Reading and Mapping Fiction: Spatialising the Literary Text 9781108766876, 9781108487450, 2020009168, 2020009169, 9781108720304

Citation preview

Reading and Mapping Fiction

Do we map as we read? How central to our experience of literature is the way in which we spatialise and visualise a fictional world? Reading and Mapping Fiction offers a fresh approach to the interpretation of literary space and place centred upon the emergence of a fictional map alongside the text in the late nineteenth and early twentieth centuries. Bringing together a range of new and emerging theories, including cognitive mapping and critical cartography, Bushell compellingly argues that this activity, whatever it is called – mapping, diagramming, visualising, spatialising – is a vital and intrinsic part of how we experience literature, and of what makes it so powerful. Drawing on both the theory and history of literature and cartography, this richly illustrated study opens up understanding of spatial meaning and interpretation in new ways that are relevant to both more traditional academic scholarship and to newly emerging digital practices.

  is Professor of Romantic and Victorian Literature in the Department of English Literature & Creative Writing, Lancaster University. She is interested in mapping texts in a range of ways (across process; empirically; digitally). She is also principal investigator on the Arts and Humanities Research Council (AHRC)-funded literary mapping project, Chronotopic Cartographies.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Reading and Mapping Fiction Spatialising the Literary Text   Lancaster University

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

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/9781108487450 DOI: 10.1017/9781108766876 © Sally Bushell 2020 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 2020 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: Bushell, Sally, author. Title: Reading and mapping fiction : spatialising the literary text / Sally Bushell, Lancaster University. Description: Cambridge ; New York : Cambridge University Press, 2020. | Includes bibliographical references and index. Identifiers: LCCN 2020009168 (print) | LCCN 2020009169 (ebook) | ISBN 9781108487450 (hardback) | ISBN 9781108720304 (paperback) | ISBN 9781108766876 (epub) Subjects: LCSH: Geographical perception in literature. | English fiction–19th century–History and criticism. | English fiction–20th century–History and criticism. | Maps in literature. | Imaginary places in literature. | Books and reading. Classification: LCC PR778.G46 B87 2020 (print) | LCC PR778.G46 (ebook) | DDC 820.9/32–dc23 LC record available at https://lccn.loc.gov/2020009168 LC ebook record available at https://lccn.loc.gov/2020009169 ISBN 978-1-108-48745-0 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.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Reading and Mapping Fiction

Do we map as we read? How central to our experience of literature is the way in which we spatialise and visualise a fictional world? Reading and Mapping Fiction offers a fresh approach to the interpretation of literary space and place centred upon the emergence of a fictional map alongside the text in the late nineteenth and early twentieth centuries. Bringing together a range of new and emerging theories, including cognitive mapping and critical cartography, Bushell compellingly argues that this activity, whatever it is called – mapping, diagramming, visualising, spatialising – is a vital and intrinsic part of how we experience literature, and of what makes it so powerful. Drawing on both the theory and history of literature and cartography, this richly illustrated study opens up understanding of spatial meaning and interpretation in new ways that are relevant to both more traditional academic scholarship and to newly emerging digital practices.

  is Professor of Romantic and Victorian Literature in the Department of English Literature & Creative Writing, Lancaster University. She is interested in mapping texts in a range of ways (across process; empirically; digitally). She is also principal investigator on the Arts and Humanities Research Council (AHRC)-funded literary mapping project, Chronotopic Cartographies.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Reading and Mapping Fiction Spatialising the Literary Text   Lancaster University

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

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/9781108487450 DOI: 10.1017/9781108766876 © Sally Bushell 2020 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 2020 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: Bushell, Sally, author. Title: Reading and mapping fiction : spatialising the literary text / Sally Bushell, Lancaster University. Description: Cambridge ; New York : Cambridge University Press, 2020. | Includes bibliographical references and index. Identifiers: LCCN 2020009168 (print) | LCCN 2020009169 (ebook) | ISBN 9781108487450 (hardback) | ISBN 9781108720304 (paperback) | ISBN 9781108766876 (epub) Subjects: LCSH: Geographical perception in literature. | English fiction–19th century–History and criticism. | English fiction–20th century–History and criticism. | Maps in literature. | Imaginary places in literature. | Books and reading. Classification: LCC PR778.G46 B87 2020 (print) | LCC PR778.G46 (ebook) | DDC 820.9/32–dc23 LC record available at https://lccn.loc.gov/2020009168 LC ebook record available at https://lccn.loc.gov/2020009169 ISBN 978-1-108-48745-0 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.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Reading and Mapping Fiction

Do we map as we read? How central to our experience of literature is the way in which we spatialise and visualise a fictional world? Reading and Mapping Fiction offers a fresh approach to the interpretation of literary space and place centred upon the emergence of a fictional map alongside the text in the late nineteenth and early twentieth centuries. Bringing together a range of new and emerging theories, including cognitive mapping and critical cartography, Bushell compellingly argues that this activity, whatever it is called – mapping, diagramming, visualising, spatialising – is a vital and intrinsic part of how we experience literature, and of what makes it so powerful. Drawing on both the theory and history of literature and cartography, this richly illustrated study opens up understanding of spatial meaning and interpretation in new ways that are relevant to both more traditional academic scholarship and to newly emerging digital practices.

  is Professor of Romantic and Victorian Literature in the Department of English Literature & Creative Writing, Lancaster University. She is interested in mapping texts in a range of ways (across process; empirically; digitally). She is also principal investigator on the Arts and Humanities Research Council (AHRC)-funded literary mapping project, Chronotopic Cartographies.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Reading and Mapping Fiction Spatialising the Literary Text   Lancaster University

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

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/9781108487450 DOI: 10.1017/9781108766876 © Sally Bushell 2020 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 2020 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: Bushell, Sally, author. Title: Reading and mapping fiction : spatialising the literary text / Sally Bushell, Lancaster University. Description: Cambridge ; New York : Cambridge University Press, 2020. | Includes bibliographical references and index. Identifiers: LCCN 2020009168 (print) | LCCN 2020009169 (ebook) | ISBN 9781108487450 (hardback) | ISBN 9781108720304 (paperback) | ISBN 9781108766876 (epub) Subjects: LCSH: Geographical perception in literature. | English fiction–19th century–History and criticism. | English fiction–20th century–History and criticism. | Maps in literature. | Imaginary places in literature. | Books and reading. Classification: LCC PR778.G46 B87 2020 (print) | LCC PR778.G46 (ebook) | DDC 820.9/32–dc23 LC record available at https://lccn.loc.gov/2020009168 LC ebook record available at https://lccn.loc.gov/2020009169 ISBN 978-1-108-48745-0 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.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

To my son, Dylan, who has shared his whole life with this book.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Contents

List of Illustrations [ix] Acknowledgements [xv]

Introduction [1] 1 A Shifting Relationship: From Literary Geography to Critical Literary Mapping [16] The Origins of Literary Geography [16] Literary Geography or Literary Cartography? Critical Cartography [28] Critical Literary Mapping [33] Material Juxtaposition: Map and Text [36]

[25]

2 Historicising the Fictional Map [44] Early Pictorial Maps [45] Mapping Utopia; Mapping Dystopia [49] Mapping the New World [57] Itinerary Maps: Paris, Ogilby, Bunyan [66] Accurate Maps: Mapping the Nation [74] Accurate Maps: Mapping Empire [81] Ending Where We Began [87]

3 Doubleness and Silence in Adventure and Spy Fiction [92] Trusting the Fictional Map: Accuracy and Use Value [94] Cartographic Silence [104] The Authenticity of the Fictional Map [108] From Adventure Story to Spy Fiction [114] Integrative Cartography in The Riddle of the Sands [116] Double Meanings; Double Intentions [123]

4 Mapping Murder

[127]

The Origins of Maps in Detective Fiction [128] Mapping Crime Scenes [130] Theorising Detection [138] Trusting and Not Trusting the Map [143] Human Geometry [151]

vii

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

viii

Contents

5 Playspace: Spatialising Children’s Fiction

[164]

Spatialising Children’s Fiction [165] The Lakes As Playspace [169] A Material and Visual Playspace [175] Referential or Non-referential? [181] Returning to Rousseau [186] Mapping Negative Playspace [191]

6 Mapping Worlds: Tolkien’s Cartographic Imagination [199] What to Do First [199] ‘Each Is Both Prior to the Other and Later Than It’ [203] Writing and Mapping: Rivers and Roads [206] Multiple Mapping in The Lord of the Rings [220] Mapping and Not Mapping [226] Post-authorial Re-mapping [228]

7 Fearing the Map: Representational Priorities and Referential Assumptions [237] Reasons for the Absence of Maps [239] Fearing Illustration [244] Realist Principles and the Absent Map [248] Why a Map Is Not an Illustration [250] Mapping Realism: Trollope [253] Mapping Realism: Hardy [262] Hardy and Trollope [265] The Earliest Map: Return of the Native [267]

8 Reading As Mapping, or, What Cannot Be Visualised Spatialising Reader-Response Theory [277] Conscious Memory Mapping [279] Two Forms for the Internal Map: Route and Locale The Cognitive Mapping of Literature [290] Mapping Literature in Digital Space [296] ‘Let Us Pretend That It Is the End’ [305]

[273]

[283]

Bibliography [307] Index [327] Colour plates can be found between pages 144 and 145

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Illustrations

Cover illustration

Map from Arthur Ransome, Swallows and Amazons (Illustration by Stephen Spurrier). © The British Library Board (NN.16780). By kind permission of Random House Publishers.

Figures 2.1

Hereford Mappamundi from Hereford Cathedral Library and Archives. © Hereford Cathedral. Reproduced with the permission of Hereford Cathedral. [48] 2.2 First edition map of Utopia from Thomas More, Libellus vere aureus nec minus salutaris quam festiuus de optimo reip. statu, deq[ue] noua insula Vtopia ([Louvain]: 1516). Beinecke Rare Book and Manuscript Library, Yale University. (If M81 r516). [51] 2.3 Fourth edition map of Utopia from Thomas More, De optimo reip[ublicae] statu, deque noua insula Vtopia (Basel: 1518). Beinecke Rare Book and Manuscript Library, Yale University. (If M81 r518). [52] 2.4 World map from Joseph Hall, Mundus Alter et Idem (London: [1605]). © The British Library Board. (684.d.5, opposite f. 20). [57] 2.5 ‘Illustration of the Bashee Islands’ from William Dampier, A New Voyage Round the World (London: James Knapton, 1697) p. 385. © The British Library Board. (1045.f.22). [61] 2.6 ‘Map of Laputa’ from Jonathan Swift, Travels into Several Remote Nations of the World in Four Parts by Lemuel Gulliver (London: 1726) Vol. 2, Part 3, Plate 4, p. 39. © The British Library Board. (1045.f.22). [62] 2.7 ‘Map of Crusoe’s Island’ from Daniel Defoe, Serious Reflections during the Life and Surprising Adventures of Robinson Crusoe (London: W. Taylor, 1720). Beinecke Rare Book and Manuscript Library, Yale University. (Defoe 50 719 3). [66] 2.8 ‘Part of an itinerary from London to Jerusalem, covering the journey from London to Beauvais, with representations of the main towns’ from Matthew Paris, Historia Anglorum III: Chronica majora (1250–9). © The British Library Board. (Royal 14 C. VII, f. 2). [69]

ix

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

x

List of Illustrations

2.9

2.10

2.11

2.12 2.13

3.1

3.2

3.3

3.4

3.5

3.6 4.1

4.2

‘The Road from London to Hith’ from John Ogilby, Britannia, vol. I (1675), p. 149. Beinecke Rare Book and Manuscript Library, Yale University. (EEdea 675P). [71] Map from John Bunyan, The Pilgrim's Progress. . . A New Edition . . . To which are now added, notes by W. Mason (London: H. Trapp; A. Hogg, 1778). © The British Library Board. (1609/2471). [71] ‘The Hound of the Baskervilles, Chapter 3: The Problem’ from The Strand, vol. 22 no. 129 (September 1901), page 247. © The British Library Board. (P.P.6004.glk). [81] Map from Robert Louis Stevenson, Treasure Island (London: Cassell & Company, 1883). © The British Library Board. (C.71.c.18). [88] ‘Portolan chart of the Mediterranean Sea . . . as far south as Sierra Leone’ Jorge de Aguiar (1492). Beinecke Rare Book and Manuscript Library, Yale University. (Art Storage 30cea 1492). [89] Map from H. Rider Haggard, King Solomon’s Mines (London: Cassell & Company Ltd, 1885) p. 27. Widener Library, Harvard University. (21484.4.2). [102] Fold-out map from H. Rider Haggard, King Solomon’s Mines (London: Cassell & Company Ltd, 1885). Widener Library, Harvard University. [111] Map from John Buchan, Prester John (Boston: Houghton Mifflin Company, 1910) p. 10. Public domain image via Wikimedia Commons. [115] ‘General Map’ from Erskine Childers, The Riddle of the Sands (London: Smith, Elder & Co., 1903). Fold-out frontispiece map © The British Library Board. (Y 155.C43). [120] ‘Chart Illustrating the Stranding of the Dulcibella’ from Erskine Childers, The Riddle of the Sands, p. 69. © The British Library Board. (Y 155.C43). [121] ‘Sketch Map for Journey in the Fog’ from Erskine Childers, The Riddle of the Sands, p.233. © The British Library Board. (Y 155.C43). [122] ‘Map of the area in Hampstead where Thomas Henry Hocker robbed and murdered James Delarue’ from Illustrated London News (12 April 1845), p. 230. © Illustrated London News Ltd/Mary Evans. [132] ‘Plan of the basement of the house at no. 3 Miniver Place, Bermondsey, where the body of Patrick O’Connor was discovered under flag stones’ from Illustrated London News (8 September 1849), p. 168. © Illustrated London News Ltd/Mary Evans. [133]

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

List of Illustrations

4.3

4.4

4.5 4.6

4.7

4.8

4.9

4.10

4.11

4.12

4.13

xi

‘Tragedy at Reading’ from Illustrated Police News Issue 779 (18 January 1879). Newspaper image © The British Library Board. All rights reserved. With thanks to The British Newspaper Archive. [134] Map from Charles Warren Adams, The Notting Hill Mystery (London: Saunders, Otley, and Co, 1865). © The British Library Board. (12631. aaa.11). [136] Map from Émile Gaboriau, Monsieur Lecoq (Paris: Dentu, 1869). Public domain image via Wikimedia Commons. [137] House plan from Agatha Christie, The Mysterious Affair at Styles (New York; London: John Lane Co., 1920) p. 42. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1920). © The British Library Board. (Cup.410.f.420). [145] Room plan from Agatha Christie, The Mysterious Affair at Styles, p. 60. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1920). © The British Library Board. [145] House plan from Agatha Christie, The Murder of Roger Ackroyd, p. 63 (London: W. Collins, Sons & Co., 1926). Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1926). © The British Library Board. (Cup.410.f.554). [149] Room plan from Agatha Christie, The Murder of Roger Ackroyd, p. 91. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1926). © The British Library Board. (Cup.410.f.554). [150] Plan A from Agatha Christie, The Murder at the Vicarage (London: W. Collins, Sons & Co., [1930]) p. 27. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) ([1930]). © The British Library Board. (W 17328). [156] Plan B from Agatha Christie, The Murder at the Vicarage, p. 44. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) ([1930]). © The British Library Board. (W 17328). [156] Plan C from Agatha Christie, The Murder at the Vicarage, p. 71. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) ([1930]). © The British Library Board. (W 17328). [157] Street plan from Margery Allingham, More Work for the Undertaker (London: William Heinemann, 1948). The Margery Allingham/Philip Youngman Carter Collection, Albert Sloman Library, University of Essex. Reprinted by permission of Peters Fraser & Dunlop (www .petersfraserdunlop.com) on behalf of Rights Limited. Image courtesy of the Albert Sloman Library, University of Essex. [160]

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

xii

List of Illustrations

5.1

Endpaper Map from Arthur Ransome, Swallows and Amazons (London: Jonathan Cape, 1930). © The British Library Board. By kind permission from Random House Ltd. [176] 5.2 Endpaper Map from Arthur Ransome, Swallowdale (London: Jonathan Cape, 1931). © The British Library Board. By kind permission from Random House Ltd. [177] 5.3 Endpaper Map from Arthur Ransome, Winter Holiday (London: Jonathan Cape, 1933). © The British Library Board. By kind permission from Random House Ltd. [179] 5.4 Detail from map for Arthur Ransome, Swallowdale (London: Jonathan Cape, 1931). © The British Library Board. By kind permission from Random House Ltd. [180] 5.5 Detail from map for Arthur Ransome, Winter Holiday (London: Jonathan Cape, 1933). © The British Library Board. By kind permission from Random House Ltd. [180] 5.6 ‘Peter Pan’s Map of Kensington Gardens’ from J. M. Barrie, Peter Pan in Kensington Gardens (London: Hodder & Stoughton, 1906). Beinecke Rare Book and Manuscript Library, Yale University. (GEN MSS 1400). [192] 6.1 Map for ‘A Part of The Shire’ from J. R. R. Tolkien, The Fellowship of the Ring: Being the First Part of The Lord of the Rings (Oxford: George Allen & Unwin, 1954). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1954). © The British Library Board. (RF.2017.a.9). [208] 6.2 ‘First sketch map of The Shire, [1938]’; MS. Tolkien Drawings 104r. Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd, 1998. [209] 6.3 ‘Sketch map of roads and rivers in The Shire, [1938]’. MS. Tolkien Drawings 105r, Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd, 2015. [210] 6.4 Back endpaper ‘Map of Wilderland’ from J. R. R. Tolkien, The Hobbit, or There and Back Again (London: George Allen & Unwin, 1937). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1937). (Cup.410.f.378). [210] 6.5 Christopher Tolkien’s ‘General Map for Middle Earth’ from J. R. R. Tolkien, The Fellowship of the Ring: Being the First Part of The Lord of the Rings (Oxford: George Allen & Unwin, 1954). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1954). © The British Library Board. (RF.2017.a.9). [211]

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

List of Illustrations

6.6

6.7

6.8

6.9

6.10

6.11

6.12

6.13

7.1

xiii

Working sketch map: ‘Sketch Map of the Misty Mountains and the Lands South and East [1939]’. MS. Tolkien Drawings 115r. Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd 1998, 2015. [214] Comparing ‘Carrock’ on the maps (created by author). Details from maps given at: Figure 4 ‘Map of Wilderland’; Figure 6: Working Sketch Map c.1939; ‘Map of the north-west of Middle Earth, [c.1948]’; MS. Tolkien Drawings 124r, Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd 2015; Figure 5 ‘Christopher Tolkien’s General Map for Middle Earth’. [215] ‘Sketch of part of the Misty Mountains showing Dimrill Dale, [1940]’. MS. Tolkien Drawings 90, fol. 11r. Bodleian Library, Oxford, UK. Copyright: The Tolkien Trust 2015. [216] ‘Redrafting of topography and description for Minas Tirith’, MSS 3/1/ 24/3b, J. R. R. Tolkien Collection, Special Collections and University Archives, Raynor Memorial Libraries, Marquette University, Wisconsin. Image reproduced with the kind permission of Marquette University and the Tolkien estate. [219] ‘Christopher Tolkien’s Reconstruction of Tolkien’s Working Practices; An Annotated Version’, from John Krygier and Denis Wood, Making Maps, A Visual Guide to Map Making 2nd ed. (The Guilford Press, 2011) p. 72. [222] Christopher Tolkien's ‘Map of Gondor, Rohan and Mordor’ from J. R. R. Tolkien, The Return of the King: Being the Third Part of The Lord of the Rings (London: George Allen & Unwin, 1955). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1955). © The British Library Board. (RF.2017.a.9). [223] ‘Map 2: Journey from Hobbiton to Bywater’, from Barbara Strachey, Journeys of Frodo: An Atlas of J. R. R. Tolkien’s The Lord of the Rings (London: George Allen & Unwin, 1981) p. 12. Reprinted by permission of HarperCollins Publishers Ltd © (Barbara Strachey) (1981). Reprinted by permission of Camphill Village Trust. © The British Library Board. (Maps 203.e.7). [232] ‘Map of the Mines of Moria’, from Karen Wynn Fonstad, The Atlas of Middle-Earth (London: Grafton, 1992) p. 128. Reprinted by permission of HarperCollins Publishers Ltd © (Karen Wynn Fonstad) (1981). © The British Library Board. (Maps 204.d.49). [234] ‘The thriving City of Eden as it appeared on paper’. Illustration from Charles Dickens, The Life and Adventures of Martin Chuzzlewit

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

xiv

List of Illustrations

7.2

7.3

7.4

7.5

7.6

(London: Chapman and Hall, 1844) p. 268. Public domain image courtesy of Hathi Trust Digital Library. https://hdl.handle.net/2027/uc2 .ark:/13960/t4vh5jp66?urlappend=%3Bseq=326, accessed 17 April 2019. [244] Choropleth map of four Trollope novels mapped onto Michael Sadleir’s Map of Barsetshire (created by author). Michael Sadleir, Trollope: A Commentary (London: Constable & Co., 1945) p. 163. © The British Library Board. (10861.aa.25). [257] ‘Anthony Trollope's Sketch Map of Barsetshire’, from Michael Sadleir, Trollope: A Commentary (London: Constable & Co., 1927) facing p. 154. (010856.cc.18). [260] Embryo Map by George Eliot, from Quarry for Middlemarch Notebook; MS Lowell 13, f. 57r, Houghton Library, Harvard University. Image reproduced by kind permission of Mr Jonathan G. Ouvry. [262] ‘The Stakes were won by Wildeve’, Illustration by Arthur Hopkins, from Belgravia: A London Magazine, Volume 36, Issue 141 (July 1878), opposite title page. © The British Library Board. (P.P.6004.gn). [269] Frontispiece Map from Thomas Hardy, The Return of the Native (London: Smith, Elder, & Co., 1878) Vol. 1. © The British Library Board. (C.71.h.7). [270]

Acknowledgement of Prior Publications In relation to the discussion of paratext in Chapter 1 and of imagetext in Chapter 5: Sally Bushell, ‘Imagetext or Paratext: Theorising the Literary Map’, Word & Image 32.2 (Summer 2016): 181–94. In relation to the discussion of Adventure Fiction in Chapter 3: Sally Bushell, ‘The Map in Victorian Adventure Fiction: Doubleness, Silence and the Ur-Map in Treasure Island and King Solomon’s Mines’, Victorian Studies 57.4 (2015): 611–37. In relation to the discussion of Agatha Christie’s The Murder of Roger Ackroyd in Chapter 4: Sally Bushell, ‘“The Slipperiness of Literary Maps”: Critical Cartography and Literary Cartography’, Cartographica 47.3 (Fall 2012): 49–61.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Illustrations

Cover illustration

Map from Arthur Ransome, Swallows and Amazons (Illustration by Stephen Spurrier). © The British Library Board (NN.16780). By kind permission of Random House Publishers.

