antichità classica’, Ricerche di Storia dell’Arte 94: 35–​40. Raeder, J. 2000. Die antiken Skulpturen in Petworth House (West Sussex). Mainz am Rhein: P. von Zabern. Raepsaet, G. 1984. ‘Transport de tambours de colonnes du Pentélique à Éleusis au IVe siècle avant notre ère’, L’antiquité classique 53: 101–​36.   2008. ‘Land Transport, Part 2: Riding, Harnesses, and Vehicles’, in J. P. Oleson (ed.), Oxford Handbook of Engineering and Technology in the Classical World. Oxford: Oxford University Press, 580–​605. Rausa, F. 1994. L’immagine del vincitore. L’atleta nella statuaria greca dall’età arcaica all’ellenismo. Rome: Viella. Rice, E. E. 1986. ‘Prosopographika Rhodiaka’, Annual of the British School at Athens 81: 109–​250. Richter, G. M. A. 1915. Greek, Etruscan and Roman Bronzes. New York: Metropolitan Museum of Art.   1931. ‘The Hermes of Praxiteles’, American Journal of Archaeology 35: 277–​90.   1935. ‘Another Copy of the Diadoumenos by Polykleitos’, American Journal of Archaeology 39: 46–​52.  1954. Metropolitan Museum of Art. Catalogue of Greek Sculptures. Cambridge, MA: Harvard University Press.  1955. Ancient Italy. A Study of the Interrelations of its People as Shown in their Arts. Ann Arbor, MI: University of Michigan Press.   1962. ‘How were the Roman Copies of Greek Portraits Made?’ Römische Mitteilungen 69: 52–​8.  1968. Korai. Archaic Greek Maidens. London: Phaidon.  1970. Kouroi. Archaic Greek Youths. 3rd edn. London: Phaidon. Ridgway, B. S. 1966. ‘Stone and Metal in Greek Sculpture’, Archaeology 19: 31–​42.   1974. ‘A Story of Five Amazons’, American Journal of Archaeology 78: 1–​17.  1984. Roman Copies of Greek Sculpture. The Problem of the Originals. Ann Arbor, MI: University of Michigan Press.  1990. Hellenistic Sculpture, vol. I, The Styles of ca. 331–​200 B.C. Bristol:  Bristol Classical Press.

245

References  2002. Hellenistic Sculpture, vol. III, The Styles of ca. 100–​ 31 B.C. Madison, WI: University of Wisconsin Press. Rizzo, G. E. 1907. ‘Il Discobolo di Castel Porziano’, Bollettino d’Arte 1: 3–​14.  1932. Prassitele. Milan and Rome: Treves, Treccani & Tumminelli. Robert, C. 1919. Die antiken Sarkophagreliefs, vol. III.3, Einzelmythen. Niobiden –​ Triptolemos. Berlin: Grote. Robert, L. 1962. ‘Les Kordakia de Nicée, le combustible de Synnada et les poissons-​ scies. Sur des lettres d’un métropolite de Phrygie au Xe siècle. Philologie et réalités, II’, Journal des Savants: 5–​74. Robert, R. 1995. ‘Immensa potentia artis. Prestige et statut des œuvres d’art à Rome à la fin de la République et au début de l’Empire’, Revue Archéologique: 291–​305. Robertson, M. 1981. A Shorter History of Greek Art. Cambridge: Cambridge University Press. Rockwell, P. 1991. ‘Unfinished Statuary Associated with a Sculptor’s Studio’, in R. R. R. Smith and K. T. Erim (eds), Aphrodisias Papers, vol. II, The Theatre, a Sculptor’s Workshop, Philosophers, and Coin-​Types. Ann Arbor, MI (Journal of Roman Archaeology, supplementary series, 2), 127–​43.  1993. The Art of Stoneworking:  A Reference Guide. Cambridge: Cambridge University Press.   1997. ‘La tecnica scultorea di Apollo e Dafne’, in K. Hermann Fiore (ed.), Apollo e Dafne del Bernini nella Galleria Borghese. Cinisello Balsamo: Silvana Editoriale, 139–​47. Rodowick, D. N. 1994. ‘Impure Mimesis, or the Ends of the Aesthetics’, in P. Brunette and D. Wills (eds), Deconstruction and the Visual Arts. Art, Media, and Architecture. Cambridge: Cambridge University Press, 96–​117. Romano, I. B. 1988. ‘Early Greek Cult Images and Cult Practices’, in R. Hägg, N. Marinatos and G. C. Nordquist (eds),  Early Greek Cult Practice. Stockholm: Åström (=Acta Instituti Atheniensis Regni Sueciae, 38), 127–​33. Rosenbaum, E. 1960. A Catalogue of Cyrenaican Portrait Sculpture. London: British Academy. Rouveret, A. 1989. Histoire et imaginaire de la peinture ancienne. Ve siècle av. J.-​C.–​Ier siècle ap. J.-​C. Rome: École Française de Rome (reprint: École Française de Rome, Classiques, 2014). Rubensohn, O. 1935. ‘Parische Künstler’, Jahrbuch des Deutschen Archäologischen Instituts 50: 49–​69. Rusnak, T. J. 2001. ‘The Active Spectator: Art and the Viewer in Ancient Greece’, unpublished PhD thesis, Bryn Mawr College. Russell, B. 2013a. The Economics of the Roman Stone Trade. Oxford: Oxford University Press.   2013b. ‘Roman and Late-​ Antique Shipwrecks with Stone Cargoes:  A New Inventory’, Journal of Roman Archaeology 26: 331–​61.   2015. ‘Transport and Distribution’, in Friedland, Grunow Sobocinski, and Gazda 2015: 189–​205.

245

246

246

References Russenberger, C. 2015. Der Tod und die Mädchen. Amazonen auf römischen Sarkophagen. Berlin: De Gruyter. Rutledge, S. H. 2012. Ancient Rome as a Museum. Power, Identity, and the Culture of Collecting. Oxford: Oxford University Press. Säflund, G. 1972. The Polyphemus and Scylla Groups at Sperlonga. Stockholm: Almqvist & Wiksell. Saladino, V. 2006. ‘The Athlete with a Strigil’, in M. Michelucci (ed.), Apoxyomenos. The Athlete of Croatia. Florence: Giunti, 35–​51. Sargent, M. L. and Therkildsen, R. H. 2010. ‘The Technical Investigation of Sculptural Polychromy at the Ny Carlsberg Glyptotek 2009–​2010. An Outline’, in J. S. Østergaard (ed.), Tracking Colour. The Polychromy of Greek and Roman Sculpture in the Ny Carlsberg Glyptotek, Preliminary Report, vol. II. Copenhagen: Ny Carlsberg Glyptotek, 11–​26 (available for download at www. trackingcolour.com). Schade, K. 1998. ‘Die zwei Gesichter der Berliner Knöchelspielerin’, Jahrbuch der Berliner Museen 40: 187–​98.   2007. ‘Ein Paragone der Künste. Betrachtungen zur Idealplastik der mittleren Kaiserzeit’, Jahrbuch des Deutschen Archäologischen Instituts 122: 163–​99. Schädler, U. 1992. ‘Antiken für das Neue Museum. Ausgrabungen Pius VI. in der Villa der Quintilier’, Antike Welt 23: 201–​18. Schapiro, M. 1969. ‘On Some Problems in the Semiotics of Visual Art: Field and Vehicle in Image-​Signs’, Semiotica 1: 223–​42. Schmidt, E. E. 1968. ‘Die Mars-​Venus-​Gruppe im Museo Capitolino’, Antike Plastik 8: 85–​94.   1973. ‘Die Kopien der Erechtheionkoren’, Antike Plastik 13. Schmidt, M. 1969. Der Basler Medeasarkophag. Ein Meisterwerk spätantoninischer Kunst. Tübingen: Wasmuth. Schmidt, S. 2009. ‘Images of Statues on Attic Vases: The Cases of the Tyrannicides’, in V. Nørskov, L. Hannestad, C. Isler-​Kerényi, and S. Lewis (eds), The World of Greek Vases. Rome: Quasar (Analecta Romana Instituti Danici. Supplementum, 41), 219–​37. Schneider, C. 1999. Die Musengruppe von Milet. Mainz am Rhein: P. von Zabern (Deutsches Archäologisches Institut. Milesische Forschungen, 1). Schoch, K. 2009. Die doppelte Aphrodite –​alt und neu bei griechischen Kultbildern. Göttingen: Universitätsverlag. Scholl, A. and Platz-​Horster, G. (eds) 2007. Die Antikensammlung. Altes Museum. Pergamonmuseum. Berlin: Staatliche Museen zu Berlin and P. von Zabern. Schrader, H. 1924. Phidias. Frankfurt: Frankfurter Verlags-​Anstalt A.G. Schröder, S. F. 2004. Katalog der antiken Skulpturen des Museo del Prado in Madrid, vol. II, Idealplastik. Mainz am Rhein: P. von Zabern.   (ed.) 2009. Verwandelte Götter. Antike Skulpturen des Museo del Prado zu Gast in Dresden. Catalogue of the exhibition. Dresden and Madrid: Staatliche Kunstsammlungen Dresden and Museo del Prado.