Figures 2.1

Hereford Mappamundi from Hereford Cathedral Library and Archives. © Hereford Cathedral. Reproduced with the permission of Hereford Cathedral. [48] 2.2 First edition map of Utopia from Thomas More, Libellus vere aureus nec minus salutaris quam festiuus de optimo reip. statu, deq[ue] noua insula Vtopia ([Louvain]: 1516). Beinecke Rare Book and Manuscript Library, Yale University. (If M81 r516). [51] 2.3 Fourth edition map of Utopia from Thomas More, De optimo reip[ublicae] statu, deque noua insula Vtopia (Basel: 1518). Beinecke Rare Book and Manuscript Library, Yale University. (If M81 r518). [52] 2.4 World map from Joseph Hall, Mundus Alter et Idem (London: [1605]). © The British Library Board. (684.d.5, opposite f. 20). [57] 2.5 ‘Illustration of the Bashee Islands’ from William Dampier, A New Voyage Round the World (London: James Knapton, 1697) p. 385. © The British Library Board. (1045.f.22). [61] 2.6 ‘Map of Laputa’ from Jonathan Swift, Travels into Several Remote Nations of the World in Four Parts by Lemuel Gulliver (London: 1726) Vol. 2, Part 3, Plate 4, p. 39. © The British Library Board. (1045.f.22). [62] 2.7 ‘Map of Crusoe’s Island’ from Daniel Defoe, Serious Reflections during the Life and Surprising Adventures of Robinson Crusoe (London: W. Taylor, 1720). Beinecke Rare Book and Manuscript Library, Yale University. (Defoe 50 719 3). [66] 2.8 ‘Part of an itinerary from London to Jerusalem, covering the journey from London to Beauvais, with representations of the main towns’ from Matthew Paris, Historia Anglorum III: Chronica majora (1250–9). © The British Library Board. (Royal 14 C. VII, f. 2). [69]

ix

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

x

List of Illustrations

2.9

2.10

2.11

2.12 2.13

3.1

3.2

3.3

3.4

3.5

3.6 4.1

4.2

‘The Road from London to Hith’ from John Ogilby, Britannia, vol. I (1675), p. 149. Beinecke Rare Book and Manuscript Library, Yale University. (EEdea 675P). [71] Map from John Bunyan, The Pilgrim's Progress. . . A New Edition . . . To which are now added, notes by W. Mason (London: H. Trapp; A. Hogg, 1778). © The British Library Board. (1609/2471). [71] ‘The Hound of the Baskervilles, Chapter 3: The Problem’ from The Strand, vol. 22 no. 129 (September 1901), page 247. © The British Library Board. (P.P.6004.glk). [81] Map from Robert Louis Stevenson, Treasure Island (London: Cassell & Company, 1883). © The British Library Board. (C.71.c.18). [88] ‘Portolan chart of the Mediterranean Sea . . . as far south as Sierra Leone’ Jorge de Aguiar (1492). Beinecke Rare Book and Manuscript Library, Yale University. (Art Storage 30cea 1492). [89] Map from H. Rider Haggard, King Solomon’s Mines (London: Cassell & Company Ltd, 1885) p. 27. Widener Library, Harvard University. (21484.4.2). [102] Fold-out map from H. Rider Haggard, King Solomon’s Mines (London: Cassell & Company Ltd, 1885). Widener Library, Harvard University. [111] Map from John Buchan, Prester John (Boston: Houghton Mifflin Company, 1910) p. 10. Public domain image via Wikimedia Commons. [115] ‘General Map’ from Erskine Childers, The Riddle of the Sands (London: Smith, Elder & Co., 1903). Fold-out frontispiece map © The British Library Board. (Y 155.C43). [120] ‘Chart Illustrating the Stranding of the Dulcibella’ from Erskine Childers, The Riddle of the Sands, p. 69. © The British Library Board. (Y 155.C43). [121] ‘Sketch Map for Journey in the Fog’ from Erskine Childers, The Riddle of the Sands, p.233. © The British Library Board. (Y 155.C43). [122] ‘Map of the area in Hampstead where Thomas Henry Hocker robbed and murdered James Delarue’ from Illustrated London News (12 April 1845), p. 230. © Illustrated London News Ltd/Mary Evans. [132] ‘Plan of the basement of the house at no. 3 Miniver Place, Bermondsey, where the body of Patrick O’Connor was discovered under flag stones’ from Illustrated London News (8 September 1849), p. 168. © Illustrated London News Ltd/Mary Evans. [133]

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

List of Illustrations

4.3

4.4

4.5 4.6

4.7

4.8

4.9

4.10

4.11

4.12

4.13

xi

‘Tragedy at Reading’ from Illustrated Police News Issue 779 (18 January 1879). Newspaper image © The British Library Board. All rights reserved. With thanks to The British Newspaper Archive. [134] Map from Charles Warren Adams, The Notting Hill Mystery (London: Saunders, Otley, and Co, 1865). © The British Library Board. (12631. aaa.11). [136] Map from Émile Gaboriau, Monsieur Lecoq (Paris: Dentu, 1869). Public domain image via Wikimedia Commons. [137] House plan from Agatha Christie, The Mysterious Affair at Styles (New York; London: John Lane Co., 1920) p. 42. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1920). © The British Library Board. (Cup.410.f.420). [145] Room plan from Agatha Christie, The Mysterious Affair at Styles, p. 60. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1920). © The British Library Board. [145] House plan from Agatha Christie, The Murder of Roger Ackroyd, p. 63 (London: W. Collins, Sons & Co., 1926). Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1926). © The British Library Board. (Cup.410.f.554). [149] Room plan from Agatha Christie, The Murder of Roger Ackroyd, p. 91. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) (1926). © The British Library Board. (Cup.410.f.554). [150] Plan A from Agatha Christie, The Murder at the Vicarage (London: W. Collins, Sons & Co., [1930]) p. 27. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) ([1930]). © The British Library Board. (W 17328). [156] Plan B from Agatha Christie, The Murder at the Vicarage, p. 44. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) ([1930]). © The British Library Board. (W 17328). [156] Plan C from Agatha Christie, The Murder at the Vicarage, p. 71. Reprinted by permission of HarperCollins Publishers Ltd © (Agatha Christie) ([1930]). © The British Library Board. (W 17328). [157] Street plan from Margery Allingham, More Work for the Undertaker (London: William Heinemann, 1948). The Margery Allingham/Philip Youngman Carter Collection, Albert Sloman Library, University of Essex. Reprinted by permission of Peters Fraser & Dunlop (www .petersfraserdunlop.com) on behalf of Rights Limited. Image courtesy of the Albert Sloman Library, University of Essex. [160]

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

xii

List of Illustrations

5.1

Endpaper Map from Arthur Ransome, Swallows and Amazons (London: Jonathan Cape, 1930). © The British Library Board. By kind permission from Random House Ltd. [176] 5.2 Endpaper Map from Arthur Ransome, Swallowdale (London: Jonathan Cape, 1931). © The British Library Board. By kind permission from Random House Ltd. [177] 5.3 Endpaper Map from Arthur Ransome, Winter Holiday (London: Jonathan Cape, 1933). © The British Library Board. By kind permission from Random House Ltd. [179] 5.4 Detail from map for Arthur Ransome, Swallowdale (London: Jonathan Cape, 1931). © The British Library Board. By kind permission from Random House Ltd. [180] 5.5 Detail from map for Arthur Ransome, Winter Holiday (London: Jonathan Cape, 1933). © The British Library Board. By kind permission from Random House Ltd. [180] 5.6 ‘Peter Pan’s Map of Kensington Gardens’ from J. M. Barrie, Peter Pan in Kensington Gardens (London: Hodder & Stoughton, 1906). Beinecke Rare Book and Manuscript Library, Yale University. (GEN MSS 1400). [192] 6.1 Map for ‘A Part of The Shire’ from J. R. R. Tolkien, The Fellowship of the Ring: Being the First Part of The Lord of the Rings (Oxford: George Allen & Unwin, 1954). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1954). © The British Library Board. (RF.2017.a.9). [208] 6.2 ‘First sketch map of The Shire, [1938]’; MS. Tolkien Drawings 104r. Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd, 1998. [209] 6.3 ‘Sketch map of roads and rivers in The Shire, [1938]’. MS. Tolkien Drawings 105r, Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd, 2015. [210] 6.4 Back endpaper ‘Map of Wilderland’ from J. R. R. Tolkien, The Hobbit, or There and Back Again (London: George Allen & Unwin, 1937). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1937). (Cup.410.f.378). [210] 6.5 Christopher Tolkien’s ‘General Map for Middle Earth’ from J. R. R. Tolkien, The Fellowship of the Ring: Being the First Part of The Lord of the Rings (Oxford: George Allen & Unwin, 1954). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1954). © The British Library Board. (RF.2017.a.9). [211]

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

List of Illustrations

6.6

6.7

6.8

6.9

6.10

6.11

6.12

6.13

7.1

xiii

Working sketch map: ‘Sketch Map of the Misty Mountains and the Lands South and East [1939]’. MS. Tolkien Drawings 115r. Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd 1998, 2015. [214] Comparing ‘Carrock’ on the maps (created by author). Details from maps given at: Figure 4 ‘Map of Wilderland’; Figure 6: Working Sketch Map c.1939; ‘Map of the north-west of Middle Earth, [c.1948]’; MS. Tolkien Drawings 124r, Bodleian Library, Oxford, UK. Copyright: The Tolkien Estate Ltd 2015; Figure 5 ‘Christopher Tolkien’s General Map for Middle Earth’. [215] ‘Sketch of part of the Misty Mountains showing Dimrill Dale, [1940]’. MS. Tolkien Drawings 90, fol. 11r. Bodleian Library, Oxford, UK. Copyright: The Tolkien Trust 2015. [216] ‘Redrafting of topography and description for Minas Tirith’, MSS 3/1/ 24/3b, J. R. R. Tolkien Collection, Special Collections and University Archives, Raynor Memorial Libraries, Marquette University, Wisconsin. Image reproduced with the kind permission of Marquette University and the Tolkien estate. [219] ‘Christopher Tolkien’s Reconstruction of Tolkien’s Working Practices; An Annotated Version’, from John Krygier and Denis Wood, Making Maps, A Visual Guide to Map Making 2nd ed. (The Guilford Press, 2011) p. 72. [222] Christopher Tolkien's ‘Map of Gondor, Rohan and Mordor’ from J. R. R. Tolkien, The Return of the King: Being the Third Part of The Lord of the Rings (London: George Allen & Unwin, 1955). Reprinted by permission of HarperCollins Publishers Ltd © (J. R. R. Tolkien) (1955). © The British Library Board. (RF.2017.a.9). [223] ‘Map 2: Journey from Hobbiton to Bywater’, from Barbara Strachey, Journeys of Frodo: An Atlas of J. R. R. Tolkien’s The Lord of the Rings (London: George Allen & Unwin, 1981) p. 12. Reprinted by permission of HarperCollins Publishers Ltd © (Barbara Strachey) (1981). Reprinted by permission of Camphill Village Trust. © The British Library Board. (Maps 203.e.7). [232] ‘Map of the Mines of Moria’, from Karen Wynn Fonstad, The Atlas of Middle-Earth (London: Grafton, 1992) p. 128. Reprinted by permission of HarperCollins Publishers Ltd © (Karen Wynn Fonstad) (1981). © The British Library Board. (Maps 204.d.49). [234] ‘The thriving City of Eden as it appeared on paper’. Illustration from Charles Dickens, The Life and Adventures of Martin Chuzzlewit

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

xiv

List of Illustrations

7.2

7.3

7.4

7.5

7.6

(London: Chapman and Hall, 1844) p. 268. Public domain image courtesy of Hathi Trust Digital Library. https://hdl.handle.net/2027/uc2 .ark:/13960/t4vh5jp66?urlappend=%3Bseq=326, accessed 17 April 2019. [244] Choropleth map of four Trollope novels mapped onto Michael Sadleir’s Map of Barsetshire (created by author). Michael Sadleir, Trollope: A Commentary (London: Constable & Co., 1945) p. 163. © The British Library Board. (10861.aa.25). [257] ‘Anthony Trollope's Sketch Map of Barsetshire’, from Michael Sadleir, Trollope: A Commentary (London: Constable & Co., 1927) facing p. 154. (010856.cc.18). [260] Embryo Map by George Eliot, from Quarry for Middlemarch Notebook; MS Lowell 13, f. 57r, Houghton Library, Harvard University. Image reproduced by kind permission of Mr Jonathan G. Ouvry. [262] ‘The Stakes were won by Wildeve’, Illustration by Arthur Hopkins, from Belgravia: A London Magazine, Volume 36, Issue 141 (July 1878), opposite title page. © The British Library Board. (P.P.6004.gn). [269] Frontispiece Map from Thomas Hardy, The Return of the Native (London: Smith, Elder, & Co., 1878) Vol. 1. © The British Library Board. (C.71.h.7). [270]

Acknowledgement of Prior Publications In relation to the discussion of paratext in Chapter 1 and of imagetext in Chapter 5: Sally Bushell, ‘Imagetext or Paratext: Theorising the Literary Map’, Word & Image 32.2 (Summer 2016): 181–94. In relation to the discussion of Adventure Fiction in Chapter 3: Sally Bushell, ‘The Map in Victorian Adventure Fiction: Doubleness, Silence and the Ur-Map in Treasure Island and King Solomon’s Mines’, Victorian Studies 57.4 (2015): 611–37. In relation to the discussion of Agatha Christie’s The Murder of Roger Ackroyd in Chapter 4: Sally Bushell, ‘“The Slipperiness of Literary Maps”: Critical Cartography and Literary Cartography’, Cartographica 47.3 (Fall 2012): 49–61.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Acknowledgements

Let me begin at the beginning and acknowledge my greatest debt to those who taught me how to be the academic I am – Gillian Beer, Nigel Leask and the late Sue Manning – sine qui nihil. Next, I thank Lancaster University, the institution which gave me my first academic post back in 2000 and has been a truly supportive and collegial environment ever since. I thank the larger University for giving me the sabbatical time to complete this book and I thank my many colleagues in the department and the Faculty who really do make a difference. Chief amongst these is Simon Bainbridge – the best and wisest colleague anyone could ever wish to have – but I also want to thank: Jo Carruthers, Simon Guy, Keith Hanley, Hilary Hinds, Liz Oakley-Brown, Sharon Ruston, Leila Atkinson and Anne Stewart-Whalley, amongst others. The postdoctoral research team on the AHRC Chronotopic Cartographies project (James Butler, Duncan Hay and Becca Hutcheon) have also fed into this book in many ways – particularly in relation to the final chapter – and I thank them, too, for patiently letting me complete it when I should have been giving more time to them. Thanks also to David Cooper for his generous comments on the final chapter. As always, I thank my family – my mother, Jackie; my father, Alan; my brother, Andrew; and my sister, Gill – for teaching me all kinds of useful things, including competitive sibling rivalry, how to construct a winning argument, how to read a map and much, much more. And I thank John Hilliard for being there throughout in his own inimitable way. Friendship is as important as family, and many people have been generous and supported me across the decades: Charlotte Avery, David Clifford, Anna Evans, Tim Fulford, Ben Quash, Sue Richardson, Jan Schramm, Bart Van Es, Wei Wei Yeo. I also want to thank those around me, day by day, week by week, who helped me through the grind: Zefar Abbas, Sally and Paul Darbon, Sobia Khan, Shehz Malik, Debbie Stubbs, Rachel-Ann Powers, Tracy Simmonds, Caroline Rose, Chris Warren, Celia Roberts and Adrian McKenzie. In preparing the book for publication I want to thank the two readers at Cambridge University Press who gave me generous and invaluable advice,

xv

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

xvi

Acknowledgements

and Bethany Thomas for guiding me and it through the publication process so kindly and well. My thanks are also due to Kate Ingle for her help in reading and proofing the text as well as her warm and encouraging moral support and to Claire McGann for her superb assistance with the many images and permissions. Finally, I want to thank my son, Dylan Bushell, for his great forbearance as I wrote this book. With him in my life it took a little longer to reach completion than it otherwise might have done, but I don’t regret it one bit.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876

Introduction

A simple question lies at the heart of this book: to what extent do we map as we read? How essential a part of the experience of a literary work is the way in which we spatialise, and visualise (externally or internally) the place and space of the fictional world? Reading and Mapping Fiction provides an integrated account of the relationship between maps and texts as part of the totality of meaning for the literary work. The core argument of this study is that this activity, whatever it is called – mapping, diagramming, visualising, spatialising – is a vital and intrinsic part of how we experience literature, and of what makes it so powerful. Rather than centring a method of mapping on the referential relationship between real-world geography and fictional place and space, as previous influential incarnations of literary mapping have tended to do, this book offers a fully dynamic interpretation of fictional map and text for the twenty-first century. The primary focus of the study is on the juxtaposition of visual and verbal representations of the same fictional world where a map is explicitly present. The book is thus about ‘Reading’ and ‘Mapping’ in a way that repeatedly moves across and between those two acts and applies both to both forms of representation (reading fictional maps and mapping fictional texts). It focuses primarily upon fictional place and space rather than adopting a larger definition of ‘the literary’ that might include forms in which maps with real-world correspondence are present (e.g. the memoir; travel writing). In such works the maps function differently and readerly perception is likely to be more highly predetermined by real-world geography. In contrast, this study is primarily focused on representations of imaginary place and space; on authorial mapping of that imaginative space in composition; and on the integration of map and text by the reader. In terms of method, Reading and Mapping Fiction combines conceptual thinking about literary spatiality and the interconnections of the fields of literature, geography and cartography, with a literary-historical account of the evolution of the fictional map over time and its emergence in popular genres. The conceptual focus is informed primarily by the dominant critical cartographic mode of the current period that enables us to read against the map. The historical focus provides a broad overview of literary and

1

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

2

Introduction

historical map history from the medieval period up to the late nineteenth century (the point at which literary maps proliferate across new forms of fiction). Whilst these two imperatives could be seen to be in tension with one another, it is to be hoped that this is a fertile tension that brings forth fruit. Close analysis of particular examples opens up thinking about meaning in relation to a specific text or author, on the one hand, while, on the other, the larger nature of the genre is able to be understood spatially in a new way, in and through the specific example. The method thus works in both directions, moving from the inside out (e.g. enlarging from analysis of an influential map in the first work of a new genre) and from the outside in (starting with the emerging genre and moving to analysis of map examples within it). The study is original and innovative primarily because of its strong focus upon empirical maps in relation to genres, but also in bringing together a number of different critical frames (literary-critical and theoretical; historical; critical-cartographic; cognitive) in order to create a model of interdisciplinary interpretation based in literary studies that reaches towards and draws upon other disciplines. Prior to the twenty-first century, there were few critical studies in the field of literary mapping. Three early collections of fictional maps first stimulated my interest: An Atlas of Fantasy by J. B. Post; Language of the Land by Martha Hopkins and Michael Buscher; and You Are Here by Katharine Harmon.1 The most famous critical-theoretical intervention into the field, and, again, one that drew me into it, is that of Franco Moretti with a series of key publications in the late 1990s and early twenty-first century.2 As we shall see in Chapter 1 (and in Chapter 8 in relation to digital mapping), the work of Moretti (individually and as part of the Stanford

1

2

J. B. Post, An Atlas of Fantasy (1973), rev. ed. (New York: Ballantyne Books, 1979); Martha Hopkins and Michael Buscher, Language of the Land: The Library of Congress Book of Literary Maps (Washington, DC: Library of Congress, 1999); Katharine Harmon, You Are Here: Personal Geographies and Other Maps of the Imagination (New York: Princeton Architectural Press, 2004). See also Alberto Manguel and Gianni Guadalupi, The Dictionary of Imaginary Places (New York: Macmillan Publishing Co. Ltd, 1980). More recent studies of a similar sort include: Peter Turchi, Maps of the Imagination: The Writer as Cartographer (San Antonio, TX: Trinity University Press, 2004); Huw Lewis-Jones, The Writer’s Map: An Atlas of Imaginary Lands (London: Thames & Hudson, 2018); and John Sutherland, Literary Landscapes: Charting the Real-Life Settings of the World’s Favourite Fiction (London: Modern Books, 2018). Another early readerly-focused work from the field of cartography that influenced this study was Phillip C. Muehrcke with Juliana O. Muehrcke, Map Use: Reading, Analysis and Interpretation (Madison, WI: J. P. Publications, 1978). Franco Moretti’s three most directly relevant books here are Atlas of the European Novel: 1800–1900 (London: Verso, 1998); Graphs, Maps, Trees: Abstract Models for a Literary History (London and New York: Verso, 2004); and Distant Reading (London: Verso, 2013).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

Introduction

3

LitLab) and of the Moretti-inspired digital mapping projects led by Barbara Piatti is highly innovative for its time, but also presents certain limitations, particularly in terms of its disciplinary value to literary studies.3 Since 2010, interest in both popular and academic literary mapping has considerably increased. There has been a recent surge in essay collections at the intersection between literature and geography/cartography, with three substantial edited books published since 2014 and strong impetus also provided by Robert Tally’s ‘Geocriticism and Spatial Literary Studies’ series with Palgrave Macmillan.4 Of the three edited collections, Tally’s Literary Cartographies has the strongest focus, reflecting his own particular double-take on ‘literary cartography’ in which the term ‘mapping narratives’ (the title of his introduction) signals that narrative is both something to map with and to be mapped. So ‘in mapping a place one also tells a story’ (2) but ‘all spaces are necessarily embedded with narratives’ (2). Anders Engberg-Pedersen’s collection, Literature and Cartography, is centred upon the ‘productive tension’ (15) between the two fields. However, when we look across the contents of these collections as a whole, we find a strong tendency towards what Emmanuelle Peraldo calls ‘a transdisciplinary global debate’ (4) that celebrates ‘plurality’ and ‘multifocal perspectives’ (6). This makes a virtue of necessity but cannot quite hide a problem of ever-expanding definitions and boundaries for the field. Engberg-Pedersen, for his part, addresses this problem directly: ‘literary studies is faced with a ubiquity problem’ but ‘the problem is not restricted to the literary field . . . [there is] general overuse of the terms mapping and remapping’ (19). His solution – to remember that maps ‘guide us and lead us astray at the same time’ (19) – is one that I wholeheartedly embrace in this volume. The organisation of Engberg-Pedersen’s edition into three sections – relating to theory and method; history and context; and genre and theme – also compares to the dominant approaches employed in this book.

3

4

See https://litlab.stanford.edu and www.literaturatlas.eu/en/. Barbara Piatti et al. also have a number of influential papers, such as ‘Mapping Literature: Towards a Geography of Fiction’, in Cartography and Art, ed. W. Cartwright, G. Gartner and A. Lehn (Berlin: Springer, 2009), 1–16. See Robert T. Tally Jr, ‘Introduction: Mapping Narratives’, in Literary Cartographies: Spatiality, Representation and Narrative, ed. Robert T. Tally (Basingstoke: Palgrave Macmillan, 2014), 1–12; Emmanuelle Peraldo, ‘The Meeting of Two Practices of Space’, in Literature and Geography: The Writing of Space throughout History, ed. Emmanuelle Peraldo (Newcastle: Cambridge Scholars Publishing, 2016) 1–19; Anders Engberg-Pedersen, ‘Introduction: Estranging the Map: On Literature and Cartography’, in Literature and Cartography: Theories, Histories, Genres, ed. Anders Engberg-Pedersen (Cambridge, MA, MIT Press, 2017), 14–32.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

4

Introduction

Any similarity between Engberg-Pedersen’s position and my own may be in part because they are both implicitly aligned with that of Christian Jacob. In his monumental study The Sovereign Map, Jacob approaches the map as: a dynamic process whose effects, power and meanings are to be found at the crossroads of production and reception, of encoding and decoding, of intentions and expectations.5

The need for a multiple approach is fully articulated by Jacob: From this double evolution, historical and theoretical, we can see that the map has become an opaque object that calls attention to itself. Representational strategies are now considered, with the result that maps have lost their ‘innocence’. (6)

Although it is not his primary focus, Jacob’s discussion of imaginary maps in terms not only of the gaze but also the non-referentiality of fictional mapping is worth consideration (272–95) and he helpfully suggests that ‘cartographical fiction probably constitutes a good point of departure for the study of mechanisms of authority that belong to maps . . . and the social convention that grounds their referential links’ (272). Emerging out of such positions, Reading and Mapping Fiction could be understood to participate in a second phase for the field that involves a more dynamic model able to both analyse and critique the map in relation to the text and to adapt cartographic concepts more broadly to the interpretation of literary forms and narrative structures. Two recent publications draw close to this one. The first is John Wyatt’s Imaginary, Historical and Actual Maps in Literature (2013).6 This offers a thorough and comprehensive survey of a wide range of maps from the medieval period onward, including real-world mapping of the Lake District and the Ordnance Survey as well as covering most famous authorial map-makers. The book is divided into three parts: the first covering books of distant lands; the second, maps made in a literary context; the third considering how social change affects the map, particularly in city spaces. For the most part, however, Wyatt’s work functions uncritically at a level of information and description, rather than analysis, and does not seek to advance a critical argument. 5

6

Christian Jacob, The Sovereign Map: Theoretical Approaches in Cartography throughout History, ed. Edward H. Dahl, trans. Tom Conley (Chicago and London: University of Chicago Press, 2006), xv. John Wyatt, The Use of Imaginary, Historical and Actual Maps in Literature (Lampeter: Edwin Mellen Press, 2013).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

Introduction

5

The second study is Christina Ljungberg’s Creative Dynamics (2012).7 Ljungberg approaches the study of literary maps from a similar starting point to my own – with an interest in how such maps are read – and, superficially, the contents of my study partly overlap with hers. Ljungberg’s interest is in the connection between ‘thinking cartographically’ and ‘thinking diagrammatically and reading diagrammatically’ (1). Thus, her focus is also on the dynamic between visual and verbal forms of interpretation. She asserts: The insertion of spatial diagrams into the verbal text produces a creative interplay between the verbal and the visual narrative which is generated by the juxtaposition of the two media. Shifting between them [are] attentive readers . . . (3)

We are both interested in the same performative relationship at the heart of a text – the relationship between reading and mapping. However, it is worth noting that her book is published in a series entitled ‘Iconicity in Language and Literature’ and, as this implies, is strongly informed by Peircean semiotics. In contrast, it might be said that my primary interest is in the literary work as a totality and in the complex interactions between the meaning of the text and of the image working together for the reading of literature. Where Ljungberg’s focus is ‘on the processes involving diagrammatic thought’ (3), mine is centred on developing an integrated literary-critical method of interpretation for the experience of full spatial and material meanings in the text. * * * It is necessary to consider the nature and function of a map for a work of literature. We can start quite simply by asking: what is a map? The OED defines it as: a drawing or other representation of the earth’s surface or a part of it made on a flat surface showing the distribution of physical or geographical features . . . also a plan of the format or layout of something, as a route, a building, etc.8

A second influential definition, in the ‘Preface’ to the first volume of the History of Cartography edited by Harley and Woodward, offers an entirely new definition of ‘map’, one that is neither too restrictive nor yet so general as to be meaningless. What has eventually emerged is a simple formulation: Maps 7

8

Christina Ljungberg, Creative Dynamics: Diagrammatic Strategies in Narrative (Amsterdam: John Benjamins, 2012). Oxford English Dictionary Online: www.oed.com.ezproxy.lancs.ac.uk/view/Entry/113853. Accessed 17 April 2019.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

6

Introduction

are graphic representations that facilitate a spatial understanding of things, concepts, conditions, processes, or events in the human world.9

Finally, Christian Jacob builds upon the findings of critical cartography to assert that: A map is defined perhaps less by formal traits than by the particular conditions of its production and reception . . . [I]n a problematic mixture, it combines the transparence of referential illusion with the opacity of a medium that materializes the geographic image. (21)10

What all these descriptions make clear is that the term ‘map’ can essentially cover any visual attempt to represent relative or absolute relations in space. This study will assume such a broad definition in its use of the term. For literary mapping, secondary activities implicit in map-reading are likely to include: viewing the world from multiple perspectives; the relative situating of objects in space; the unfolding of a route; the locating of the self in relation to other people and places; a sense of ‘mapping things out’ (in relation to the fictional world, movements of characters or the literary work). How then can we describe a literary map? I would define it as: a representation of spatial relations between places, people or objects (real or imagined) that corresponds visually to the world that the text purports to represent verbally.

Such a map is authorial if it appears at the time of first publication, but there may also be post-authorial maps created subsequently over time, and certain texts (e.g. Dante’s Inferno) seem to demand and generate such maps. For the purposes of this study, my focus is primarily on fictional maps. Recent critics have attempted to determine a range of map-types for literature that allow for the relationship between verbal, visual and material elements.11

9

10

11

J. B. Harley and David Woodward, ‘Preface’, in History of Cartography, Vol. 1. Cartography in Prehistoric, Ancient and Medieval Europe and the Mediterranean (Chicago: University of Chicago Press, 1987), xvi. Jacob gives a full answer to the question ‘What is a Map?’ in his first chapter (The Sovereign Map, 11–102). See also Tania Rossetto’s summary of different kinds of literary cartography in ‘Theorizing Maps with Literature’, Progress in Human Geography 38 (2014): 513–30; 515–16; and Ryan’s taxonomy of narrative cartography (referenced in Rossetto as well): Marie-Laure Ryan, ‘Toward a Visual Narratology’, in What is Narratology? Questions and Answers Regarding the Status of a Theory, ed. T. Kindt and H. H. Muller (Berlin: Walter De Geuyer and Co., 2003), 333–64.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

Introduction

7

Robert Stockhammer (as summarised by Ryan, Foote and Azaryahu) identifies four kinds of relationship: 1. 2. 3. 4.

Map precedes the text. Author draws maps during writing. Publisher puts a map in second or third edition. Map is drawn by readers and critics.12

These are fairly broad categories, centred upon the relative temporality of map production and authorship – although the question of when the map was produced does bear directly upon meaning. For example, the primary concern of this study is with the integration of meaning across and between visual and verbal forms, with the result that the focus is primarily on authorial (or authorially approved) maps that appear alongside the text from the first edition onward. Such maps are likely to be more fully integrated with the spatial dimensions of the work than a later post-authorial map. A second, more comprehensive, categorisation is given by Matthew Graves who distinguishes between: Para-textual maps: the frontispiece or appended map as an ideographic emblem which prefigures the text Intra-textual maps: maps which are embedded in the narrative, intradiegetic place-markers Intertextual maps: which refer to an external geography, referential or imaginary Logo-textual maps: word maps or narrated maps that are pure text, bereft of graphic form.13 Graves’s categories are of considerable relevance since they are centred upon the problematic nature of representation and reference and include both paratext and visual/verbal forms. His inclusion of verbal descriptions as ‘language constructions whose spatial extension is left to the imagination of the reader’ is directly relevant to the last two chapters of this book.14

12

13

14

Marie-Laure Ryan, Kenneth Foote and Maoz Azaryahu, Narrating Space / Spatializing Narrative: Where Narrative Theory and Geography Meet (Columbus: Ohio State University Press, 2016), 59. From Robert Stockhammer, Kartierung der Erde, Macht und Lust in Karten und Literatur (Munich: Wilhelm Fink Verlag, 2007), 63. Matthew Graves, ‘Maps and Texts: Reading Literary Maps – The Case of The Riddle of the Sands’, Trans 16.9.6 (July 2006), Internet-Zeitschrift für Kulturwissenschaften (Knowledge Networking in Cultural Studies): www.inst.at/trans/16Nr/09_6/graves16.htm. Graves has ‘Intertextual’ as a subcategory within ‘Intratextual’ but they seem to me to be distinct. Graves, ‘Maps and Texts’.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

8

Introduction

Finally, Marie-Laure Ryan in ‘Toward a Visual Narratology’ provides a full attempt to define and classify different kinds of space relevant to the generation of ‘concrete visual maps’ (335) as an integral part of the text. She denotes: the space of production; the represented space of the text; spatial form; the space of the reader; the material place and space of the book (335). Ryan then classifies maps (in a way that can be crossreferenced to these spaces) as either: internal – ‘designed by an author or illustrator as part of the interface between the text and the reader’ (336) – or external: ‘designed by readers . . . usually critics’ and ‘in rare cases . . . also . . . drawn by authors themselves as an aide to the imagination’ (336). Such a division loosely corresponds to that between writerly/readerly maps and partly anticipates the primary division between inward-facing or outward-facing maps positioned on the threshold of the paratext, as discussed below. It is essential to consider what might be unique about maps for literature as opposed to any other kind of map. As this study will go on to illustrate, there are three primary characteristics that I would identify as fundamental to the literary map: the expository nature of the map/text relationship; the map as a spatial visualisation of a sequential narrative; the doublefacedness of the map.15 I want to dwell a little more on each of these here. A map is a thing of use; what appears on (and is left off ) a map – and thus what we understand a map to be – is determined by anticipated audience and function. For literary maps, the existence of a map alongside a text naturally leads to the assumption that its main purpose (or use value) is one of visual elucidation of spatial elements in the written text. As such it is highly likely to be considered as secondary to the text and expository in nature. Mark Monmonier’s account of just such a relationship for maps in academic texts in Mapping It Out proves helpful.16 The purpose of exposition is to explain or describe something more clearly; so the presentation of a map alongside a text in a scholarly work (for example, maps of battlefields, or naval battles, in a history text) operates fairly straightforwardly as a visual aid that clarifies a verbal account centred upon multiple complex movements in relation to a core event. In considering the power 15

16

A fictional map is also a literary map, but a literary map need not be fictional. Since my main focus is on fictional maps, I use both terms more or less interchangeably. Mark Monmonier, Mapping It Out: Expository Cartography for the Humanities and Social Sciences (Chicago, IL, and London: University of Chicago Press, 1997). Monmonier’s book is primarily intended as a bridge between cartography and those working in the humanities who seek to develop visual and spatial approaches.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

Introduction

9

of the expository map, Monmonier discusses the way in which the linguistic code of the map (words and font types), provides ‘a needed link between the cartographic symbols and the natural language of authors and readers’ (93). Monmonier argues that when a map and text occur alongside each other, the words on the map have a particular function: Words that reflect the author’s ideas and terminology are needed to tie the map to the written text and to integrate the structurally diverse realms of cartographic and literary representation. (93)

This suggests that, in terms of the system of codes on the literary map, the linguistic code is likely to be of heightened importance. It is worth remembering, then, that we do always literally ‘read’ the map. A closely related issue is the way in which the map functions in relation to narrative, but also displays narrative elements itself. Monmonier states that: ‘By organizing information chronologically as well as spatially, maps can support a variety of historical narratives addressing long or short periods of time’ (204). The fictional map often does attempt to do this in a highly distinctive way, by giving far more text than would ordinarily appear on a map, often added into the boxes intended for the ‘Title’ or ‘Key’ but also linking to major events in the narrative on the map itself.17 Text on the map describes dramatic events within the narrative and so ties together the hermeneutic codes of map and text as the simultaneity of the map anticipates and stimulates interest in the temporal unfolding of the story. The literary map thus raises questions that only the text (and a rereading of the map in the light of the text) can resolve – thus establishing a dynamic temporal relationship between the simultaneity of the map and the sequentiality of the narrative. Monmonier also makes the important point that the expository map is easily overlooked in favour of the text so that: ‘the first goal of integrative cartography is to get the reader to look at the map’ (242). Literary mapping is a particular form of ‘integrative cartography’ (242) but achieving these goals is not always straightforward, so that a range of explicit and implicit tactics are employed. For maps in academic writing, to ensure that the map fulfils its primary purpose as a thing of use, Monmonier advises that a text needs to explicitly direct its readers towards the map: ‘Expository cartography works best when the author tells the reader to look at the map’ (94). In literary mapping, for certain genres in which the map is presented as 17

See for example the endpaper map for Arthur Ransome, We Didn’t Mean to Go to Sea (London: Jonathan Cape, 1937).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

10

Introduction

part of an apparatus around the text, there is no difficulty with a similar use of explicit instructions by means of footnotes or direction within the text.18 Elsewhere, readerly use of the map is less directly indicated by proxy (e.g. Jim Hawkins pores over the map of Treasure Island in anticipation of journeying there and so raises anticipation in the reader). The expository nature of the literary map directly relates to a second major feature, mentioned above: its identity as the spatialisation of a sequential narrative. Historically, a comparable relationship between visual and verbal representational forms finds its starting point in the classical debate initiated by Horace’s Ars Poetica. As part of an attempt to accord poetry the same degree of careful analysis as the visual arts, Horace makes the famous statement ut pictura poeisis (‘as is painting so is poetry’).19 From such a perspective, the shared goal of literature and visual art is understood to be representation through imitation of nature (mimesis) so that ideas of equivalence underpin aesthetic principles right through to the mid-eighteenth century. A major intervention occurs with Lessing’s ‘Laocoon: An Essay on the Limits of Painting and Poetry’ (1766) in which he asserts a distinction between the two art forms on the grounds of time and space: there is this essential difference between them: one is a visible progressive action, the various parts of which follow one another in time; the other is a visible stationary action, the development of whose parts take place in space.20

Lessing goes on to develop a full account of this distinction, affirming that ‘succession in time is the province of the poet, co-existence in space that of the artist’ (109). This position opens the way for a larger shift from a mimetic to an expressive aesthetics with a consequent redetermining of the visual/verbal relationship analogically in the Romantic period and beyond. Visual and verbal forms of representation are no longer understood to be directly equivalent but now work in a comparable, but fundamentally different way (with the space/time distinction as a key element of that difference). 18

19

20

E.g. in travel writing. For an example within fictional maps, see Erskine Childers’s The Riddle of the Sands: A Record of Secret Service Recently Achieved (London: Smith, Elder & Co., 1903). The reader is repeatedly directed back to the maps at the front of the book by means of footnotes or references within the text. Horace: Satires, Epistles and Ars Poetica, trans. H. R. Fairclough (London: Loeb Classical Library, 1936), 2. Gotthold Ephraim Lessing, Laocoön: An Essay on the Limits of Painting and Poetry (1766), trans. Edward Allen McCormick (Baltimore, MD, and London: Johns Hopkins University Press, 1984), 90.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

Introduction

11

The grounds for Lessing’s distinction are particularly pertinent for understanding the relationship between a literary map and the text to which it corresponds. What does the map give us that the text does not? It provides a simultaneous experience of places, events at key sites and the relations between them which stands against/alongside the successive nature of the text. Lessing’s description of the two larger art forms as ‘like two just and friendly neighbours . . . who exercise mutual forbearance on the borders’ (110) also proves helpful. For the literary map, the two forms coexist physically within the covers of the book but also remain vitally connected at a level of meaning and interpretation since they purport to represent (visually and verbally) the same place. Because the map corresponds to a text, and its use value is tied into reading that text, the literary map is bound up with the temporality of narrative to a far greater degree than would normally be true (even for an expository map). For the literary map, a full understanding of what is represented cannot be immediately achieved because the text has not yet been read (assuming a normative position for the map at the front of the book). Equally, linkage to the narrative means that the map needs to be returned to at key points within the sequence to be fully utilised and activated. In a range of ways, then, the literary map draws attention to the need for it to be read primarily in and through a juxtaposed relationship. The third dominant characteristic of a literary map is less obvious but is undeniably present (as we shall see in a range of ways across this study): its two-facedness, or doubleness. The fictional map, in particular, frequently functions at two levels simultaneously: at one level, it is represented within the text, involving use of the map by the characters; at another it is outside the text and faces toward the reader. Wherever the map is explicitly referred to within the text, it is immediately doubled. This element is handled most adroitly in fictional works (e.g. Thomas More’s Utopia) and also relates to the identity and status of the map as part of the paratextual apparatus of the book. Christina Ljungberg articulates three key ideas that are relevant to this double-facedness. In Peircean terms, a map is a sign system employing at least three major forms of code: iconic; indexical and symbolic.21 Indexical and iconic signs appear more ‘natural’ (and thus more ‘motivated’) than