247

References Scigliano, E. 2005. Michelangelo’s Mountain. The Quest for Perfection in the Marble Quarries of Carrara. New York: Free Press. Sena Chiesa, G. (ed.) 2012. Costantino 313 d.C. Catalogue of the exhibition. Milan: Electa. Sepe, C. 2008. ‘Regiones VII–​VIII’, in A. Carrella, L. A. D’Acunto, N. Inserra and C. Sepe, Marmora pompeiana nel Museo Archeologico Nazionale di Napoli: gli arredi scultorei delle case pompeiane. Rome: ‘L’Erma’ di Bretschneider, 109–​53. Settis, S. 1992. ‘Fortuna del Diadumeno. I testi’, Numismatica e Antichità Classiche. Quaderni Ticinesi 21: 51–​68.  1999. Laocoonte. Fama e stile. Rome: Donzelli. Sichtermann, H. and Koch, G. 1975. Griechische Mythen auf römischen Sarkophagen. Tübingen: Wasmuth. Silva Maroto, M. P. 1995. ‘La escultura en Madrid en la época de Carlos II. Importación de obras y coleccionismo’, Anales de Historia del Arte 5: 205–​24. Simmel, G. 1902. ‘Der Bildrahmen. Ein ästhetischer Versuch’, Der Tag 541, 18 Nov. (=‘The Picture Frame: An Aesthetic Study’, tr. M. Ritter, Theory, Culture and Society 11 (1994): 11–​17). Simon, E. 1957. ‘Zur Augustusstatue von Prima Porta’, Römische Mitteilungen 64: 46–​68. Simonato, L. 2005. ‘Ludovico Sergardi, Agostino Cornacchini e la statua vaticana del Carlo Magno’, Prospettiva 119–​20: 23–​63. Slater, N. W. 1998. ‘Passion and Petrifaction. The Gaze in Apuleius’, Classical Philology 93: 18–​48. Slavazzi, F. 1996. Italia verius quam provincia. Diffusione e funzione delle copie di sculture greche nella Gallia Narbonense. Naples: Edizioni Scientifiche Italiane.   2002. ‘Immagini riflesse. Copie e doppi nelle sculture di Villa Adriana’, in A. M. Reggiani (ed.), Villa Adriana. Paesaggio antico e ambiente moderno. Elementi di novità e ricerche in corso. Milan: Electa, 52–​61. Smith, A. H. 1892–​1904. A Catalogue of Sculpture in the Department of Greek and Roman Antiquities. British Museum, 3 vols. London: British Museum. Smith, R. R. R. 1991. Hellenistic Sculpture. A Handbook. London: Thames & Hudson.   (ed.) 2006. Aphrodisias II. Roman Portrait Statuary from Aphrodisias. Mainz am Rhein: P. von Zabern. Soleti, V. M. 2010. La scultura ideale romana nella Regio Secunda (Apulia et Calabria). Rome: Giorgio Bretschneider. Spinola, G. 2004. Il Museo Pio Clementino, vol. III. Vatican City: Musei Vaticani 2004. Spyropoulos, G. 2001. Drei Meisterwerke der griechischen Plastik aus der Villa des Herodes Atticus zu Eva/​Loukou. Frankfurt am Main: Peter Lang. Squire, M. 2007. ‘The Motto in the Grotto. Inscribing Illustration and Illustrating Inscription at Sperlonga’, in Z. Newby and R. Leader-​Newby (eds), Art and Inscriptions in the Ancient World. Cambridge: Cambridge University Press, 102–​27.

247

248

248

References  2009. Image and Text in Graeco-​ Roman Antiquity. Cambridge: Cambridge University Press.   2013. ‘Embodied Ambiguities on the Prima Porta Augustus’, Art History 36: 242–​79. Stähli, A. 1999. Die Verweigerung der Lüste. Erotische Gruppen in der antiken Plastik. Berlin: Reimer.   2003. ‘Erzählte Zeit, Erzählzeit und Wahrnehmungszeit. Zum Verhältnis von Temporalität und Narration, speziell in der hellenistischen Plastik’, in P. C. Bol (ed.), Zum Verhältnis von Raum und Zeit in der griechischen Kunst. Möhnesee: Bibliopolis, 239–​64. Steinhauer, G. 2001. The Archaeological Museum of Piraeus. Athens: Latsis Group (= G. Steinhauer, Ta mnēmeia kai to Archaiologiko Mouseio tou Peiraia. Athens: Toumpēs, 1998). Stemmer, K. (ed.) 1995. Standorte. Kontext und Funktion antiker Skulptur. Berlin: Freunde & Förderer der Abguss-​Sammlung Antiker Plastik e.V. Stephens, S. A. 2015. Callimachus: The Hymns. Oxford: Oxford University Press. Stewart, A. 1977a. Skopas of Paros. Park Ridge: Noyes Press.   1977b. ‘To Entertain an Emperor: Sperlonga, Laocoon and Tiberius at the Dinner-​ Table’, Journal of Roman Studies 67: 76–​90.  1990. Greek Sculpture. An Exploration. New Haven: Yale University Press. Stewart, P. 2003. Statues in Roman Society. Representation and Response. Oxford: Oxford University Press. Stirling, L. M. 1996. ‘Divinities and Heroes in the Age of Ausonius: A Late-​Antique Villa and Sculptural Collection at Saint-​Georges-​de-​Montagne (Gironde)’, Revue Archéologique: 103–​45.   1997a. ‘Late-​Antique Goddesses and Other Statuary at the Villa of La-​Garenne-​ de-​Nérac (Lot-​et-​Garonne)’, Echos du Monde Classique /​Classical Views 41, ns 16: 149–​75.   1997b. ‘Statuary Collecting and Display in the Late Antique Villas of Gaul and Spain: A Comparative Study’, in F. A. Bauer and C. Witschel (eds), Statuen in der Spätantike. Wiesbaden: Reichert, 307–​21.  2005. The Learned Collector. Mythological Statuettes and Classical Taste in Late Antique Gaul. Ann Arbor, MI: University of Michigan Press.   2008. ‘Pagan Statuettes in Late Antique Corinth:  Sculpture from the Panayia Domus’, Hesperia 77: 89–​161. Stone, I. and Stone, J. 1962. I, Michelangelo, Sculptor. An Autobiography through Letters. Garden City, NY: Doubleday & Co. Strocka, V. M. 1999. ‘Zur Datierung der Sperlonga-​Gruppen und des Laokoon’, in Gedenkschrift für Andreas Linfert. Mainz am Rhein: P. von Zabern, 307–​22. Strong, D. and Claridge, A. 1976. ‘Marble Sculpture’, in D. Strong and D. Brown (eds), Roman Crafts. London: Duckworth, 195–​207. Stuart Jones, H. 1912. A Catalogue of the Ancient Sculptures Preserved in the Municipal Collections of Rome:  The Sculptures of the Museo Capitolino. Oxford: Clarendon Press.