21

An index is a sign that has an actual physical connection with its object (e.g. a weathercock for wind direction; smoke for a fire). An icon is a sign based only on perceived likeness between sign and referent (e.g. road signs). A symbol is a sign determined by convention with no necessary resemblance (e.g. language; mathematics).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

12

Introduction

the symbolic.22 The spatial reader who makes full use of the map as part of the experience of the literary work must be moving across and between different semiotic systems on the map itself (which contains images, symbols and text) as well as between map and text. In a highly spatial text that wants to aid the reader’s ability to visualise a particular world, landscape or route, the presence of a map can help literature to communicate the iconicity it seeks. Peircean semiotics also makes clear that the power of iconic resemblance lies in its ability to ‘reveal truth’ by presenting the same information in a different form.23 Such a comment is immediately relevant to the role of maps alongside texts in works of fiction where they may serve to reveal or cover up the truth. Ljungberg strongly situates her own ‘diagrammatic’ work in relation to Peirce’s second subcategory for the icon, defining literary maps as: spatial embodiments of knowledge with a peculiar potential to stimulate new cognitive engagements. This lies in their capacity to abstract and to generalize, which makes them ideal not only for activities such as solving problems, hypothesizing or just tracing imaginary journeys on a map – or in the mind. (Creative Dynamics, 5)

Ljungberg argues that the main way in which the map is iconic is in terms of its representation of the relations between things: ‘A map is a “diagram” example of a territory because the relationships in it are presented as analogous to that arrangement in the territory itself’ (18–19).24 A second important point concerns her account of any map as having two referents or objects: In Peircean semiotics, a map has at least two objects, one dynamic and one immediate object. The dynamical object is the reality of the geographical facts (in the case of a map) whereas the immediate object is what could be called the mode of imaging intrinsic to the map itself, which will influence the ways in which the map will reflect our cultural and personal knowledge.25

22 23

24

25

‘Motivation’ concerns the degree to which the signifier is determined by the signified. Peirce describes ‘this capacity of revealing unexpected truth’ (Charles Peirce, Collected Papers of Charles Sanders Peirce. Vol. I Principles of Philosophy and Vol. II Elements of Logic, ed. Charles Hartshorne and Paul Weiss (Cambridge, MA: The Belknap Press of Harvard University Press, 1960), II: 158). Peirce notes: ‘Every picture . . . is essentially a representation of that kind [iconic]. So is every diagram, even although there be no sensuous resemblance between it and its object, but only an analogy between the relations of the parts of each’ (II, 158). Christina Ljungberg, ‘Cartographies of the Future: Julie Mehretu’s Dynamic Charting of Fluid Space’, The Cartographic Journal 46 (2009): 308–15; 309.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

Introduction

13

The map ‘represents’ or ‘is like’ a geographical region (say) that forms its object in terms of what it purports to depict out in the world but, at the same time, it is also interpreted through the mind of the interpretant (drawing upon all kinds of personal and intertextual associations) because, as an icon, its power lies in its ability to evoke a corresponding image of itself. It has both an internal and an external referent. The crucial point to realise is that reading empirical maps within literature thus means ‘working with two sets of references, one referring to the universe of the text, the other one to the lifeworld and the memories of the reader’.26 This doubleness in terms of map correspondence (to places and objects in the fictional world and to the reader’s internalised experience of place and space) proves to be a fundamental feature of fictional literary mapping which finds expression in a range of ways. As already noted, the most obvious is the way in which the reader is ostensibly provided with ‘the same map’ (or a copy; or the final version) of that used by the characters. This enables an extension of map use from character to reader with the intended result that when the characters consult the map within the story, the reader is also prompted to turn to his or her copy at the front of the book. In a slightly contradictory way, the reader is taken out of the narrative and act of reading in order to undertake a mimetic act that draws him or her further back into the narrative. The map thus straddles representational and material borderlines; it is both inside and outside the text. This can also create representational confusion and draw attention to the illusory representational nature of both visual and verbal forms. * * * An implicit three-part structure underlies the eight chapters of Reading and Mapping Fiction. The first two chapters seek to contextualise acts of literary mapping in two ways: conceptually and literary-historically. The middle four chapters of the book focus closely on the integral function of maps in new forms of fiction as these genres emerge in the late nineteenth century. The final two chapters turn towards problems of representation and the internal mapping of the reader. The first chapter contextualises the dynamic model of reading and mapping articulated here in terms of the evolving interdisciplinary relationship between literature, geography and cartography. It outlines the

26

Christina Ljungberg, ‘Dynamic Instances of Interaction: The Performative Function of Iconicity in Literary Texts’, Sign Systems Studies 38 (2010): 271–93; 276.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

14

Introduction

ways in which forms of literary mapping have emerged out of that connection before turning to critical cartography and its potential in relation to a more critical form of literary mapping. The final sections of the chapter use Gérard Genette’s concept of the paratext to analyse the material nature of the juxtaposition of map and text. Chapter 2 then contextualises the fictional map itself by aligning early examples to the history of cartography, centred on major turning points and correspondence (or noncorrespondence) between real-world developments and works of literature. Chapters 3, 4, 5 and 6 focus on the emergence of explicit maps as an integral element within popular genres and thus follow a rough chronology from the mid-nineteenth to the mid-twentieth century. Chapter 3 explores the map in Adventure Fiction as it emerges in the late nineteenth century at a point when literary maps proliferate across texts and genres. This chapter offers a detailed reading of two iconic maps in Treasure Island and King Solomon’s Mines and seeks to show that visual and verbal meanings are fully integrated. In the final sections of the chapter the concept of doubling in map and text is taken to its furthest extreme and works to create a new genre – the spy thriller. Chapter 4 examines the emergence of detective fiction in relation to real-world crime and how the reporting and mapping of it within newspapers bears upon fictional representations. It offers a sceptical reading of the map as something not to be trusted in the search for truth. A final section on maps and human geometry (relations between people and place) analyses the social and human dimensions of such maps in the work of Agatha Christie and Margery Allingham. Chapter 5 explores the spatial dimensions of children’s fiction, arguing that the explicit presence of the map in many canonical texts signals the importance of place and space for the genre and that a specific form of bounded space between adult and child (playspace) emerges. The spatial play that this enables is exemplified in Arthur Ransome’s Lake District Series whilst the – equally powerful – negation of playspace is embodied in Peter Pan. Chapter 6 considers the importance of the map for fantasy, with particular reference to J. R. R. Tolkien as the creator of that genre. Tolkien’s cartographic imagination in the process of creative composition is unique and so the chapter is largely devoted to a full exploration of this. The last section of the chapter draws attention to a post-authorial context for literary mapping, reflecting the larger cultural power and influence of Tolkien’s maps. The final two chapters of the study turn from consideration of ways in which the material presence of the map bears upon authorial and readerly meaning-making, to ways in which the absence, or internalisation, of the

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

Introduction

15

map affects the reader’s engagement with the text. Literary mapping is unusual by comparison with maps in other disciplines, in that the question of why a map is not present, or is withheld, can be of as much interest as its presence. Chapter 7 addresses a question that implicitly emerges from the earlier chapters: why do maps occur so frequently in popular genres but extremely infrequently in canonical texts (especially the realist novel)? After exploring this issue through debates around realism and representation in France and Britain, the chapter considers two rare canonical authors who do use maps in relation to the realist novel (Trollope and Hardy). The final chapter seeks to approach the intangible subject of the cognitive map for literature from a number of overlapping directions. The study concludes by considering ways forward for the mapping of literature in and through the digital medium. Across my research, in exploring aspects of the literary work that are commonly considered marginal or subsidiary (in this case the fictional map) I aim to enlarge the boundaries of what we understand ‘the literary work’ to be. In larger terms I seek to achieve this through new modes of interpretation and by shifting the focus of literary criticism from interpretation of the semantic content onto understanding of other aspects of the work that not only illuminate traditional means of interpretation but potentially redetermine those means.27 In this study, I attempt to do this by interpreting a paratextual element (the map) not always present within a work in order to explore visual/verbal relations when the map is there as well as when it is not, and to consider the role of the spatial writer and reader as connective agent; reading and mapping. Overall, then, this book is intended to raise larger questions concerning the nature of literature, meaning and interpretation and the role of the critic, in ways that connect to my own previous and ongoing interests.

27

This book is the second of three proposed works. The first, Text as Process: Creative Composition in Wordsworth, Tennyson and Dickinson (Charlottesville: University of Virginia Press, 2009), explored the margins of textuality by developing a method for interpreting works in a state of process. The third book, provisionally entitled A Spatial and Material Hermeneutics for Literature, will conclude the series.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.001

1

A Shifting Relationship: From Literary Geography to Critical Literary Mapping

Many things on maps do not exist in reality . . . many things in reality are missing from maps. P. C. Muehrcke and J. O. Muehrcke, Map Use

This first chapter approaches the central concept of literary mapping by seeking to locate its interdisciplinary origins in an earlier, broader understanding of ‘literary geography’, with an evolving sense of the nature of the connection involved. Over time, this term becomes increasingly complex and allows for a wide range of definitions and approaches. The chapter also articulates a particular kind of relationship between the two disciplines of literature and cartography that allows for the direct application of shared theoretical principles informing cultural geography and critical cartography back onto literature. For this study, with its strong empirical and generic focus on the fictional map, a knowing, selfconscious approach (‘critical literary mapping’) enables complex ways of interpreting the relationship between visual and verbal meanings held within the literary work that connect map to text in order to read them with, and against, each other. The first three parts of this chapter thus provide a broader conceptual groundwork and context that establishes the dominant theoretical model for the later genre chapters. This leads into a final section on the all-important material juxtaposition of map and text for literature.

The Origins of Literary Geography In her first review article for the Times Literary Supplement (1905), Virginia Woolf wrote a brief piece entitled ‘Literary Geography’ in which she stated: A writer’s country is a territory within his own brain; and we run the risk of disillusionment if we try to turn such phantom cities into tangible brick and mortar. We know our way there without signposts . . . No city indeed is so real

16

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

The Origins of Literary Geography

17

as this that we make for ourselves . . . and to insist that it has any counterpoint in the cities of the earth is to rob it of half of its charm.1

Woolf was reviewing two books (Thackeray’s Country and Dickens’ Country) concerned with mapping fictional space onto actual locations in the world that readers could then go and visit in acts of literary pilgrimage and tourism. For her, any attempt to map one onto the other had to be misplaced since, as her comment makes clear, the imaginative occupation of literary space far exceeds the experience of actual place in the world. The term ‘literary geography’ used here by Woolf had first been employed just one year earlier as the title of an essay collection by William Sharp. His essays were strongly authorial, primarily centred upon the major writers of the day, with photographs of their houses and relevant landscapes. The concept of literary geography thus first came into existence as a direct extension of biographical interest in the topographical background of the author.2 At the same time, Sharp’s collection acknowledges the tension identified by Woolf between the unlimited reach of a writer’s imagination and physical geography and he frequently undertakes a debate over which of these two realms constitutes literary geography. When determining the extent of Scott’s Country, for example, he momentarily wavers between ‘the wider country . . . the country of his genius’ (60) and ‘the obvious alternative of taking as Scott’s country solely the lands intimately related to him by natural ties – the region of his birth and upbringing and habitual domicile’ (59).3 However, he always concludes in favour of the biographical: It will, I think, be found demonstrable that in by far the greater number of instances the early environment of a writer is what counts most in his mature expression of nature as a background to the play of human emotions and passions and life lived. (129) 1

2

3

Virginia Woolf, ‘Literary Geography’, in The Essays of Virginia Woolf, Vol. 1: 1904–1912, ed. A. McNeillie (London: Hogarth Press, 1986), 32–6; 35. See also Barbara Piatti and Lorenz Hurni (‘Editorial: Cartographies of Fictional Worlds’, The Cartographic Journal 48 (2011): 218–23) where they play Woolf ‘s position off against Joyce (218–19); and Andrew Thacker, Moving through Modernity: Space and Geography in Modernism (Manchester: Manchester University Press, 2003) whose chapter on Woolf opens with this quotation. For another study of maps in modernism, see Eric Bulson, Novels, Maps, Modernity: The Spatial Imagination, 1850–2000 (New York and London: Routledge, 2000). William Sharp, Literary Geography (London: Pall Mall Publications, 1904), 23. This was a collection of articles he had written as a series for the Pall Mall Magazine in 1903–4. In The Literary Tourist: Readers and Places in Romantic and Victorian Britain (Basingstoke: Palgrave Macmillan, 2006) Nicola Watson tells us that ‘Scott country’ was ‘the first and for a while the most influential of these literary countries’ (170).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

18

From Literary Geography to Critical Literary Mapping

In its first manifestation, then, ‘literary geography’ is strongly authorial and national in nature, informed by a model of interpretation centred on authorial intention that ultimately works to ‘situate canonical literary texts in equally canonical landscapes’.4 In her study of late Victorian literary tourism, Nicola Watson poses the question of ‘how far literary tourism emerged as a side-effect of cultural nationalism, with the emerging literary canon seized upon in order to effect a sort of interiorised national mapping?’ (14). In fact, as late as 1979, we find works such as Margaret Drabble’s A Writer’s Britain or David Daiches’s Literary Landscapes of the British Isles operating in just this way.5 Both books articulate major socio-historical changes to Britain that shape actual and literary landscapes and feed into a larger cultural sense of Englishness (or Scottishness or Irishness) strongly associated with particular landscapes. It is notable that Daiches also frequently uses clean outline maps of Britain, Europe and elsewhere, onto which he maps thematic information (in ways that strongly anticipate Franco Moretti’s later map forms). Is there anything wrong with literary geography functioning primarily in support of literary tourism? The danger lies in the fact that the kind of mapping undertaken by the literary tourist, out in the landscape, privileges ‘accurate’ use of the map in terms of finding an exact match between imaginary and actual sites. It then attempts to assert knowledge of the actual site back onto the text which may be found wanting. As a result, it is in danger of inverting the relationship between imagined and actual geography because of an inability to recognise that the same space can have both a real and an imaginary existence. In the second half of the twentieth century, this authorial understanding of literary geography began to shift as it became the focus of early attempts at interdisciplinarity. In a 1978 article on ‘Literature and Geography’, Yi-Fu Tuan identified three ways in which ‘Literary art serves the geographer’: As thought experiment on possible modes of human experience and relationship, it provides hints as to what a geographer might look for when he studies, for instance, social space. As artifact it reveals the environmental perceptions and values of a culture: it serves the geographer who is also a historian of ideas. Finally, as an

4 5

Watson, Literary Tourist, 1. David Daiches, Literary Landscapes of the British Isles: A Narrative Atlas (London: Paddington Press, 1979); Margaret Drabble, A Writer’s Britain (London: Thames & Hudson, 1979).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

The Origins of Literary Geography

19

ambitious attempt to balance the subjective and the objective it is a model for geographical synthesis.6

In his introduction to the edited collection Humanistic Geography and Literature (1981), Douglas Pocock also attempted to determine different forms of engagement.7 These included the use of ‘literary landscape depiction’ (12); ‘the geography or topography behind literature’ (13); ‘the search for geography in literature, treating writings as a literary quarry from which to construct a more general literary topography’ (13); and landscape depiction as witness to the ‘cultural refraction of reality’ (13). As both accounts make clear, the nature of the interdisciplinary relationship articulated at this time was not an equal, reciprocal one, mainly because the primary focus for making a connection initially was from geography across to literature. This resulted in what Joanne Sharp has more recently criticised as a rather simplistic engagement with literary studies, involving ‘an ignorance of literary conventions’ and ‘a naivete about the form of literary writing’.8 Marc Brosseau also notes ‘the partial silencing of the literary text as a text’ that results from this onesided engagement.9 In an article on ‘Geography’s Literature’ Brosseau outlines a move from a historical geography which treats ‘regional novels as complements to regional geography’ (334) and which strongly privileges ‘the nineteenth-century realist novel tradition’ (337) to a humanistic geography centred upon ‘sense of place’ drawing on the work of early spatial theorists such as Edward Relph and Tuan, and to a more self-reflective practice that takes into account ‘the ways in which geography’s discourse constructs its objects’ as well as ‘a particular type of geography’ (349) for literature. In one of many articles offering more recent (2011) surveys of literary geography (again from the geographic side) Fernando Cabo Adequinolaza gives a clear account of the spread of spatial theory across disciplines, determined by ‘the spatial nature of the contemporary era’.10 Adequinolaza 6

7

8

9

10

Yi-Fu Tuan, ‘Literature and Geography: Implications for Geographical Research’, in Humanistic Geography: Prospects and Problems, ed. David Ley and Marwyn S. Samuels (Chicago, IL: Maaroufa Press, 1978), 194–206; 194, 205. Douglas C. D. Pocock, ‘Introduction: Imaginative Literature and the Geographer’, in his Humanistic Geography and Literature: Essays on the Experience of Place (London: Croom Helm, 1981): 11–26. See also his ‘Geography and Literature’, Progress in Human Geography 12 (1988): 87–102. Joanne P. Sharp, ‘Towards a Critical Analysis of Fictive Geographies’, Area 32.3 (2000): 327–34; 328. Marc Brosseau, ‘Geography’s Literature’, Progress in Human Geography 18.3 (1994): 333–53; 349. Fernando Cabo Adequinolaza, ‘The Spatial Turn in Literary Historiography’, Comparative Literature and Culture 13.5 (2011): 1–9; 3. This in turn draws upon an article in French by

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

20

From Literary Geography to Critical Literary Mapping

helpfully outlines ways in which the relationship between literature and space is being redetermined, identifying first ‘a geocriticism . . . related to research on cultural spaces and their association with the study of (in the language of Pierre Bourdieu) literary fields’ (4); second ‘a geopoetics whose sphere would coincide with the confluence of an element of external reality’ (4); and third ‘the sphere of the geosymbolic or, to be more exact, the geographic imaginary’ (4). When understood in such ways, the one-to-one relationship between either author and topographical region or between the two disciplines of literature and geography is simultaneously undermined and overpowered by a fully interdisciplinary entity drawing upon history, art, cultural and social studies in a ‘spread of spatial imaginaries’ (4). As the nature of the debate shows, in the late twentieth and early twentyfirst centuries the emergence of cultural geography as a force within human geography, alongside the development of spatial theory, has meant that disciplinary approaches are increasingly underpinned by shared underlying theoretical models, with key philosophers and theorists being drawn upon in different subject-specific ways (Heidegger; Merleau-Ponty; Foucault; Tuan; Certeau; Jameson; Lefevbre; Soja). Cultural geographers, coming out of social, political and spatial theory, share overlapping concerns and interests when they connect to literature out of their discipline. The work of figures such as Denis Cosgrove, Rob Kitchin, Chris Perkins, Nigel Thrift, Rob Shields and others bears upon this study when they explore the spatial as a key aspect of social and cultural practices.11 As a result of such conceptual and theoretical enrichment, the concept of literary geography might have been expected to fall out of use, but, instead, its meaning has become further refined and applied in a particular way to forms of literary mapping. Whilst earlier manifestations of literary geography frequently used maps to illustrate particular points, but did so largely in passing (Daiches being an excellent example) these become a central, redefining element of the concept of literary geography in the 1990s with the work of Franco Moretti and the subsequent extension of

11

Daniel-Henri Pageaux, ‘Eléments pour une géosymbolique. Littérature générale et comparée et géographie’, in Savoirs et Litterature / Literature, the Humanities and the Social Sciences, ed. Jean Bessière (Paris: Sorbonne Nouvelle, 2002), 75–92. Three of the most useful overviews of applied critical cartography in recent years are provided by: Rob Kitchin, Chris Perkins and Martin Dodge, ‘Thinking about Maps’, in Rethinking Maps: New Frontiers in Cartographic Theory, ed. Kitchin et al. (London: Routledge, 2009), 1–25; Jeremy W. Crampton, Mapping: A Critical Introduction to Cartography and GIS (Oxford: Wiley-Blackwell, 2010); and Robert T. Tally, Jr, Spatiality (London: Routledge, 2013). See also Tania Rossetto’s survey article, ‘Theorizing Maps with Literature’.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

The Origins of Literary Geography

21

his work into the digital medium by means of Geographical Information Systems (GIS) technology. In approaching Moretti with some sense of a genealogy for literary geography (as he continues to call it) both the strengths and weaknesses of the kinds of literary mapping he advocates can be seen more clearly. We can also see how Moretti’s work builds upon an earlier cruder conceptualisation of the relationship between literature and place in order to enable a far richer, deeper academic engagement with what it means to ‘map’ within this discipline. Moretti calls the first chapter of Atlas of the European Novel ‘Towards a Geography of Literature’ and immediately makes clear the nature of the interdisciplinary connection as he conceives it: [G]eography is not an inert container, is not a box where cultural history ‘happens’, but an active force, that pervades the literary field and shapes it in depth. Making the connection between geography and literature explicit, then – mapping it: because a map is precisely that, a connection made visible – will allow us to see some significant relationships that have so far escaped us.12

This primary aim – to make the map active, to make it a part of the interpretative process for literature – lies at the heart of what he is trying to do and, from the perspective of literary geography as outlined above, represents a great leap forward. Moretti himself makes this clear when he compares his work with previous literary Atlases ‘all with one thing in common: maps play in them a wholly peripheral role . . . they are colorful appendixes, that don’t intervene in the interpretative process’ (7). Moretti thus takes a relationship between map and text that is inherent, but latent, in previous works (many of which do contain maps of a similar kind to his) and develops a model of thematic mapping of literary data so that the visualisation of the map becomes part of the argument, with maps used ‘as intellectual tools’ (4). Moretti also significantly enlarges the focus of mapping itself (the content being mapped) from regions of significance to spatial authors or national literary landscapes of cultural value – the previous domain of maps in ‘literary geography’ – to a wider range of spatial fields. Moretti’s overarching aim is to use maps to uncover ‘how geography shapes the narrative structure of the European novel’ (7) which he applies in two ways: to ‘the study of space in literature’ and ‘literature in space’ (3). Within texts, the study uses maps to explore the intrinsically spatial, the ‘internal logic of 12

Franco Moretti, Atlas of the European Novel: 1800–1900 (London: Verso, 1998); 3. For a full account of Moretti’s larger aims across his works, see Tally, Spatiality, 99–111.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

22

From Literary Geography to Critical Literary Mapping

narrative: the semiotic domain around which a plot coalesces’ (5) but it also creates maps to illustrate larger patterns or a spread of movement for particular genres and forms such as: routes or character movements within a text; the spatial distribution of generic forms or chosen settings across Europe; or references to global places in periodical magazines. In seeking to show that the spatial is a vital and intrinsic part of literature, Moretti argues that maps are linked to plot action (as we would expect) but also that we need to understand ‘Geography as the foundation of narrative form’ (38), that ‘Space acts upon style’ (43) and – drawing upon Bakhtin’s account of the chronotope – that genre itself is strongly spatialised: ‘Each genre possesses its own space, then – and each space its own genre: defined by a spatial distribution – by a map – which is unique to it’ (35). These are strong claims, but ones that I agree with and (albeit in different ways) that I want to make in this book. However, where Moretti generates his own maps and projects them onto real-world spaces, my focus is on pre-existing maps and the dynamic between map and text for writer and reader. In his later study, Graphs, Maps, Trees Moretti again articulates his method, although the tendency towards quantitative methods emerges more strongly and defensively than in Atlas. His account of ‘what literary maps do’ is worth giving at some length: What do literary maps do . . . First, they are a good way to prepare a text for analysis. You choose a unit – walks, lawsuits, luxury goods, whatever – find its occurrences, place them in space . . . or in other words: you reduce the text to a few elements, and abstract them from the narrative flow, and construct a new, artificial object like the maps that I have been discussing. And with a little luck, these maps will be more than the sum of their parts: they will possess ‘emerging’ qualities, which were not visible at the lower level. (Graphs, 53)

From a literary-critical perspective, Moretti’s tone here is deliberately provocative (in a way that partly accounts for the mixed reception his work has received within that discipline).13 From the self-conscious arbitrariness of ‘You choose a unit . . . whatever’ to the italicised terms ‘reduce’, ‘abstract’, artificial’ the author goes out of his way to make his method

13

Moretti has a tendency to attack the underlying methods and raison d’être of his home discipline. In Atlas he suggests: ‘one day, who knows, a literary criticism finally transformed into a comparative historical morphology may be able to rise to the challenge’ (195). At this point he is still only really making a rationale for a more ‘scientific’ way of working with texts. In Graphs his position hardens into a definition of ‘Distant reading’ which is set against ‘close reading’ (3) and he calls for ‘a more rational literary history’ (4). See also Distant Reading and Tally, Spatiality, 99–111.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

The Origins of Literary Geography

23

sound far more reductive than it actually is. At bottom his aim is simple and effective, posing central questions for any form of literary map: ‘There is a very simple question about literary maps: what exactly do they do? What do they do that cannot be done with words?’ (35). Another secondary debate in Graphs that is relevant to a larger understanding of literary mapping concerns the nature of the maps themselves. Moretti responds to a criticism of the first book that ‘the figures of the Atlas . . . are not really maps but diagrams’ (54). The critic questions whether what Moretti is discussing is really ‘geography’ at all or just ‘geometry’. In response Moretti acknowledges: The diagrams look like maps, yes, because they have been ‘superimposed on a cartographic plane’ but their true nature emerges unmistakably from the way I analyse them, which disregards the specificity of the various locations, to focus almost entirely on their mutual relations; which is indeed the way to read diagrams, but certainly not maps. (54)

Although he does not directly articulate this, Moretti is effectively undertaking thematic mapping in relation to data derived from literary texts. In cartographic terms, a thematic map is usually quantitative and involves the comparison of a single element distributed across different areas; ‘themeoriented maps stress a relatively small number of variables. Frequently only one variable is shown.’14 So, what Moretti describes here as a diagram could equally well be understood as a thematic map, the focus of which might be, for example, the place of habitation of protagonists of Parisian novels for which ‘relations among locations are more significant than locations as such’ (Graphs, 55). The variable upon which the thematic map is focused does not exist in actuality, but only within the realm of fiction. If it is true that Moretti superimposes geometry onto geography, as he happily acknowledges (‘If I keep making diagrams, then, it is because for me, geometry ‘signifies’ more than geography’ (56)) his position is easily defensible. These are still ‘maps’ in any broad definition of what a map can be (as given here in the Introduction). Moretti’s emphasis on geometrical relations is significant, though, because – for the particular form being mapped (the literary work) – this often provides the most immediate use value of the map in ways that are central to the meaning and interpretation of a literary work. More problematic for Moretti than the (not-that-relevant) question of whether his ‘maps’ are ‘diagrams’ is the way that his approach to the 14

Muehrcke and MueHrcke, Map Use, 80.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

24

From Literary Geography to Critical Literary Mapping

mapping of texts leaves little room for the full exploration and interpretation of the rich and dynamic layers of meaning that occur in relation to what might be called a ‘sense of place’ for literature or, more philosophically, ‘literary spatiality’. This is the major weakness of his work in relation to literary studies because this aspect of the text is fundamental to the ‘human geometry’ that connects humans to each other and to their environment. The structuralist nature of his enterprise is reflected here. For literary studies more broadly, structuralism works well at the larger level but fails to fully integrate itself with actual textual materials. In just the same way, for Moretti, there is little ongoing dialogue between the maps and the texts to which they correspond (in fact there is little direct engagement with the language of texts at all). Instead, their dominant usage is to present a dramatic visual finding with consequences for the overall spatial shape or dynamic of a work. In 2011 Barbara Piatti and Lorenz Hurni, as co-directors of the project ‘A Literary Atlas of Europe’, offered a definition of literary geography as: the observation that fictional plots are set along a scale of localisations that range from the realistically rendered, highly recognisable to the completely imaginary. The geography of fiction follows its own distinct rules, since literature can create any space, without physical restrictions. . . It belongs to the ambitious goals of literary geography to find out more about those rules and to demonstrate that the spatial dimension of fictional accounts can actually be one key to the understanding of the whole plot.15

The aims are again structuralist; they involve determining the underlying conventions by which the ‘rules’ operate and, from this, increasing understanding of spatial meaning. One can see why this kind of project is open to accusations of reductiveness from literary critics. For example, there is a tendency towards self-contradiction: if literature can ‘create any space without physical restrictions’ then surely there are ‘no rules’ or at least no limit to them? But Piatti and Hurni have to be programmatic – they are creating programmes after all. Because they are centred upon map-making rather than interpretation, their focus is on the limits of what can be done, what is able to be visualised and spatialised in order to produce maps. It is unsurprising then that, for them, Woolf’s comment in support of unmappable imaginative extent (with which this chapter opened) is set against that of James Joyce (which provides the underlying rationale) that ‘I want to give a picture of Dublin so complete that if the city one day suddenly 15

Piatti and Hurni, ‘Editorial’, 218.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Literary Geography or Literary Cartography?

25

disappeared from the earth it could be reconstructed out of my book.’16 Piatti and Hurni also introduce a new term – that of ‘literary cartography’ – defined as a ‘subdiscipline’ or ‘ancillary science’ to literary geography: Obviously the two terms are linked in a logical, hierarchical way: while literary geography is the overall topic, literary cartography provides one possible approach by using a symbolic language; spatial elements of fictional texts are translated into cartographic symbols, which allows new ways in exploring and analysing the particular geography of literature. (‘Editorial’, 218)

Here, literary cartography is centred upon (digital) map-making with realworld reference points in a highly practical way rather than in terms of a layered and nuanced process for all participants. Piatti and Hurni conclude: It goes without saying that literary cartography follows the ‘Joyce-line’. One of its traditional starting points is precisely the assumption that a large part of fiction indeed refers to the physical/real world . . . by using an almost infinite variety of options to do so. (218)

This kind of statement seems to be in danger of returning to the representational confusions of the literary tourist, conflating real and imagined place and space by assuming (or, more pertinently, leading the user of the map to assume) a direct correspondence between actual and fictional worlds that cannot exist.

Literary Geography or Literary Cartography? Moretti’s (and Piatti’s) repositioning of maps at the centre of literary geography shifted the methodological focus from analytical comparison between topography in the world and spatial description in the text, to a comparison between different forms of representation (and ultimately towards a position that questions all such forms and any sense of priority).17 Literary cartography also now emerges as a new term that complicates the original interdisciplinary duality. The intersection of literature – geography – cartography – and digital humanities (by means of GIS technologies) means that a relatively straightforward initial interdisciplinary relationship now 16

17

Frank Budgen, James Joyce and the Making of Ulysses, quoted in Piatti and Hurni, ‘Editorial’, 218–19. The concept of ‘post-representational’ mapping emerges here. See particularly Kitchin, Perkins and Dodge, ‘Thinking about Maps’ (10) and their visualisation of Alan MacEachran’s ‘cartography cube’ (8).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

26

From Literary Geography to Critical Literary Mapping

becomes tetradic and thus increasingly complex. Not only are there four different disciplinary starting points but, within these, there are multiple positions and understandings of that relationship as well. So, for example, geographers do not generally consider themselves cartographers; cartographers traditionally divide into practical map-makers and theorists; and GIS represents a major threat to both camps that may or may not be understood to be housed within the discipline.18 In one of a series of survey articles for literary geography by various commentators, from the late 1980s onward, Angaharad Saunders states that ‘Little more than a decade ago it was possible to identify a literary geographic tradition which stretched back to the early twentieth century’ but this ‘has been subject to some profound epistemological transformations’.19 As a result, methods and approaches begin to proliferate to a dizzying degree. While Moretti ‘remains a major reference point’, now his ‘literary geography’ stands for just one kind of method, not the sole determining one.20 Questions begin to emerge about the core relationship between the two primary subjects. Saunders asks: ‘[I]s it a geography of literature or a literary geography?’ (439) while Sheila Hones notes that, ‘One of the most fundamental issues in literary geography has been (and still is) the question of whether its key terms refer to discipline or to subject matter.’21 A comparison of two recent attempts to define literary geography for the twenty-first century reveals the complexities involved.22 In Spatiality Robert Tally employs three terms – ‘literary cartography’; ‘literary geography’; ‘geocriticism’ – to structure his study, each referring to specific ways of understanding and responding to literary place and space. For him, literary cartography concerns the writer as map-maker – ‘the act of writing might be considered a form of mapmaking or a cartographic

18

19

20 21

22

For a full account of the debate around GIS within geography, see Crampton, A Critical Introduction to Cartography and GIS. Angharad Saunders, ‘Literary Geography: Reforging the Connections’, Progress in Human Geography 34 (2010): 436–52; 436, 437. A list of key survey articles (all referenced elsewhere in this chapter) might read: Pocock (1988); Brosseau (1994); Sharp (2000); Ogborn (2005–6); Thacker (2005–6); Hones (2008); Saunders (2010); Rossetto (2014). Rossetto, ‘Theorizing Maps’, 516. Sheila Hones, ‘Text As It Happens: Literary Geography’, Geography Compass (2008): 1301–17; 1303. This comparison is suggested by Hones’s own observation in relation to Spatiality that ‘the space-time of literary geography looks completely different depending on where one is standing’ (Literary Geographies: Narrative Space in Let the Great World Spin (Basingstoke: Palgrave Macmillan, 2018), 167).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Literary Geography or Literary Cartography?

27

activity’ (45) – literary geography relates only to the reader – ‘the critical reader becomes a kind of geographer’ (79) (with a strong focus on Moretti’s model) – and geocriticism (pace Bernard Westphal) concerns spatial and cultural theory as ‘itself a crucial domain of spatiality’.23 In comparison, for Sheila Hones in Literary Geographies, everything is embraced by the term ‘literary geography’. She articulates three kinds of literary-geographical space: the ‘fictional space generated in the event of the text’; ‘intertextual literary space’; and ‘the sociospatial dimension of the collaboration of author, editor, publisher, critic and reader’.24 Hones shifts the emphasis from geography to literature to such a degree that ‘geography’ ceases to be topographical and becomes that of the production and reception of the text itself. Equally, this space is post-representational to the point of being asserted as ‘real’. As Hones articulates it: ‘the literary-geographical space in which fiction happens is a real space’ (Literary Geographies, 9). As the interdisciplinary relationship becomes more complex, then, it is in danger of breaking down altogether, with practitioners from different individual disciplines sharing some degree of common ground, but defining the concept in radically different ways. The current solution to this problem appears to be offered by a redetermining of the approach at a macro-level that can potentially apply across the whole, in terms of a shift to an ‘experiential’ or ‘practice-based’ method. This shift was introduced by Rob Kitchin, Chris Perkins and Martin Dodge and moves the field on from ‘a representational to a processual understanding of maps’ (clearly underpinning both approaches above).25 Kitchin et al. articulate a move away from ‘ontology’ and towards ‘ontogenesis: maps as practices’, arguing that both ‘Mapmaking and map use are understood as processual in nature, being both embodied and dynamic’ and that such mapping should be ‘conceptualized as a suite of cultural practices involving action and affect’.26 As Tania Rossetto puts it (in her thorough attempt to survey the field): ‘Maps are practices, they proceed from action, rather than being grounded in power’.27

23

24 25

26 27

Tally, Spatiality, 45, 79, 112. Tally employs the term ‘cartographics’ to ‘designate a set of critical practices that seek to engage with the issues of spatial relations in connection with cultural and social theory’ (113). Hones, Literary Geographies, 8. Kitchin, Perkins and Dodge, ‘Thinking about Maps’, 1–25. See, for example, David Cooper and Gary Priestnall, ‘The Processual Intertextuality of Literary Cartographies: Critical and Digital Practices’, The Cartographic Journal 48 (2011): 250–62. Kitchin, Perkins and Dodge, ‘Thinking about Maps,’ 16, 17. Rossetto, ‘Theorizing Maps’, 514.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

28

From Literary Geography to Critical Literary Mapping

Where does this book position itself in relation to such shifts? The strong focus on reading maps and on the visual/verbal juxtaposed relationship means that this study was always about an active engagement with cartographic forms. Chapter 6, on Tolkien, not only understands maps as practices but explores the act of mapping in and through creative process; while the final chapter attempts to articulate a new model for digital literary mapping coming out of the cognitive mapping of literary place and space. In all cases, however, the study addresses actual empirical maps and actual acts of (internal or external) mapping. In this sense, it both embraces an active model and partakes of the kind of ‘re-cartographization of literary mapping practices’ that David Cooper calls for in his article on ‘Critical Literary Cartography’ (see ‘Ending Where We Began’ below).28

Critical Cartography Whilst reading against the map is the current mode du jour – and vitally informs this book – it is still worth heeding the cartographers and remembering, as John Krygier and Denis Wood point out, that this involves both a ‘conflation of cartography . . . with the whole of map-making’ and that ‘mapmaking has been perpetually transformed, all but dialectically, by successive critiques’.29 In other words, map-making exceeds the profession of cartography, and the most recent ‘cartographic turn’ is only one of many. In part, as with all critiques, it stems from a reaction to a preceding paradigm that it is then used to condemn: in this case, that of the post-war period which sought to assert a more scientific model and focused on the map as a communicative form. Critical cartography shifts the focus from the map itself to the social forces with which it is bound up and asks ‘how the body of the subject was constructed by the map, that is, how the map oppressed, subjugated, or otherwise impinged on people’ 342). One way in which this particular ‘turn’ is distinctive, however, is that it took place within (and was directly aligned with) a larger spatial turn for the Humanities at a time when shared theoretical underpinnings enabled new forms of interdisciplinarity – and

28

29

David Cooper, ‘Critical Literary Cartography: Text, Maps and a Coleridge Notebook’, in Mapping Cultures: Place, Practice, Performance, ed. Les Roberts (Basingstoke: Palgrave Macmillan, 2012): 29–52; 49. J. Krygier and D. Wood, ‘Critical Cartography’, in International Encyclopedia of Human Geography, ed. Rob Kitchin and Nigel Thrift (Amsterdam: Elsevier Ltd, 2009), 340–4; 340.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Critical Cartography

29

the spread of methods and approaches across disciplines – in ways that were not previously possible. In his account of ‘the spatial turn’ in Spatiality, Robert T. Tally suggests that the shift from a late nineteenth-century concern with time, linearity and teleology to a late twentieth-century concern with (modernist and postmodern) space also finds its origins in the post-war period: The geopolitical organization and disruptions in the postwar era called attention to the distinctively political essence of geography, as the forces of decolonization, as well as those of neocolonization, made clear that the spaces of the map were not uncontested. (13–14)

This shift picks up speed with ‘the new spatiality of the postmodern’ (41) encouraged by globalisation and the forces of late capitalism as articulated by Frederic Jameson and others.30 In this sense, then, the ‘cartographic turn’ is an inevitable part of the larger spatial one, fully aligned with it. In a traditional account of the history of cartography, a history of maps, centred upon Enlightenment principles, identifies key discoveries and turning points in a journey towards ever-increasing systematisation and the development of more and more precise means of accurately representing the topography of the world. The next chapter traces major turning points in the history of cartography onto the emergence of the literary map. Jess Edwards sums up the established position: The traditional positivist history of cartography is teleological and idealist. Traditional cartographic history assumes consistent development towards a modern scientific practice founded on the discipline of geometric measurement and projection and treats individual maps as neutral contributions to a Platonic archive of geographic knowledge.31

Such an understanding of cartography is centred upon ways in which the content of the map is defined by its intended purpose. Crampton and Krygier assess the mid-twentieth-century influence of the cartographer Arthur Robinson in just such terms: Robinson’s great achievement is that he included the map user in the equation. This design focus had the goal of improving the efficiency and functionality of maps as communication devices via empirical experimentation.32 30 31

32

I will return to Jameson in Chapter 8 in relation to cognitive mapping. Jess Edwards, ‘Study, Marketplace and Labyrinth: Geometry as Rhetoric’, New Formations 57 (2005–6): 126–44; 127. Jeremy W. Crampton and John Krygier, ‘An Introduction to Critical Cartography’, ACME: An International E-Journal for Critical Geographies 4.1 (2006): 11–33; 20.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