249

References  1926. A Catalogue of the Ancient Sculptures Preserved in the Municipal Collections of Rome. The Sculptures of the Palazzo dei Conservatori. Oxford: Clarendon Press. Studniczka, F. 1888. ‘Die archaische Artemisstatuette aus Pompeii’, Römische Mitteilungen 3: 277–​302.   1910. ‘Zur Augustusstatue der Livia’, Römische Mitteilungen 25, 27–​55.  1926. Artemis und Iphigenie. Marmorgruppe der Ny Carlsberg Glyptothek. Leipzig: S. Hirzel. Swetnam-​Burland, M. 2015. Egypt in Rome. Visions of Egypt in Roman Imperial Culture. Cambridge: Cambridge University Press. Tanner, J. and Osborne, R. 2007. ‘Introduction: Art and Agency and Art History’, in R. Osborne and J. Tanner (eds), Art’s Agency and Art History. Malden, MA: Blackwell (New Interventions in Art History), 1–​27. Themelis, P. G. 1998–​ 1999. ‘Statuenfunde aus dem Gymnasion von Messene’, Nürnberger Blätter zur Archäologie 15: 59–​84.  2003. Das antike Messene. Athens: Ministry of Culture.   2013. ‘The Doryphoros of Messene’, in V. Franciosi and P. G. Themelis (eds), Pompei/​Messene. Il Doriforo e il suo contesto. Naples:  Università degli Studi Suor Orsola Benincasa, 125–​209. Thouvenot, R. 1927. Catalogue des figurines et objets de bronze du Musée Archéologique de Madrid, vol. I, Bronzes grecs et romains. Bordeaux and Paris: Feret and de Boccard. Tuchelt, K. 1979. Frühe Denkmäler Roms in Kleinasien. Beiträge zur archäologischen Überlieferung aus der Zeit der Republik und des Augustus, Teil 1, Roma und Promagistrate. Tübingen: Wasmuth (Istanbuler Mitteilungen. Beiheft 23). Treu, G. 1897. Die Bildwerke von Olympia in Stein und Thon, vol. III. Berlin: Asher & Co. Trimble, J. 2011. Women and Visual Replication in Roman Imperial Art and Culture. Cambridge: Cambridge University Press. Van Keuren, F., Cox, J. E., Attanasio, D., Prochaska, W., Herrmann, J. J. Jr. and Abramitis, D. H. 2015. ‘Parian Lychnites and the Badminton Sarcophagus in New York’, in P. Pensabene and E. Gasparini (eds), Asmosia X. Proceedings of the 10h International Conference of ASMOSIA. Rome: ‘L’Erma’ di Bretschneider, 403–​11. Vermeule, C. C. 1964. ‘Aphrodisiaca:  Satyr, Maenad and Eros. A  Romano-​ Hellenistic Marble Group of the Third Century AD in Boston’, in L. Freeman Sandler (ed.), Essays in Memory of Karl Lehmann. New York: Institute of Fine Arts, 359–​74.  1977. Greek Sculpture and Roman Taste. The Purpose and Setting of Graeco-​ Roman Art in Italy and the Greek Imperial East. Ann Arbor, MI: University of Michigan Press. Vermeule C. C. and Brauer, A. 1990. Stone Sculptures. The Greek, Roman and Etruscan Collections of the Harvard University Art Museums. Cambridge, MA: Harvard University Art Museums.