30

From Literary Geography to Critical Literary Mapping

The dominant position prior to the 1980s, then, could be described in terms of a ‘map communication model’ (24). However, earlier cartographers were fully aware of the distorting effects of selection and generalisation necessary as part of the map-making process. Muehrcke and Muehrcke state that ‘what makes a map so useful is its genius of omission’ whilst Robinson points out that ‘the great power of the mapping process lies in its ability to provide fresh, insightful perspectives, sometimes even distorted ones, on our environment’.33 The primary difference between these cartographers and the critical cartography that comes after them is that they view such effects positively, as a central element of how a map works, rather than approaching them as in any way problematic. Critical cartography, on the other hand, involves a paradigm shift from trusting the map (implicitly understood as a scientific, objective entity) to mistrusting it (for the subjective, human, cultural and ideological elements that produced it). Once the map interpreter is no longer in thrall to an Enlightenment privileging of the scientific drive towards perfection he, or she, is also compelled to acknowledge that any map is partial and biased by its very nature (even a map that does not appear to be so). Monmonier states that: ‘A good map tells a multitude of little white lies; it suppresses truth to help the user see what needs to be seen’.34 In other words, if all the information that could be included on the map was on the map, it would be useless. Monmonier calls this the ‘cartographic paradox’: ‘To present a useful and truthful picture, an accurate map must tell white lies’ (1). This shift directly replicates that made earlier in literary studies, from a decontextualised new critical reading of the text, to a fully historicised self-aware contextualisation of it by new historicism and a sceptical deconstructive reading of the text against itself. As is well known, critical cartography applies theoretical principles from other disciplines (most notably literary and art theory) back onto cartographic representation. The leading figure here is, of course, Brian Harley, but the work of those coming after him in his discipline further consolidated the field in the late twentieth and early twenty-first centuries. In ‘Thinking about Maps’, Kitchin, Perkins and Dodge provide a concise outline of different approaches post-Harley in terms of: a critique of the social constructivist model (Crampton); ‘maps as inscriptions’ (Pickles)

33

34

Muehrcke and Muehrcke, Map Use, 10; and A. H. Robinson et al., Elements of Cartography, 5th ed. (New York: John Wiley & Sons, 1984), 16. Mark Monmonier, How to Lie with Maps (Chicago, IL, and London: University of Chicago Press, 1991), 25.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Critical Cartography

31

(12); ‘maps as propositions’ (Woods and Fels) (13–14); ‘maps as mobiles and actants’ (Bruno Latour); before articulating their own account of ‘maps as practices’ (16–23). Harley’s early Foucauldian critique in terms of power extends over time into exploration of different aspects of power in relation to particular national, racial or ethnic groups, gender and so on, with counter-hegemonic mapping emerging to challenge dominant official map forms.35 Christian Jacob almost literally picks up where Harley leaves off, acknowledging his direct influence – ‘Harley’s theoretical papers were for me a major inspiration’ (xvi) – but offering a less highly theorised and more reflective approach: ‘The Sovereign Map is not a dogmatic book. It does not construct a theory of maps . . . it implies attempts, strategies, playing with questions and problems’ (xvi)’. This approach reincorporates the idea of the map as a form of communication but from a theorised position that understands it as ‘a medium of communication which implies at the same time an attempt to encode values and meanings and the various strategies of reception on the part of the user’ (xv). For Jacob, this is approached primarily through a meeting point between the two sides of map-maker and map-reader, both understood visually: ‘The map originates in the meeting of graphic gesture and the visual pathway that identifies it’ (2). In a comparable way, but for maps in literature, we can interpret the map as a meeting point for the making of meaning by both author and reader. A semiotic approach self-evidently underpins the cartographic selfcritique. Indeed, it is this that enables Harley’s application of approaches more common to the Humanities, to his own traditionally scientific field. In his seminal essay ‘Maps, Knowledge, and Power’, Harley explicitly adopts an iconological approach, defining maps as ‘images with historically specific codes’.36 At the same time, he almost immediately qualifies this approach by noting the limitations of a narrow semiotic reading and suggesting an alternative: The idea of a cartographic language is [also] preferred to an approach derived directly from semiotics, which, while having attracted some cartographers, is too blunt a tool for specific historical enquiry (53).

35

36

The term ‘counter-mapping’ was coined by Nancy Peluso. See N. L Peluso, ‘Whose Woods are These? Counter-Mapping Forest Territories in Kalimantan, Indonesia’, Antipode 27 (1995): 383–406. J. B. Harley, ‘Maps, Knowledge, and Power’, in The New Nature of Maps, ed. Paul Laxton, intro. J. H. Andrews (Baltimore, MD: Johns Hopkins University Press, 1997), 52–81.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

32

From Literary Geography to Critical Literary Mapping

In other words, we should not simply be interpreting the use of codes on any individual map or kind of map but thinking in terms of maps as their own visual language, or discourse. Unsurprisingly, Harley rapidly shifts his focus from the sign system in play for any given map and onto the context surrounding it, placing emphasis on the external ideological conventions that determine our reading of the map. Thus, his primary interest is in how maps give voice to the implicit conventions that produce them: ‘They are a class of rhetorical images and are bound by rules which govern their codes and modes of social production, exchange, and use just as surely as any other discursive form’ (54). In Saussurean terms, Harley treats the map as ‘synchronic’ in order to prioritise the importance of understanding the full context of the map’s production rather than analysing the object itself: In any iconological study it is only through context that meaning and influence can properly be unravelled. Such contexts may be defined as the circumstances in which maps were made and used. They are analogous to the ‘speech situation’ in linguistic study and involve reconstructions of the physical and social settings for the production and consumption of maps (56)

Harley’s iconological reading essentially distinguishes between two major levels of communicated meaning in a way that proves highly relevant to maps in fiction. He identifies: a surface or literal level of meaning but also a ‘deeper’ level, usually associated with the symbolic dimension in the act of sending or receiving a message . . . It is often on this symbolic level that political power is most effectively reproduced, communicated and experienced through maps. (54)

His interest lies in uncovering the ‘hidden rules’ of that discourse by exploring the different ways the map manipulates the information it presents (e.g. through geometrical distortion; silence; reification of symbols). Only towards the end of the paper does he acknowledge that: ‘We still have to grapple with maps as unique systems of signs with codes that may be at once iconic, linguistic, numerical, and temporal, and as a spatial form of knowledge’ (79). This challenge is taken up by Denis Wood in the first edition of The Power of Maps where he enlarges Harley’s four codes (iconic, linguistic, numerical and temporal) to ten.37 Following Harley, but with even more explicit and direct application of literary theory to cartography, Wood 37

Denis Wood with John Fels, The Power of Maps (New York: Guilford Press, 1992), 111–13.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Critical Literary Mapping

33

famously used Roland Barthes’s account of ‘myth’ to deconstruct the meanings held in a ‘North Carolina Official Highway Map 1978–79’. He distinguishes between ‘a two-tiered semiological system in which the simpler is appropriated by the more complex’ (103). In other words, the map presents itself as a neutral, factual, representation of roads in a particular bounded area but, beneath this, what it really seeks to do is to project the attractiveness of North Carolina as a state. He concludes that ‘maps constitute a semiological system (that is a system of values) that are ever vulnerable to seizure or invasion by myth’ (107). The first-order level of signification is easily appropriated by a second-order one (although this is not always consciously apparent to the user). Finally, Jacob revitalises this issue when he considers the relationship between maps and writing in a full material sense, comparing ‘writing and drawing’ (190). Jacob distinguishes between the ‘archipelago text’ and ‘the written map’ (190) as two extremes for the relationship between visual and verbal on the map itself: ‘In one case the text turns into a map; in the other, the map turns into a text’ (191).

Critical Literary Mapping Perhaps the first attempt to articulate a ‘critical literary geography’ in a way that bears directly upon this study is that made by Andrew Thacker who poses the question: ‘What would [such] a “critical literary geography” look like in practice? Would it be a critical practice based on cartography?’38 As well as identifying limitations in Moretti’s objective response to maps, Thacker considers other ways in which literature and cartography can relate. These involve the most common metaphorical approach in which a literary text is ‘mapped’ or spatial metaphors are linked to thematic elements, as well as one in which geographic representations of space are compared to their literary representations. Thacker’s (essentially Lefebvrean) model is not specifically cartographic in focus, ultimately being concerned with broader ‘interaction between spatial forms and social space’ (63). However, he does helpfully identify different meaningful literarycritical ways in which this interaction might be explored in terms of ‘Typography and layout on the page; the space of metaphor and the shifting between different senses of space within a text; or the very shape 38

Andrew Thacker, ‘The Idea of a Critical Literary Geography’, New Formations 57 (2005–6): 56–73; 60. See also Moving through Modernity. Damian Walford Davies also develops his argument out of Thacker’s influential article in Cartographies of Culture (6–7).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

34

From Literary Geography to Critical Literary Mapping

of narrative forms’ (63). He raises the possibility of mapping material spaces, suggesting that ‘we should reconnect the representational spaces in literary texts not only to the material spaces they depict but also reverse the movement’ (63). Thacker also directly anticipates my work here when he suggests that ‘another line of investigation for a critical literary geography might be to analyse the occurrence of maps and mapping in specific texts, analysing how cartography functions as an instance of visual culture in such texts’ (64). Cooper draws upon Thacker when he makes a stand against metaphorical forms of literary-critical cartography in favour of a return to spatial readings of actual maps in texts: As part of this spatialization of literary studies . . . the verb ‘to map’ became increasingly synonymous with the concatenated practices of reading, textual analysis and critical taxonomy rather than the surveillance and representation of the geographic landscape.39

Cooper advocates the use of critical cartography as the primary theoretical approach for the ‘geocritic’ interested in ‘thinking about the symbolic textuality of “actual” literary cartographies’ (33) in ways that relate to this study. Finally, Rossetto, in the last part of ‘Theorizing Maps with Literature’, starts to explore ‘alternative directions for thinking [about] maps across different cartographic theoretical “domains”’ (521). She shifts the focus towards the ‘challenging relationship between cognitive and postrepresentational cartographic theories’ (521) and also undertakes an exemplary analysis of Cormac McCarthy’s The Road from a perspective which views ‘literary repertoires’ as ‘an immense archive of living maps and emergent mapping practices’ (524). It seems likely that the intersection between literature and geography/ cartography is heading in an increasingly subjective direction with its focus on maps and texts as sites for experiences and with a degree of liberation from levels of representational priority this brings with it. While the experiential approach is attractive (and to some extent I employ it here), taken to its logical extreme it has the danger of making all aspects of literature ‘mappable’ and thus of losing the interdisciplinary focus for literary mapping which has been simultaneously problematic and invaluable to both sides. If we seek to retain that focus then it may be more helpful to return to core elements held in common, or running in parallel,

39

Cooper, ‘Critical Literary Cartography’, 30.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Critical Literary Mapping

35

between literature and geography or cartography – elements that brought them together in the first place. In Elements of Cartography, Arthur Robinson himself employed a dramatic metaphor to describe the internal structure of the discipline: We can liken cartography to a kind of drama played, at a minimum, by two performers with two stage properties. The actors are (1) the cartographer, and (2) the map user. In some instances the same person can take both parts, but the parts are played sequentially.40

Such an image reminds us that the map is a connective object between what lies behind it – its own historicity, the context of its production and of the meanings it presents or withholds – and what it projects (a communicative object that holds meanings and signs to be read and used). We might also note that this core division at the heart of cartography – which is also that between practical map-making (and advice on this) as opposed to theoretical or historical interpretation of maps – is replicated in literary studies by the distinction between a pre- and postdeconstructive critical model that is complicated by the second distinction between work and text (made, material object as opposed to semantic content). It is the compatibility of the disciplines that enables Harley and others so easily to apply theories developed from History and Literature back onto Cartography in the first place. This in turn creates a kind of virtuous circle, predisposing critical cartography for its reapplication to literature. So, in ‘Maps, Knowledge, and Power’ Harley describes how: In any iconological study, it is only through context that meaning and influence can properly be unraveled. Such contexts may be defined as the circumstances in which maps were made and used. They . . . involve reconstructions of the physical and social settings for the production and consumption of maps, the events leading up to these actions, the identity of map makers and map users, and their perceptions of the act of making and using maps in a socially constructed world. (56)

What Harley describes in relation to cartography can apply directly and easily to both literary works and to maps in literature. The literary map has its own context of production and use: that of its making (by the author, or by others with the author’s permission and approval); as well as its relation as a map form to other historical map forms; to the historical context of the

40

Robinson et al., Elements of Cartography, 16.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

36

From Literary Geography to Critical Literary Mapping

novel (the time of its writing or the period in which it is set); and to the reader’s own historicity. Such contexts can easily be brought to bear upon literary interpretation. When we consider the parallels between the disciplines, then, and the fact that there is both an innate disciplinary comparison and a shared theoretical core for analysis, this seems to provide a good platform from which to develop a more advanced interdisciplinary model. Such an approach needs to be capable of analysing text or map (or both) within each discipline and across and between them, in a full context historically and generically. It also needs to be capable of analysing the represented form in a way that allows for the full range of meanings held in the multiple levels of hermeneutic code; intended and unintended meanings; and in terms of visible and hidden agendas. It needs to incorporate the full materiality of text, or map, or both, and the interaction of visual and verbal meanings on and off the map. Finally, it needs also to allow for acts of creative mapping and re-mapping by reader and writer – and here the processual model can come to the fore and connect to cognitive narratology. Critical literary mapping is about exploring the relationship between verbal and visual representations of space and place in a fully dynamic way to open up the nature of any literary work, but particularly one that contains a strongly spatial or topographical element (or for which the presence of a map alongside the text signals the centrality of spatial meaning).

Material Juxtaposition: Map and Text A critical literary mapping need not only concern itself with actual maps given alongside texts but, where it does so, as in this study, the implications of that material juxtaposition for analysis must be taken into account. The final section of this chapter therefore necessarily focuses on the material dimensions of literary mapping. How do we address a juxtaposed visual– verbal relationship within a predominantly verbal form? Since the empirical presence of the map within the covers of the book is a vital element bearing upon meaning, the relationship between fictional map and text here is primarily considered in material terms by applying the concept of the paratext to the fictional map. Across a number of books, French critic and theorist Gérard Genette defines different ways in which texts form relationships with other texts, articulating a larger structure of transtextuality, of which paratextuality is

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Material Juxtaposition: Map and Text

37

merely a part.41 Genette defines paratextual materials as ‘accompanying productions’ to the text that ‘surround it and extend it, precisely in order to present it’ (Paratexts, 1). In Genette’s account the paratext is centred upon the concept of ‘a threshold, or – a word Borges used apropos of a preface – a “vestibule” that offers the world at large the possibility of either stepping inside or turning back’ (2).42 This central spatial sense determines a distinction between the two secondary forms of ‘paratext’ that Genette is concerned with: a ‘peri-text’ (within the book) and an ‘epitext’ (outside it).43 However, as Genette makes clear elsewhere, ‘the paratext is neither on the interior nor on the exterior: it is both; it is on the threshold’.44 The paratextual space is not just a territory or zone but also possesses distinctive characteristics and operates for both writer and reader with a clear and simple function: ‘to present and comment on the text’ (345). As this function suggests, the concept of the paratext is also strongly authorial. In his conclusion, Genette states: ‘the main issue of the paratext is not to “look nice” around the text but rather to ensure for the text a destiny consistent with the author’s purpose’ (407). If the paratext is concerned with authorial mediation, it is also open to the possibility of manipulation: The effect of the paratext lies very often in the realm of influence – indeed, manipulation – experienced subconsciously. This mode of operation is doubtless in the author’s interest, though not always in the reader’s. (409)

As we shall see, this point is particularly relevant to the fictional map as a form of paratext that may, or may not, actually operate on the reader in the way intended by the author. In view of the complexities involved in working across visual and verbal forms it is perhaps not surprising that Genette’s study deliberately chooses to focus on textual paratexts.45 Ostensibly, this is because he wishes to keep the focus on material that ‘shares the linguistic status of the text’ but that,

41

42 43

44

45

Genette articulates this theory across a trilogy: The Architext: An Introduction, trans. Jane E. Lewin (Berkeley, CA: UCLA Press, 1992); Palimpsests: Literature in the Second Degree, trans. Channa Newman and Claude Doubinsky (Lincoln and London: University of Nebraska Press, 1997) and Paratexts: Thresholds of Interpretation, trans. Jane E. Lewin (Cambridge: Cambridge University Press, 1997). The French title of the book is simply Seuils (‘thresholds’). Technically this entire discussion should be adopting the more specific secondary term ‘peritext’, but since ‘paratext’ is the more familiar general term, I have used this instead. Gérard Genette, ‘Paratexts’, Poétique 69 (Paris, 1987) (quoted in Richard Macksey’s foreword to Paratexts, xvii). Paratexts, 7. Genette states: ‘almost all the paratexts I consider will themselves be of a textual, or at least verbal, kind’ (7).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

38

From Literary Geography to Critical Literary Mapping

in turn, raises the question of whether a visual paratext (which is not of the same order or status) functions in the same way.46 It is at least possible that one reason that a map in a literary work is often overlooked or undervalued may be because it is defined as a paratext, which allows it to be treated as marginal, even where this is clearly not the case. (We may want to consider, then, whether the meaning and value of a literary map is constrained by its paratextual identity or potentially exceeds it.) Nonetheless, while bearing such issues in mind, the concept of the paratext can still help to identify a number of core characteristics that the literary map possesses, as well as the potential contradictions it embodies. Genette usefully identifies key factors that relate to and determine paratextual materials: [D]efining a paratextual element consists of determining its location (the question where?); the date of its appearance and, if need be, its disappearance (when?); its mode of existence, verbal or other (how?); the characteristics of its situation of communication – its sender and addressee (from whom? to whom?); and the functions that its message aims to fulfil (to do what?). (4)

This definition identifies three core aspects of the material relationship between map and text within the covers of a book: their relative location as part of the literary work; relative duration and temporality; and the situation of communication established. By briefly considering each element in turn, I want to examine both the extent to which the literary map truly does partake of such paratextual characteristics and the extent to which the meaning of the map is bound up with the meaning of the text (and vice versa) to a greater extent than this concept perhaps implies. The normative position for a literary map might be determined as at the front of the book where it is most likely to appear after the title page as a ‘specific illustration’ in Genette’s terms (24). However, there are at least three other locations that occur regularly: as endpapers (on the inside front, and often also back, cover); within the text; at the back of the book. To what extent does the physical position of the map affect the dynamic between map and text? Where a map appears as an endpaper, it is likely to involve a strong design element intended to have an aesthetic appeal. Monmonier reminds us of the tensions in play here between the iconic power of the map (and thus its value to publishers in making a literary

46

In his conclusion, Genette clearly authorises the definition of visual materials as paratext. He also refers to three maps or plans (Perec’s floor plan in Life: A User’s Manual; Faulkner’s map of Yoknapatawpha County; Eco’s plan of the abbey in The Name of the Rose) (405).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Material Juxtaposition: Map and Text

39

work attractive and aesthetically pleasing) and the communicative value of its content in relation to the narrative: ‘Visual balance is the prime goal in page layout, so maps are positioned to make the book look good rather than to make the reader look at the maps’ (243). At the same time, the likely emphasis on design in an endpaper map misleadingly encourages the reader to respond to it as primarily illustrative or decorative and thus serves to devalue its integrative meaning. Two further non-standard locations that affect the integration of visual and verbal meaning occur when the map is either given within the text or at the back of the book. The choice of location within a text is not that common and is closely associated with particular genres, such as detective fiction, in which the map is frequently presented in a physically integrated way. This encourages the reader to view the map as a kind of ‘clue’ or puzzle and to treat it as something of use at a particular stage of the investigation rather than to read it more holistically (as we shall see in Chapter 4). Paradoxically, this suggests that the map is ephemeral (to be used and discarded) in relation to the working out of narrated events but such a map is often more respected (retaining a fixed form and location over time) than one at the front of the book. In contrast, the map at the back of the book is somewhat perilously positioned. As already noted, Genette cites George Perec’s book, Life: A User’s Manual, in which the plan of a Paris apartment and its different inhabitants over time is given as part of a mock-academic appendix after the narrative. The danger (or deliberate provocation) of placing the map at the back is obvious: readers may not know it is there until they finish reading, in which case the map cannot have an active function during the reading process. To mitigate this, authors who place a map at the back, but want to ensure that the reader will be aware of it, are likely to emphasise its materiality, drawing attention to the map as a physical object by presenting it as a fold-out map or in a pocket. The disadvantage of placing the map after the narrative is thus compensated for by the advantage of physical juxtaposition of text and image whilst reading. The way that such maps physically relate to the book – being partially stuck onto the inside back cover or entirely removable – also emphasises the capacity of the fictional map to exist apart from and beyond the text to which it corresponds (in opposition to Genette’s affirmation that ‘The paratext is only an assistant, only an accessory of the text’ (410)). A second major aspect of the fictional map when considered as part of the paratext concerns its vulnerability over time in comparison with the text. Genette denotes an ‘original paratext’ (5) which is that of the first