249

250

250

References Vermeule, C. C. and Neuerburg, N. 1973. Catalogue of the Ancient Art in the J. Paul Getty Museum. The Larger Statuary, Wall Paintings and Mosaics. Los Angeles: The J. Paul Getty Museum. Verri, G., Opper, T. and Deviese, T. 2000. ‘The Treu Head. A Case Study in Roman Sculptural Polychromy’, The British Museum. Technical Research Bulletin 4: 39–​54. Verri, G., Opper, T. and Lazzarini, L. 2014. ‘In picturae modum variata circumlitio? The Reconstruction of the Polychromy of a Roman Ideal Female Head (Treu Head)’, in P. Liverani and U. Santamaria (eds), Diversamente bianco: la policromia della scultura romana. Rome: Quasar, 149–​83. Vierneisel-​Schlörb, B. 1979. Glyptothek München. Katalog der Skulpturen, vols. I–​II, Klassische Skulpturen des 5. und 4. Jahrhunderts v. Chr. Munich: Beck. Visconti, E. Q. 1782–​96. Museo Pio-​Clementino. 7 vols. Rome: L. Mirri. Vlachogianni, E. 2012a. ‘Sculpture’, in A. Christopoulou, A. Gadolou, and P. Bouyia (eds), The Antikythera Shipwreck. The Technology of the Ship, the Cargo and the Mechanism. Athens: National Archaeological Museum, 38–​45.   2012b. ‘Sculpture’, in Kaltsas, Vlachogianni, and Bouyia 2012: 62–​79. von Blanckenhagen, P. H. 1957. ‘Narration in Hellenistic and Roman Art’, American Journal of Archaeology 61: 78–​83.   1969. ‘Laokoon, Sperlonga und Vergil’, Archäologischer Anzeiger 84: 427–​42.   1973. Review of:  G. Säflund, The Polyphemus and Scylla Groups at Sperlonga (Stockholm, 1972). American Journal of Archaeology 77: 456–​60.   1976. Review of:  B. Conticello and B.  Andreae, Die Skulpturen von Sperlonga (Berlin, 1974). American Journal of Archaeology 80: 99–​104. von den Hoff, R. 2004. ‘Horror and Amazement. Colossal Mythological Statue Groups and the New Rhetoric of Images in Late Second and Early Third Century Rome’, in B. E. Borg (ed.), Paideia: The World of the Second Sophistic. Berlin: De Gruyter, 105–​29. von Gerkan, A. and Krischen, F. 1928. Thermen und Palaestren. Berlin: Hans Schoetz & Co. (Milet. Ergebnisse der Ausgrabungen und Untersuchungen, vol. I.9). von Moock, D. W. 1998. Die figürlichen Grabstelen Attikas in der Kaiserzeit. Studien zur Verbreitung, Chronologie, Typologie und Ikonographie. Mainz am Rhein: P. von Zabern. Vorster, C. 1993a. Römische Skulpturen des späten Hellenismus und der Kaiserzeit, vol. I, Werke nach Vorlagen und Bildformeln des 5. und 4. Jahrhunderts v. Chr. Mainz am Rhein: P. von Zabern.   1993b. ‘Torsorepliken des Leidener Pan’, in Beck and Bol 1993: 193–​219.   2007. ‘Die Plastik des späten Hellenismus –​Porträts und rundplastische Gruppen’, in P. C. Bol (ed.), Die Geschichte der antiken Bildhauerkunst, vol. III, Klassische Plastik. Mainz am Rhein: P. von Zabern, 273–​331.   2012–​ 2013. ‘Spätantike Bildhauerwerkstätten in Rom. Beobachtungen zur Idealskulptur der Nachkonstantinischen Zeit’, Jahrbuch des Deutschen Archäologischen Instituts 127–​8: 393–​496.

251

References Waetzoldt, H. 2007. ‘The Use of Wool for the Production of Strings, Ropes, Braided Mats, and Similar Fabrics’, in C. Gillis and M.-​L. B. Nosch (eds), Ancient Textiles. Production, Craft and Society. Oxford: Oxbow, 112–​21. Waldhauer, O. 1936. Die antiken Skulpturen der Ermitage, vol. III. Berlin: De Gruyter. Wallace, W. E. 2015. ‘An Impossible Task’, in Helmstutler Di Dio 2015: 47–​58. Wardropper, I. 1999. ‘Cristoforo Stati’s Samson and the Lion. Florentine Style and Spanish Patronage’, Apollo 150.451: 30–​7.  2011. European Sculpture, 1400–​1900 in the Metropolitan Museum of Art. New York and New Haven: Metropolitan Museum of Art and Yale University Press. Waywell, G. B. 1978. The Free-​Standing Sculptures of the Mausoleum at Halicarnassus in the British Museum. London: British Museum.  1986. The Lever and Hope Sculptures. Berlin: Gebr. Mann.   1996. ‘Scilla nell’arte antica’, in B. Andreae and C. Parisi Presicce (eds), Ulisse. Il mito e la memoria. Rome: Progetti Museali Editore, 108–​19. Webb, R. 2008. ‘Inside the Mask: Pantomime from the Performers’ Perspective’, in E. Hall and R. Wyles (eds), New Directions in Ancient Pantomime. Oxford: Oxford University Press, 43–​60. Weber, M. 1996. ‘Zum griechischen Athletenbild. Zum Typus und zur Gattung des Originals der Apoxyomenosstatue im Vatikan’, Römische Mitteilungen 103: 31–​49. Weege, F. 1929. Der einschenkende Satyr aus Sammlung Mengarini. Berlin:  De Gruyter. Weinstock, H. 2012. Statuenstützen an vorkaiserzeitlichen Marmorstatuen. Ein Beitrag zur Chronologie der griechischen Plastik. Hamburg: Dr. Kovač. Weis, H. A. 2000. ‘Odysseus at Sperlonga. Hellenistic Hero or Roman Heroic Foil?’, in N. T. de Grummond and B. S. Ridgway (eds), From Pergamon to Sperlonga. Sculpture and Context. Berkeley, Los Angeles and London: University of California Press, 111–​65. Weitmann, P. 2012. ‘Das Messinstrument aus dem Hanghaus 2 in Ephesos  –​der älteste erhaltene Pantograph?’, Jahreshefte des Österreichischen Archäologischen Institutes 81: 331–​6. Welcker, F. G. 1849. ‘Der Myronische Diskuswerfer’, in Alte Denkmäler, vol. I, Die Giebelgruppen und andere Griechische Gruppen und Statuen. Göttingen: Dieterichsche Buchhandlung, 418–​29. Wenzel, C. H. 2005. An Introduction to Kant’s Aesthetics. Core Concepts and Problems. Malden, MA, and London: Blackwell. Wittkower, R. 1961. ‘Cornacchinis Reiterstatue Karls der Grossen in St. Peter’, in Miscellanea Bibliothecae Hertzianae. Munich: Schroll, 464–​73.  1997. Bernini. The Sculptor of the Roman Baroque. 4th edn. London: Phaidon.  1999. Art and Architecture in Italy 1600–​1750. 6th edn, revised by J. Connors and J. Montagu. 3 vols. New Haven: Yale University Press. Wixom, W. D. 1967. ‘Early Christian Sculptures at Cleveland’, Bulletin of the Cleveland Museum of Art 54.3: 67–​88.

251

252

252

References   1979. ‘Portrait Busts, Good Shepherd, and Jonah Figures’, in K. Weitzmann (ed.), Age of Spirituality Late Antique and Early Christian Art, Third to Seventh Century. New York: Metropolitan Museum of Art, 406–​10. Wollheim, R. 1980. Art and its Objects. 2nd edn. Cambridge: Cambridge University Press. Wrede, H. 1986. Die antike Herme. Mainz am Rhein: P. von Zabern (Trierer Beiträge zur Altertumskunde). Yfantidis, K. 1984. Die Polychromie der Hellenistischen Plastik, unpublished PhD thesis, University of Mainz. Younger, J. G. and Rehak, P. 2009. ‘Technical Observations on the Sculptures from the Temple of Zeus at Olympia’, Hesperia 78: 41–​105. Zanker, P. 1974. Klassizistische Statuen. Studien zur Veränderung des Kunstgeschmacks in der römischen Kaiserzeit. Mainz am Rhein: P. von Zabern.  1988. The Power of Images in the Age of Augustus. Ann Arbor, MI: University of Michigan Press.   1989. ‘Ein Augustusenkel als klassischer Pais’, in H.-​U. Cain, H. Gabelmann and D. Salzmann (eds), Festschrift für Nikolaus Himmelmann. Beiträge zur Ikonographie und Hermeneutik. Mainz am Rhein: P. von Zabern, 351–​4.   1992. ‘Nachahmen als kulturelles Schicksal’, in C. Lenz (ed.), Probleme der Kopie von der Antike bis zum 19. Jahrhundert:  vier Vorträge. Munich:  Bayerische Staatsgemäldesammlung, 9–​24.   1997. ‘In Search of the Roman Viewer’, in D. Buitron-​ Oliver (ed.), The Interpretation of Architectural Sculpture in Greece and Rome. Washington, DC: National Gallery of Art, 179–​91.  2002. Un’arte per l’impero. Funzione e intenzione delle immagini nel mondo romano. Milan: Electa. Zanker, P. and Ewald, B. C. 2012. Living with Myths. The Imagery of Roman Sarcophagi, tr. J. Slater. Oxford: Oxford University Press (= Mit Mythen leben. Die Bilderwelt der römischen Sarkophage. Munich: Hirmer, 2004). Zevi, F. (ed.) 1983. Baia. Il ninfeo imperiale sommerso di Punta Epitaffio. Benevento: Banca Sannitica. Zikos, D. 1996. ‘Antikenkopien in Bronze des Massimiliano Soldani Benzi’, in V. Krahn (ed.), Von  allen Seiten schön. Bronzen der Renaissance und des Barock. Rückblicke auf Ausstellung und Kolloquium, vol. I, Dokumentation zu Ausstellung und Kolloquium. Cologne: Letter-​Stiftung, 125–​38. Zimmer, K. B. 2014. Im Zeichen der Schönheit. Form, Funktion und Stellenwert klassischer Skulpturen im Hellenismus am Beispiel der Göttin Aphrodite. Rahden: Verlag Marie Leidorf.