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

40

From Literary Geography to Critical Literary Mapping

published edition. This moment in time is of great importance for the fictional map because it marks its first appearance in the world at a point when it has the greatest degree of authorial intention invested in it and its fully integrated (visual/verbal) meaning is most able to be interpreted. However, few modern editions respect the integrity of the first edition map over time, with the result that subtle meanings held within that map form, or in the dynamic between original map and text, are lost. Genette notes that ‘If, then, a paratextual element may appear at any time, it may also disappear, definitively or not, by authorial decision or outside intervention or by virtue of the eroding effect of time’ (6). This suggests that paratextual material is more vulnerable to the ravages of time than the text to which it is connected, and this is certainly true for the literary map. A key factor here is the degree of authorial intention that is associated with the map and respect (or a lack of respect) for that intention. Modern editions rarely, if ever, respect the original (first publication) form of the map and some omit the map altogether.47 One major reason for this is the distinction between the printing of hardback and paperback editions of a text. The move into paperback for large, cheaper print runs also involves a reduction of page size which has significant consequences for the quality and size of the map and thus its usefulness. The map is thus far more vulnerable to change, reduction and even removal over time than is the text to which it corresponds. The temporality of the map not only relates to duration (its ability to last over historical time) but is also bound up more immediately with the reading process in relation to the sequentiality of the narrative. It may be helpful to distinguish three phases of use for the fictional map during the reading process: a phase of anticipation; of active experience and repeated return; of retrospect. If we assume a normative position for the fictional map at the front of the book, then it represents spatial relations that will strongly bear upon events within a sequential narrative that the reader has not yet encountered. Only once the fictional world within the book becomes meaningful can the map have an integrative meaning. Dismissal of the map as mere illustration stems from this. The reader with limited spatial interest glances at the map initially and then disregards it so that the map is never actually ‘used’ and thus remains in a state of passive illustration rather than being engaged with actively to enlarge spatial understanding (and internalised mapping) of the literary space for the reader. Here,

47

E.g. popular editions of The Swiss Family Robinson.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Material Juxtaposition: Map and Text

41

then, an assumption of the map’s paratextual status interferes with its function as a powerful form of communication capable of working with the text. In contrast, the spatial reader, alert to the potential the map offers for a more grounded and richer experience of the spatial dimensions of the text, will make use of the map at multiple points in multiple ways during reading, as well as potentially returning to it afterwards to review the entirety of the narrative. The ‘two-faced’ nature of the paratext and thus, by extension, its potential for double meaning also proves highly relevant for the literary map where an apparent primary use value is often undermined by how it actually functions (consciously or unconsciously for the characters; intentionally or unintentionally for the author) as well as whom it addresses within or outside the text. When understood as a speech-act, the paratext is a problematic utterance, ‘situated as it is between the first-order illocutionary domain of the public world and that of the second-order speechacts of fiction’.48 As Genette points out, certain paratextual features (such as dedications) are explicitly performative, easily directing their message outwards (‘I dedicate this book to you’) but for other paratextual elements, the two ‘orders’ create difficulties and tensions (11). For the fictional map, this characteristic of two-facedness emerges strongly in relation to the third paratextual element defined by Genette: illocutionary force. The question of who has provided the map bears strongly upon its communicative nature and on how much it can be relied upon and trusted. (This is true for both sender–addressee relationships within and outside the text.) The map that faces towards the reader is most commonly presented by the narrator, who (ostensibly) gives it in order to further illuminate and authenticate the narrative he or she is relating (e.g. Jim Hawkins provides the map for Treasure Island). In such cases, the sender–addressee relationship ties map and narrative closely together. At the same time, the corresponding inward-looking map, represented within the narrative, is often introduced by means of a lengthy account of its own history and a full description of its materiality and contents. The fictional sender–addressee relationship is felt to be strongly arbitrary: the map does not know in whose hands it will eventually land. This also means that the map within the narrative has to prove itself worthy (or reveal itself not to be so) to the person who now owns it. Sometimes, the map attempts to reduce confusion by facing in only one direction; that is, it presents itself as explicitly

48

Genette, Paratexts, foreword by Richard Macksey, xix.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

42

From Literary Geography to Critical Literary Mapping

readerly. Where there is no corresponding use of the map within the text, this reduces the likelihood of confusion between the two different orders of communication (and the underlying referents) but can also create a visual/ verbal contradiction.49 It is clear that the materialism of Genette’s approach proves extremely useful for this particular conjunction, where much of the meaningproduction is centred upon direct juxtaposition of map and text within the covers of the book and where a material approach to the various tensions between the two forms proves highly productive in terms of enriching interpretation of the work. However, this is still to understand the application of the paratext only within the remorseless logic of its own terms rather than to consider its limitations. Perhaps a key issue here is that of subordination. Understanding the map’s primary function in relation to the text to be expository (as discussed above) naturally fits with Genette’s account of the paratext as always subordinate to the text, that is: dedicated to the service of something other than itself that constitutes its raison d’être. This something is the text . . . the paratextual element is always subordinate to ‘its’ text, and this functionality determines the essence of its appeal and its existence. (12)

However, to define the fictional map in such a way marginalises and limits its spatial power, which not only bears upon and shapes that text in generic and narrative ways but also illuminates it and potentially extends beyond it. Fictional maps are capable of acquiring meanings and values far beyond their primary ‘use’ value. For Genette, though, the paratext always operates at a second-order level, one step further away from reality than fictional representation, with the result that: the discourse on the paratext must never forget that it bears on a discourse that bears on a discourse, and that the meaning of its object depends on the object of this meaning, which is yet another meaning. (410)

On the one hand, this might be said to be particularly true of the fictional map, as a visual representation of a verbal representation with no basis in real-world reference. But, on the other, what we see with the most powerful and influential literary maps is that they have a capacity to release themselves, to work against the text in a number of ways, to become cultural icons, acquiring meanings beyond themselves and independently of the text. This

49

E.g. the presence of a map at the front of Richard Adams’s Watership Down (Bristol: Rex Collins Ltd, 1972).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

Material Juxtaposition: Map and Text

43

also, of course, applies to authoriality (a problematic issue in any case for the literary map which is sometimes, but not always, drawn directly by the author). When the meaning of the map ultimately works against the meaning of the text (as is quite frequently the case, for example, in detective fiction) it is unclear whether or not this is intentional on the part of the author. Instead, the authority of the map overrides the authority of the author. We might also note that, as an inevitable consequence of the assumption of subordination, paratextual elements are understood to have been written after, and in relation to, a pre-existing text. But in some of the most highly integrated examples of literary mapping, the map either precedes (The Hobbit), is used alongside (Hardy’s Map of Wessex) or actively brings into being (Treasure Island; The Lord of the Rings) the literary work to which it corresponds, immediately complicating this issue and undermining assumptions of referential priority. In Topographies, J. Hillis Miller comments that: [I]f the landscape is inside the novel, then it is determined by it and so cannot constitute its ground. The same thing may be said of the relation of any two members of the series: novel and map; real map and imaginary map; landscape and map. Each is both prior to the other and later than it . . .50

Just as critical cartography suggests that maps can invert assumed priority and work to bring constructions of the world into being, rather than functioning only as a secondary representation of the ‘real’, so for the literary work since there is no ‘real’ space for the fictional world either, the map corresponds to something which does not exist any more or less than it does. If neither is prior, and neither is underpinned by any referential reality, then the relationship between map and text is less about the hierarchy of the paratext and more about a complex totality that enables and brings about a deep experience of literary space and place. From one perspective, this leads to the inevitable paradox that ‘the effort of mapping is interrupted by an encounter with the unmappable’ as the reader engages with what Hillis Miller calls ‘a place that is everywhere and nowhere, a place you cannot get to from here’ (7). From another, or rather, at the self-same time, the paradox itself attests to the extraordinary nature of the spatial experience of literature – an active experience that enables a response to either map or world as if it were real, even though the reader knows that it is not.51 50 51

J. Hillis Miller, Topographies (Stanford, CA: Stanford University Press, 1995), 21. In fairness, Genette is not blind to the possibility of the paratext exceeding its intended role. He states: ‘like all relays, the paratext sometimes tends to go beyond its function and to turn itself into an impediment, from then on playing its own game to the detriment of its text’s game’ (Paratexts, 410).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.002

2

Historicising the Fictional Map

Then thought I with my self, who, that goeth on Pilgrimage, but would have one of these Maps about him, that he may look when he is at a stand, which is the way he must take? Bunyan, The Pilgrim’s Progress, part II

Where the previous chapter outlined the interdisciplinary underpinnings of this book by considering the interrelations of literature, geography and cartography and the material nature of the juxtaposition of map and text, this chapter adopts a more historical approach. It offers a brief history of the development of the fictional map from the twelfth to the nineteenth centuries, focusing on rare early examples that existed prior to the explosion of literary maps across genres in the mid-late nineteenth and early twentieth centuries (which forms the main focus of later chapters and of the study as a whole). It aligns the best-known fictional examples to major turning points in the history of cartography in order to uncover the extent to which the literary map partakes of the particular ‘map-consciousness’ of its time (or does not do so) and in order to contextualise how such maps were read. Fully historicising the fictional map reveals not only the extent to which such maps are subject to shifts in process and practice in the world but also enables a response to the fictional map itself, making it the focus of investigation. This allows us to see core characteristics and patterns of meaning that run right across genres and illuminates unlikely connections between those literary maps that might not otherwise be noticed. Such an approach has not, to the best of my knowledge, been previously attempted (although the limited number of early fictional maps means that these have been considered individually in a number of prior studies).1 1

44

The ‘hottest’ fictional maps are those for Utopia, Gulliver’s Travels, Robinson Crusoe and Treasure Island. These maps are discussed by most commentators. See particularly: Wyatt, Imaginary, Historical and Actual Maps; Ljungberg, Creative Dynamics; Ricardo Padrón, ‘Mapping Imaginary Worlds’, in Maps: Finding Our Place in the World, ed. James Akerman and Robert Karrow (Chicago, IL, and London: University of Chicago Press, 2007) 255–87; Ryan et al. Narrating Space / Spatialising Narrative.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Early Pictorial Maps

45

The chapter also considers the extent to which the early fictional map is strongly integrated with the fictional work to which it corresponds and, even in these earliest examples, the extent to which the map seeks to exert a knowing manipulation upon its readership (one of its later chief characteristics). In terms of the core dynamic between fictional map and text that forms the focus of later chapters, this chapter offers more of a survey, since its primary focus is the relationship between map form and style in comparison to real-world equivalents and other fictional maps over time.2 However, that survey proves essential since the overview it provides leads into the need for the comparative generic chapters which follow.

Early Pictorial Maps I want to begin with the mappaemundi – world maps of the twelfth to fifteenth centuries that represent the major cartographic achievement of their time.3 As the first example to be considered here they are highly unusual since a mappamundi is both a real-world map and one that corresponds to a text (the Bible). As such it functions more as a precursor than as a true fictional map, but it is certainly worthy of consideration, and draws attention to some distinctive characteristics of its later descendants. The making of such maps strongly asserts the dominance of the pictorial. Woodward makes the point that ‘Since the vast majority of these maps were produced for manuscript books, the techniques involved are indistinguishable from those used in manuscript illumination.’4 At this point the map-maker was essentially an artist: Mappaemundi were regarded as paintings in the early Middle Ages. Since their makers were map painters rather than cartographers in the modern sense of the

2

3

4

Both real-world and fictional examples are also predominantly British, revealing that Britain has often been a highly innovative nation in terms of cartographic developments, which may in turn indirectly bear upon literary maps. In English Maps: A History (Toronto: University of Toronto Press, 1969), Catherine DelanoSmith and Roger J. P. Kain state that ‘Some two dozen English mappaemundi are known . . . of which fewer than half are extant’ (38). Around 1,100 mappaemundi survive in the world but many of these are small and consist largely of T and O diagrams, ‘which present the world as ‘a “T” within a circle containing three continents, Asia, Europe and Africa, surrounded by water. The land masses are divided by three waterways which make up the “T”’ (see Jerry Brotton, A History of the World in Twelve Maps (London: Penguin, 2013), 96. David Woodward, ‘Medieval Mappaemundi’, in History of Cartography: Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean, vol. 1, ed. J. B. Harley and David Woodward (Chicago, IL: University of Chicago Press, 1987), 324.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

46

Historicising the Fictional Map

word, the methods, tools and materials used for these maps were those of the medieval artist in general. (Woodward, ‘Mappaemundi’, 324)

The fact that the content of the map is not primarily centred upon accurate representation of the real world itself tells us a lot about the mapconsciousness of the period and the way in which maps were read. Indeed, it is important to bear in mind that: There is no word for map in any ancient European language. In some modern European languages . . . the word used derives from late Latin mappa, a cloth, by way of mappa mundi, a cloth painted with a representation of the world. In most others it derives from late Latin carta, which meant any sort of formal document, resulting in ambiguities that persist to this day.5

The lack of terminology for such an object brings home to us the key point that most people in this period had no need of a map. Jeremy Harwood reminds us that ‘the majority of medieval people were unlikely to have travelled more than 15 miles from their birthplaces during the whole of their lifetimes’.6 Even where a map would be a far more efficient means of representing information, it was not adopted: Harwood continues, ‘If they did venture outside their own localities, medieval travellers seem to have used descriptions and gazetteers rather than maps’ (51). People simply did not think spatially as we do today: most would never have seen a map and, even if they did, they would not know how to read it. Thus, what we retrospectively call ‘maps’ are best understood as vast frameworks of spatial information to be gazed upon as rare objects rather than used: [T]he vast majority of medieval world maps are scarcely maps at all. They are diagrams – diagrams of the world – and are best understood as an open framework where all kinds of information might be placed in the relevant spatial position, not unlike a chronicle or narrative in which information would be arranged chronologically.7

Such a map stands as a graphic representation which privileges relative over absolute spatial meaning. It is interested in the relations between places or objects, rather than in fixing accurate points to actual sites on the earth’s surface or creating a uniform scale. All of these factors strongly determine the pictorial mode which dominates for the mappaemundi and 5

6

7

P. D. A. Harvey, The History of Topographical Maps: Symbols, Pictures and Surveys (London: Thames & Hudson, 1980), 10. Jeremy Harwood, To the Ends of The Earth: 100 Maps that Changed the World (London: Marshall Editions, 2006), 51. P. D. A. Harvey, Medieval Maps (London: British Library, 1991), 19.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Early Pictorial Maps

47

which, in turn, anticipates the dominant pictorial mode for the later fictional map, seen most clearly in well-known maps for children’s books in the early twentieth century (e.g. E. H. Shepard’s endpaper maps for Winnie-the-Pooh and The Wind in the Willows).8 The mappaemundi represent real-world geography in a spiritual context that affirms God’s presence. Their task is to ‘fit the world and what they thought they knew of it into the prevailing philosophical and religious viewpoints of the time’.9 Spatial meaning is thus distorted by the spiritual in a way that plays actual locations in the world off against the biblical and Christian valuing of such places. Such maps are framed by biblical sites not even locatable on earth – most notably the terrestrial paradise of the Garden of Eden. Delano-Smith and Kain tell us: The beginning of time, signified by some aspect of the Earthly Paradise . . . is always featured and so too, usually, is the end of time, signified by Christ Enthroned (at the top of the map), at the Last Judgement. (English Maps, 37)

This is true for the Hereford Map, a famous British example and now the largest surviving mappamundi in the world (see Fig. 2.1).10 The outline map of the world is held within a narrative frame that incorporates the start and end of time, whilst geography is distorted to ensure that Jerusalem is at its centre and other elements of biblical narrative are illustrated upon it. The historical and cultural remoteness of this form is signalled above all by its interest in time over place and not chronological time either: In a mappamundi, space and time are equal components. That is to say, the world that the mappa-mundi portrays is not synchronous but one in which events take place according to spatial rather than chronological order. (37)

Again, the medieval form anticipates the distinctive nature of the much later fictional map. Literary maps are essentially chronotopic – strongly concerned with time as well as space – in ways that create problems. Exactly the same problem (and a similar solution) is illustrated by the medieval map: [H]ow to show the sequence of events when everything appears simultaneously on the map? The problem was usually addressed (if not solved) by blocks of text,

8

9 10

A. A. Milne, Winnie-the-Pooh (London: Methuen & Co., 1926); Kenneth Grahame, The Wind in the Willows [1908] (London: Methuen & Co., 1931). Harwood, To the Ends of the Earth, 32. The largest mappamundi was the Ebstorf map in Germany, but this was destroyed in a bombing raid on Dresden in the Second World War.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

48

Historicising the Fictional Map

Fig. 2.1. Hereford Mappamundi (For the colour version, please refer to the plate section.)

either on the map itself or surrounding it, explaining how the various features were to be interpreted.11

This central, shared characteristic means that such maps often function in a juxtapositional way that directly anticipates the dynamic between fictional map and text that forms the focus of this study. Although the most famous mappaemundi are large wall maps, most surviving maps are much smaller: The vast majority of the maps that survive were produced as ipso facto book illustrations. In the late Middle Ages . . . there was a tendency to place maps on the 11

Evelyn Edson, Mapping Time and Space: How Medieval Mapmakers Viewed Their World (London: British Library, 1997), 165.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping Utopia; Mapping Dystopia

49

first or second page of a codex, which may reflect the growing importance of maps in giving the reader an overview of the text.12

In terms of a material juxtaposition of map and text then, these smaller mappaemundi function exactly like a literary map. So we can see that, despite their historical remoteness, many core characteristics of the later literary map chime strongly with this earliest of Western forms: a lack of interest in accuracy; a strongly pictorial mode; the importance of time as well as space; relational meanings and networks of meaning; an expository relationship to a text; the need to incorporate text and narrative with image; decorative, visual and cultural power.

Mapping Utopia; Mapping Dystopia In the medieval period, map-making was an extension of manuscript copying involving artistic skill and undertaken on the same materials and with the same tools. However, the invention of moveable type (as well as the printing press) in the mid-fifteenth century meant that monks lost control over map production. With the shift from hand-drawn, hand-painted maps to printed maps, challenges for the printers began to emerge, as Woodward makes clear: Map printing is a specialized activity with requirements that are often different from those for the printing of books. In the incunable period of map printing (technically through the end of the year 1500), the most important of these were the ease of making corrections, the ability of the medium to hold fine detail . . . and the versatility in combining lettering and linework.13

Maps were commonly printed using relief printing and a woodcut technique which did not require a press. Blocks were able to be printed alongside type, which was an advantage, but the difficulty of carving lines and letters in wood meant that the appearance of the images was relatively crude and the inset nature of the lettering sometimes apparent. The move away from the pictorial and geometric nature of medieval maps drew attention to another core element of mapping, affecting all kinds of map: that of perspective, the angle from which we look down upon the represented surface. Three main perspectives can be adopted: horizontal, oblique (‘bird’s-eye view’) and vertical (plan view) with a gradual move from the first to the last over time. Early maps, from primitive through to 12 13

Woodward, ‘Mappaemundi’, 286. Woodward, ‘Map Engraving’, 592.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

50

Historicising the Fictional Map

medieval times, tend to use symbols or pictogram representation to present a small version of the object as if seen from ground level (as with animals and cities on the Hereford Map). However, by the fifteenth and sixteenth centuries the bird’s-eye view predominates, particularly in Italy, the centre for map-making at that time. The map provides a plan view for everything seen directly from above, and an increasingly oblique angle for areas further away. The wealth of the Italian city states in the late medieval period led to major advances in cartography: the production of extraordinarily detailed city maps at home; accurate mapping of the inland sea of the Mediterranean with the beautifully detailed coastlines found on portolan charts; and maps of exploration into unknown realms by sailor explorers. The number of maps able to be produced also increased significantly: ‘By 1500 there were approximately 60,000 individual printed maps in circulation in Europe. By 1600 this number had risen to a staggering 1.3 million’.14 The earliest example of a literary work to contain an explicit fictional map corresponding to the text is Sir Thomas More’s Utopia which partakes of the new form of perspective offered by Italian city states, presenting Utopia by means of a low oblique view of the island and its capital. The first edition map of Utopia of 1516 was made by an unknown artist and published in Louvain on the reverse of the title page in the form of a small and simple woodcut entitled ‘Utopiae Insulae Figura’ (see Fig. 2.2). Although the map is pleasing and relatively ornate, the limitations of woodcut can be seen in the blocked and inserted Gothic lettering and the difficulty of creating fluid lines in the wood to represent the sea. This first map was then reproduced for the second Basel edition in 1518 by Ambrosius Holbein (brother of Hans Holbein the younger).15 As an article in the British Dental Journal points out, adaptations made to the image of the map – particularly the relocation of the ship, bottom right, allow the ship and island together to be interpreted as a skull (the ship’s side representing teeth) and thus turn it into a memento mori (see Fig. 2.3).16 As such it

14 15

16

Brotton, History of the World in Twelve Maps, 158. In 1518 the last word of the title was changed from ‘Figura’ (figure) to ‘Tabula’ (tablet and, by extension, map, painting or document drawn on a tablet). See M. Bishop, ‘Ambrosius Holbein’s Memento Mori for Sir Thomas More’s Utopia. The Meanings of a Masterpiece of Early Sixteenth Century Graphic Art’, British Dental Journal 199 (2005): 107–12; 108. As Bishop explains, it is Holbein’s later extraordinary use of the anamorphic skull for his painting of Henry VIII in The Ambassadors (1553) that retrospectively allows the possibility that he used the same idea in the woodcut map for Utopia (107). He also notes, ‘More himself made a similar pun on his own name when dealing with a man who was indebted to him’ (109).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping Utopia; Mapping Dystopia