253

Index

Achilles Tatius The Story of Leucippe and Clitophon, 66 Addison, Joseph, 28 Amazon type Sciarra /​Berlin /​Lansdowne, 48, 95, 97–​8 Aphrodisias, Bouleuterion, 207–​8 Aphrodite Capitoline Venus, 28 Cnidian Aphrodite, 14, 28, 49, 62, 181 Medici Venus, 27, 28–​9 Apollo Sauroctonos, 42, 49 Apoxyomenos, 14, 34, 89 Ephesian type, 157 Vatican, 14–​15, 75, 82, 114, 115, 117, 157, 165 Apuleius Apologia, 176 The Golden Ass, 3–​4 Aristotle Parts of Animals, 4 Arrian Periplus, 118 Artemis and Iphigenia, Copenhagen, 10, 56–​8, 96 Artemis from Pompeii, 99–​101 Athena Varvakeion, 48–​9 Athens Acropolis, 45, 46 agora, 128 Kallithea monument, 52–​4 Kerameikos, 29 Augustus from Prima Porta, 29, 38–​9, 95 Baiae, nymphaeum of Punta Epitaffio, 66, 70, 82, 83, 164, 183–​5 Bartman, Elizabeth, 145, 193, 194 Bernini, Gianlorenzo, 1–​2, 11, 15, 105 Boy Strangling a Goose, 14, 18–​19 bronze, 108–​11 Bunraku, 199 Caesar The Civil Wars, 5 Callimachus Aetia, 96

Hymn to Apollo, 96 Hymn to Delos, 96 Canina, Luigi, 14 Canova, Antonio, 11 Capestrano Warrior, 33 Carradori, Francesco, 148 Cassiodorus Variae, 168 Castel Gandolfo, Villa of Domitian, 69, 87, 143–​4, 149, 154, 182 Cellini, Benvenuto, 8 Cicero Brutus, 175 Letters to Atticus, 118 On Divination, 175 Cole, Michael, 105 coloured stones, 167 Columella On Agriculture, 5 Cornacchini, Agostino, 1–​2 dance, 218–​19 Dancing Faun Athens, Kerameikos, 29 Florence, Uffizi, 27, 29 Rome, Palazzo Corsini, 29, 128 de Vries, Adriaen, 108 Delos, 36, 54, 55, 96, 120 Diadoumenus. See Diadoumenus from Delos group of Aphrodite, Pan and Eros, 54–​5, 98–​100 pseudo athlete, 55, 82, 84–​6 Delphi, Daochos monument, 51–​4 depiction of struts, 208–​10 Derrida, Jacques, 19–​20, 139 Diadoumenus, 34, 89, 126–​7, 144–​5 from Delos, 34–​6, 61, 95–​6, 144 New York, 74, 144 Digest, 119 Dio Chrysostom Olympic Discourse, 175 Diodorus of Sicily Library of History, 4 Dirce bound to the Bull, 169, 170

253

254

254

Index Discobolus, 34, 78, 91, 136–​8, 219 Castel Porziano, 68, 75, 82, 114, 207 Lancellotti, 5, 68, 147 Discophorus, 34 Doryphorus, 34, 190, 216 Dresden Artemis, 70, 138, 159 Dresden Youth, 143–​4, 146–​7, 149, 182 ephemeral features, 204–​5 equestrian statue. See horse ex uno lapide /​ ex eodem lapide. See monolithic sculpture focal awareness /​subsidiary awareness, 217 fountain, 18, 29, 66 frames, 19, 129–​32, 139, 208 Gell, Alfred, 213 genital strut, 82–​3 Geominy, Wilfred, 22, 138, 142, 145, 178 Giambologna, 114–​15 gilding, 36, 95–​6 Gombrich, Ernst, 218 Guattani, Giuseppe Antonio, 5, 78, 107 hair strut, 83, 194 Hermes and the infant Dionysus, 9, 13, 14, 15, 49–​51 Hollinshead, Mary B., 23, 83, 132 horse, 1–​2, 45, 55, 97 Houdon, Jean-​Antoine, 12 interdigital strut, 78–​82, 109–​11, 194 joins, 75, 120–​2, 150, 159–​64, 187, 193 advantages, 121, 125–​6, 160 history, 120–​1, 165 Kant, Immanuel, 139 ergon and parergon, 19–​20 korai, 45, 120 kouroi, 45–​6 Kritios Boy, 46 Laocoon, 168–​9, 170–​1 Large Herculaneum Woman, 88, 89, 187 late antique sculpture, 135–​4, 140–​39, 194–​8 Leochares, 5 Libanius Orations, 219 lifting technologies, 125 Lucian The Lover of Lies, 78, 219 Lycurgus Cup, 196–​7 Lysippus, 14, 34, 39, 75, 157

Maffei, Paolo Alessandro, 105 marble Aphrodisias, 95, 191, 195 Docimium, 118, 122, 170, 172, 181, 197 Greek islands, 91, 195 Luna /​Carrara, 152, 179, 182, 190, 196 Parian, 34, 46, 55, 62, 80, 118, 122, 191, 196 Pentelic, 62, 117, 159, 181 Proconnesian, 182, 191 provenance, 185, 189, 190 Mausoleum of Halicarnassus, 45, 120 Maviglia, Ada, 6, 33 Medici Venus. See Aphrodite Michelangelo, 170, 179 Miletus, Baths of Faustina Melpomene, 20–​2, 70, 89, 96 Venus, 75, 160–​4 miniatures, see small-​scale sculpture mirror copy, 154, 159 Mochi, Francesco, 10–​11 monolithic sculpture, 126, 159, 160, 168–​71, 203 Myron, 34, 75, 219 nape strut. See neck strut naturalism, 216–​17 neck strut, 20, 66, 88–​9, 144, 187, 188–​90, 196 Nollekens, Joseph, 11–​12 Nonnus of Panopolis Dionysiaca, 38 Olympia, nymphaeum of Herodes Atticus, 181, 185–​7, 188 Ovid Metamorphoses, 221 pantomime. See dance Pausanias Description of Greece, 49 pedimental sculpture, 44 pendants, 154, 182–​3 Philostratus the Elder Images, 32, 176 Philostratus the Younger Life of Apollonios of Tyana, 118 photography, 15 piecing. See joins plinths, 116, 134, 210 Pliny the Elder Natural History, 5, 96, 157, 168–​9, 175, 203–​4 Pliny the Younger Letters, 5 Plutarch Simpotic Questions, 218–​19

255

Index pointing technique, 146–​7, 148–​9 Polanyi, Michael, 217 polychromy, 16, 94–​5, 96–​101, 177 Polyclitus, 34, 96, 143, 144 Pontormo, 9 portraits, 88–​9, 181, 186–​7 Pouring Satyr, 33, 69, 87, 133–​4, 147, 149–​55, 182 Palermo, 13, 63, 87, 150, 154–​5, 213 Praxiteles, 9, 28, 42, 49, 149, 181 Pseudo-​Lucian Amores, 181 Quintilian Institutes of Oratory, 78, 175 Resting Satyr, 42–​3, 193 restoration ancient, 161–​4, 207–​8 modern, 5, 14–​15, 59, 61, 78, 86, 91, 112, 134, 146–​7 Rhodes, 109, 120, 122, 169, 171 Rome, 1, 39, 179 ancient cityscape, 168 Basilica of San Paolo fuori le Mura, 10 Basilica of Sta Maria sopra Minerva, 10 Baths of Agrippa, 157 Baths of Caracalla, 91, 170 found in, 70, 133, 149, 152, 190, 196 Horti Sallustiani, 39, 57 monumenta of Asinius Pollio, 169 produced in, 91, 191, 196 Temple of Apollo Sosianus, 44 Villa of the Quintilii, 18–​19 sarcophagi, 68, 91–​2, 191, 209–​12 schema, 218–​19 Seneca Moral Letters to Lucilius, 215 shipping, 206–​7

shipwreck Antikythera, 55–​7, 69, 80, 116–​17, 118, 121–​2, 125, 183 Mahdia, 116–​17, 118 signature, 28, 80, 122, 150, 171, 182 Silenus with the infant Dionysus, 39–​42 Simmel, Georg, 132 small-​scale sculpture, 88, 134, 140–​39, 188, 191–​6, 203 Soldani Benzi, Massimiliano, 27–​8 Sperlonga group of Polyphemus, 69, 78–​80, 97, 112–​14 group of Scylla, 171, 177 sculptural groups, 16, 79–​8, 97, 123, 171–​5, 181 spiral strut, 90–​2, 132–​4, 152, 190–​1 statuesque, 129, 217, 221 Stewart, Peter, 129, 134, 217 storage, 116 Strabo Geography, 170 Studniczka, Franz, 57, 96, 100 terminology Greek and Latin, 3–​5 modern, 8 Three Graces, 13 from Cyrene, 61–​2, 143 Thucydides The Peloponnesian War, 4 Toro Farnese. See Dirce bound to the Bull Treu, Georg, 13–​14, 15 Tyrant Slayers, 128–​9 Visconti, Ennio Quirino, 27 Vitruvius On Architecture, 5 von Blanckenhagen, Peter, 217 Wollheim, Richard, 218 wounded warrior, so-​called Protesilaos, 63, 75, 134–​7 Xenophon Memorabilia, 175 On Horsemanship, 4

255

256