51

Fig. 2.2. First edition map of Utopia (1516)

functions as a visual pun on More’s name, very much in keeping with the many levels of deliberate play present in both book and map, and also perhaps pointing to a darker side below the surface of this perfect communist state. Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

52

Historicising the Fictional Map

Fig. 2.3. Fourth edition map of Utopia (1518)

By its nature, a Utopian fiction contains certain characteristics that are naturally spatial and geographical, suggesting the need for a map. It is commonly presented as a traveller’s tale retold by an outsider after returning from an alien environment. Within the text, the narrator, Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping Utopia; Mapping Dystopia

53

Raphael Hythloday, opens the second part of the book with a detailed geographic description: The island of the Utopians is two hundred miles across in the middle part, where it is widest, and is nowhere much narrower than this except toward the two ends, where it gradually tapers. These ends, drawn toward one another as if in a fivehundred-mile circle, make the island crescent-shaped, like a new moon. Between the horns of the crescent, which are about eleven miles apart, the sea enters and spreads into a broad bay. Being sheltered from the wind by the surrounding land, the bay is never rough, but quiet and smooth instead, like a big lake. Thus, nearly the whole inner coast is one great harbor, across which ships pass in every direction, to the great advantage of the people. What with shallows on one side and rocks on the other, the entrance into the bay is perilous. Near the middle of the channel, there is one rock that rises above the water, and so presents no danger in itself; on top of it a tower has been built, and there a garrison is kept. Since the other rocks lie underwater, they are very dangerous to navigation. The channels are known only to the Utopians, so hardly any strangers enter the bay without one of their pilots.17

This is not a cartographic period overly concerned with accuracy or proportionate scale representation, so the map-image at the front could be said to loosely correspond to this description.18 At the start of book II, Raphael provides a succinct description of the rigid urban-planning design that determines the occupation of space: There are fifty-four cities on the island, all spacious and magnificent . . . So far as the location permits, all of them are built on the same plan and have the same appearance. The nearest are twenty-four miles apart, and the farthest are not so remote that a person cannot go on foot from one to the other in a day. (39–40)

In the case of Utopia, the strict geometry and geography of the island and the layout of cities extends from its core principle: the importance of holding everything in common. For this reason, ‘If you know one of their cities, you know them all, for they’re exactly alike, except where geography itself makes a difference’ (41). This highly regularised geometric ordering of space is partly represented on the map with the capital city

17

18

Thomas More, Utopia, ed. with a revised translation George M. Logan (New York and London: W.W. Norton, 2011), 38–9. In fact, as Brian Goodey makes clear in his literalistic attempt to reproduce an accurate map of Utopia, the geography More describes is impossible, not least because ‘A circle with a circumference of five hundred miles cannot contain a diameter of two hundred miles’ (21). See Brian Goodey, ‘Mapping “Utopia”: A Comment on the Geography of Sir Thomas More’, Geographical Review 60.1 (1970): 15–30.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

54

Historicising the Fictional Map

(Amaurotum) at the centre, divided by the river Anydri and other smaller towns spaced out symmetrically in a stylised way. The map is presented at the start of the work along with a sample of Utopian poetry and the Utopian alphabet (added by More’s correspondent, Peter Giles) and then followed by a number of letters, varying in different editions of Utopia, but designed, again, to reassure the reader about the worth of the work and also its veracity.19 The publication of such letters of recommendation from wealthy patrons at the start of a work was common in the early sixteenth century, but More also uses this space expressly to make claims for the ‘truth’ of his account and thus, crucially, to blur the boundaries of fact and fiction in his satiric account of an imaginary place.20 In his own prefatory letter to Giles (a real friend of More’s) he repeatedly does this, describing the writing of the book as mere transcription: ‘there was nothing for me to do but simply write down what I had heard’ (6) and emphasising his ‘particular pains to avoid having anything false in the book’ (7). The map thus forms part of a documentary apparatus that all contributes strongly to the illusion that Utopia is a real place on the edge of the known world. As this suggests, the entire work is deeply contradictory: ultimately More is not advocating an ideal communist society but making the point that such a place could never exist on earth. There is also a further ongoing joke about the difficulty of pinpointing the exact location of the island. The initial, prefatory letter by More, raises a problem that he hopes his friend, Peter Giles, will be able to help him with: [A]nother problem that has cropped up – whether through my fault, or yours, or Raphael’s, I’m not sure. For it didn’t occur to us to ask, nor to him to say, in what area of the New World Utopia is to be found. I wouldn’t have missed hearing about this for a sizable sum of money, for I’m quite ashamed not to know even the name of the ocean where this island lies about which I’ve written so much. Besides there are various people here, and one in particular, a devout man and a professor of theology, who very much wants to go to Utopia. (7)

19

20

As such it is also an early example of the fictional map as ‘doceme’: part of a framework of gathered material relating to the text. Genette describes such ‘documentary texts’ in the ‘Conclusion’ to Paratexts (404). See also Stefan Ekman’s definition (referencing Genette and drawing on Niels Windfeld Lund) in Here Be Dragons: Exploring Fantasy Maps and Settings (Middletown, CT: Wesleyan University Press, 2013) where he argues that ‘fantasy maps can be fruitfully interpreted as both paratexts and docemes’ (22). See also Ricardo Padrón’s reading of Utopia in ‘Mapping Imaginary Worlds’. He interprets the map as part of the larger play of the work: ‘meant to blur the line between the fictive and the real’ (269).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping Utopia; Mapping Dystopia

55

The map portrays an absolutely enclosed, bounded space that abides by its own rules but is remote and inaccessible. It is not located in relation to a known world and will not enable you to travel to Utopia – which is surely what the reader who seeks to share in the perfect communal life (the supposed ‘professor’ above) needs the map to do. The map is thus contextualised stylistically in terms of the spatial and cartographic knowledge of its time period, but it is also directly aligned to the core aims and purpose of the fictional narrative: to present a parallel world that critiques the writer’s own and one that represents an ideal commonwealth spatially as well as philosophically (but ultimately asserts that this is not a place you could ever travel to or visit). Despite being unlocatable on a map of the world, however, More tied his traveller’s discovery of Utopia directly into contemporary accounts of sea exploration, using the voyages of Italian explorer Amerigo Vespucci as his starting point. Vespucci had travelled to South America on four voyages between 1497 and 1502, getting as far as Cape Frio, seventy miles east of Rio de Janeiro. Accounts of his travels, the first entitled Mundus Novus (New World), were written as letters home (attributed to Vespucci himself, though the originals were lost). The four voyages had been gathered together in 1507, only nine years prior to the publication of Utopia. They were extremely popular and made the discovery of the New World widely known although (perhaps crucially) they did not contain any maps.21 In Utopia, More integrates Raphael’s fictional narrative with the end of Vespucci’s account of his fourth and final voyage, as if his companions went on a little further without him. Raphael is described as follows: He was Vespucci’s constant companion on the last three of his four voyages, accounts of which are now common reading everywhere; but on the last voyage, he did not return home with the commander. After much persuasion and expostulation he got Amerigo’s permission to be one of the twenty-four men who were left in a fort at the farthest point of the last voyage . . . After Vespucci’s departure, he traveled through many countries with five companions from the garrison. (11)

The positioning of Utopia just beyond the edge of the known world cleverly enfolds imaginary place into only-just-discovered real-world geography. Thus, More uses the limits of cartographic and geographic knowledge of his time – with its extremely fuzzy outer edges – to his advantage. 21

In 1507 Vespucci’s accounts were collected together by German cartographer Martin Waldseemüller in Cosmographiae Introductio, a book published alongside the wall map Universalis Cosmographia. This was the first world map to include the newly discovered lands and name them after Vespucci in the feminised form of ‘America’.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

56

Historicising the Fictional Map

This very first fictional map for Utopia also draws attention to the recurring presence of maps in certain genres. Where a fictional map appears very early on within a genre, or in the first work in a particular form (as with this example), this authorises the ongoing presence of a map as an intrinsic element of that genre.22 We see this not long after in the first dystopia: Joseph Hall’s Mundus Alter et Idem (An Old World and a New) first published anonymously in Latin in 1600.23 A work that could well be seen as bridging the gap between Utopia and Gulliver’s Travels – thus enabling us to see a clear line of descent in a tradition of fictional mapping – Hall’s narrative concerns a voyage in the ship Fantasia to the four regions of Crapulia (a land of excess), Viraginia (land of the viragoes), Moronia (fools-land) and Lavernia (land of thieves and cheats), whilst close by is the land of Terra Sancta that remains pointedly empty (see Fig. 2.4). Like More (and in anticipation of Swift), Hall uses the unknown limits of the world to create his own fictional space below the newly discovered America, in the imagined continent of Terra Australis (not actually discovered until 1770 by Captain Cook) off Terra del Fuego. Unlike More he provides a world map locating his New World and gives an overview map of the entire region and then four more detailed maps of each of the four regions (an early example of multiple mapping).24 Hall’s ‘nesting’ of the imaginative space alongside or within recent mappings of the real (though remote and alien) world initiates a major mode for literary mapping – in which larger well-known place names (countries; major cities) are retained to create a realistic frame within which the more localised fictional world is fitted. In spite of its historical distance, the map for Utopia contains a surprising number of central characteristics that emerge in much later fictional maps. It appears to be provided in order to authenticate and ground the narrative, but has the capacity to mislead and withholds as much as it gives; it is fully integrated into the knowing cleverness of the text to which it corresponds; it signals a strongly spatial text and initiates the use of a map for the genre it establishes; it plays with ideas of authenticity. In relation to the idea of a lineage for fictional maps, a claim could be made that, until the 22

23

24

Subsequent utopian and dystopian fictions also frequently contain maps. E.g. J. V. Andreae Christianopolis (1619); Francis Bacon’s New Atlantis (1626); Gulliver’s Travels (1726); Erewhon (1884). When we come to the twentieth century, however, maps seem less de rigueur: Brave New World (1932); 1984 (1949); Lord of the Flies (1954). Joseph Hall, Mundus Alter et Idem (An Old World and a New), ed. H. J. Andersen (London: George Bell & Sons, 1908). Also unlike More, he used his satiric antipodean space to attack Roman Catholic orthodoxy.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping the New World

57

Fig. 2.4. World map from Mundus Alter et Idem

nineteenth century, there is a sense in which almost all fictional maps look back to it. Thus, it exceeds its own genre to initiate a discrete form in its own right: the mapped fictional world as an island, an ‘othered’ space that is then used satirically, or in another way, to create a rift between one way of living and another and to enable reflection back onto the world or the life lived in it. Later, apparently very different, literary works containing maps, such as Gulliver’s Travels and Robinson Crusoe, would still more or less fall under this category.

Mapping the New World It could reasonably be argued, then, that the model for fictional mapping set by Utopia (itself a possible descendant of the medieval depiction of Terrestrial Paradise as an island off the edge of the world) extends its influence directly into the eighteenth century with the imaginary geographies of Defoe and Swift. In all of these examples of early fictional mapping, a vital element is the fluid nature of the interplay between fact and fiction in

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

58

Historicising the Fictional Map

terms of geography and the extension of what is known into what can only be imagined. In this way, the larger geographic and cartographic worldview bears directly upon the development of the fictional map and of literary place and space. From the sixteenth century onwards, cartographic developments were centred on the problem of map projection. Whilst local and regional maps did not have to worry about distortion caused by the fact that the earth is round, world maps had to address this issue by creating mathematical transformations that enabled a spherical world to be represented on a flat plane by means of a geometrical figure such as a cone or cylinder. This problem was solved, famously, by Flemish map-maker Gerard Mercator in 1569 when he produced what ‘may be the most influential map in the history of geography’ (Brotton, History of the World in Twelve Maps, 247).25 His Mercator projection was the first to achieve conformality – ‘the maintenance of accurate angular relations at any point on the map’ (252), invaluable to sea voyagers since it allowed them to plot a straight line on a chart and then follow it using a compass. With the onset of increasingly confident world exploration came the emergence of a closely linked generic form: the voyage narrative. The first widely read compendium of travel narratives was that of Richard Hakluyt, The Principal Navigations, Voyages, and Discoveries of the English Nation (1589–1600), which, with its accounts of the travels of Raleigh, Drake, Gilbert, Cavendish and others, set out to celebrate and anticipate national achievements and the acquisition of foreign lands. This was followed in 1625 by Samuel Purchas’s Haklytus Posthumus or Purchas his Pilgrims which drew heavily upon Hakluyt (Purchas had worked for Hakluyt as his assistant) and organised voyages into different categories for easy reading, whilst also having a far more religious attitude towards the voyage as an allegory of the soul’s journey.26 However, the key work of voyage literature for our purposes here – the most popular voyage narrative of the eighteenth century – was William Dampier’s New Voyage Round the World (1697). Dampier was a buccaneer who spent 1679–91 circumnavigating the globe on a series of adventures and took considerable time and care over writing them up. His personality and style was perfectly suited to the form: ‘Dampier

25

26

Brotton has an entire chapter on Mercator (218–59) and gives an illuminating explanation of Mercator’s projection by means of imagining a balloon inside a cylinder (251–2). See C. R. Steele, ‘From Hakluyt to Purchas’, The Hakluyt Handbook, ed. D. B. Quinn, 2 vols. (London: Hakluyt Society, 1974). See also Wyatt, Imaginary, Historical and Actual Maps, 22–36.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping the New World

59

was a natural Baconian scientist, fascinated by everything he saw’, resulting in a text that is ‘both a narrative of happenings and an extremely detailed account of weather, scenery, the people . . . birds, beasts and fish’.27 Dampier’s attentiveness to the needs of the ‘armchair traveller’ rather than the seaman was undoubtedly a key element in his success but this also indicates the extent to which the form lent itself naturally to fictionalisation. As Phillip Edwards reminds us in The Story of the Voyage: [T]he reading public could not get enough in the way of accounts of all the maritime activity involved in extending Britain’s knowledge of the globe and her control of territories old and new. (They literally could not get enough, so fiction writers supplied them with more.) (2)

Edwards also makes the perceptive point that (with the exception of Dampier), ‘Sailors did not know how to make the most of their own stories’ (8). Numerous voyage narratives were published, but the reasons for doing so were not entirely clear, which allowed for a large degree of slippage between formal record and entertaining read: ‘Schizophrenic dithering between the demands of science and the claims of the general reader was never resolved’ (7). However, this indeterminacy proved very useful for fictional writers and fictional mapping: ‘The uncertainty of the voyagers about their aim and their public made the gap which Defoe could enter’ (8). Both Defoe and Swift capitalised on the popularity of the form to create their travel narratives by imaginative extension out of the preexisting real-world genre. We could also compare this to Purchas’s allegorisation of the voyage – giving the act of being ‘cast away’ secondary layers of meaning (spiritual in Defoe’s case or satirical in Swift’s). Where Defoe draws directly upon actual accounts by Dampier and his descendants, Swift makes a direct connection in ‘A Letter From Capt. Gulliver to his Cousin Sympson’ prefixed to the 1735 edition of Gulliver’s Travels, in which Gulliver complains at being rushed into publishing his papers and so not having time to ‘to hire some young Gentlemen of either University to put them in Order, and correct the Style, as my Cousin Dampier did by my Advice, in his Book called, A Voyage round the World’’.28. Cartographically speaking, Defoe and Swift are both directly connected through one man: the famous eighteenth-century British map-maker,

27

28

Philip Edwards, The Story of the Voyage: Sea Narratives in Eighteenth Century England (Cambridge: Cambridge University Press, 2004), 20. Jonathan Swift, Gulliver’s Travels, ed. Albert J. Rivero (New York and London: W.W. Norton & Co. Ltd., 2002), 253.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

60

Historicising the Fictional Map

Herman Moll. In an early, but informative, article on ‘The Maps in Gulliver’s Travels’ Fredrick Bracher states: Moll’s maps probably gave the average Englishman of the day his picture of the world, and they were so popular that imitation and copying became a serious problem . . . It is ironical that Moll’s name should have survived in literary history largely because of a publication in which his maps were copied, Gulliver’s Travels.29

Moll is also one of the few contemporary figures mentioned by name by Gulliver in the final part of the book: I arrived in seven Hours to the South-East Point of New-Holland. This confirmed me in the opinion I have long entertained, that the Maps and Charts place this Country at least three Degrees more to the East than it really is; which Thought I communicated many Years ago to my worthy Friend, Mr. Herman Moll, and gave him my Reasons for it, although he hath rather chosen to follow other Authors.30

Gulliver’s authority as imaginary traveller here purports to override accurate real-world information (acquired from Dampier and others) that Moll drew on in creating his up-to-date maps using Mercator’s projection. Moll’s fortunes were closely tied to Dampier’s since he produced the maps that accompanied the volume, particularly that on the title page ‘A Map of the World Shewing the Course of Mr Dampier’s Voyage Round it. From 1679 to 1691’.31 Moll was also using the new copperplate engraving style favoured by Mercator and made popular by him.32 The maps for such travel narratives were an intrinsic part of their popularity, so it is unsurprising that when Defoe added a double hemisphere ‘Map of the World’ to the fourth edition of Robinson Crusoe in August 1719 he chose Moll as map-maker.33 The two maps (in Dampier and Defoe) are extremely similar, but in Robinson Crusoe the map is not strongly integrated with the text so it would seem that its addition is likely to be more a result of larger cultural and economic forces than anything 29

30

31

32

33

Frederick Bracher, ‘The Maps in Gulliver’s Travels’, Huntington Library Quarterly 6.1 (1944): 59–74; 60. Part IV, ch. XI, 239. As Bracher points out, this accolade should not be taken at face value: ‘he is described, no doubt with irony, as “my worthy friend Herman Moll”’ (60). Moll’s name was clearly marked at the bottom of the map (‘H. Moll Fecit’) and gave the collection additional caché. See Brotton, History of the World in Twelve Maps, 225. For a full account see Coolie Verner, ‘Copperplate Printing’, in Five Centuries of Map Printing, ed. David Woodward (Chicago, IL: University of Chicago Press, 1975), 51–75. In A Tour Through the Whole Island of Great Britain, Defoe also used details from Moll’s A Set of Fifty New and Correct Maps of England and Wales (1724) reproduced opposite letters I–IV and VI–XIII.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping the New World

61

Fig. 2.5. ‘Illustration of the Bashee Islands’ from A New Voyage Round the World

else: cartographic association of Robinson Crusoe and Dampier could only help to sell more books.34 All of the coastlines of newly discovered lands in Gulliver’s Travels are copied from Moll’s 1719 world map (as Bracher convincingly argues (62)). Moll’s maps had a very distinctive clean style with a flowing italic hand and block capitals for place names which was directly imitated in the maps for Gulliver’s Travels. It is worth pausing to compare one of Moll’s detailed maps for Dampier’s New Voyage Round the World with the map of Laputa (the flying island) and its corresponding land below (Figs. 2.5 and 2.6). When we do this, we can see how stylistically similar the resemblance is (although the map-maker for Gulliver does not use the free hand writing style). The map replicates Moll’s shading around the edges of the land, representation of hills on the islands and his printed font style for names. I chose

34

Ljungberg in Creative Dynamics discusses this map as an example of how ‘Crusoe, as the prototypical homo economicus, puts his mark on the world, leaving traces of his travels and appropriating it’ (64).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

62

Historicising the Fictional Map

Fig. 2.6. ‘Map of Laputa’ from Travels into Several Remote Nations of the World . . . by Lemuel Gulliver

the map for Laputa as the one that creates the most difficulty for the cartographer (since he has to find a way to represent an island in the sky). He does this by adapting Moll’s habit of marking routes around the islands with dotted lines, but, here, these dotted routes trace Laputa’s movement up and down in the air, across the land below (although Laputa itself is oddly represented as if seen from above as well, when it should be viewed horizontally).35 For Laputa, the movement of the island across its base domain by manipulation of magnetic force used to repel, and thus push upwards, or attract and pull downward could also be seen to work as a kind of ironic counterpoint to the use of the magnetic force of the earth for a compass bearing. Where, in real-world cartography, magnetic force enables accurate self-location on the earth by aligning to magnetic north, in Laputa the magnetic force (working vertically rather than horizontally) works to propel a remote king along above his people. In relation to Gulliver’s Travels, a range of issues emerges around critical engagement with the maps, and these issues provide a good example of how literary-critical accounts of mapping change across the twentieth century. Initially, the maps proved a heated focus of debate in the 1940s and 50s with three main critics (Moore; Case; Bracher) offering detailed accounts of the problems in aligning the geography of the islands with those of remote coastlines of the day.36 This early discussion is entirely centred on accuracy in relation to real-world cartographic knowledge of the time. Major issues emerge as ‘problems’: the fact that, although Swift 35

36

See also Nicole E. Didicher, ‘Mapping the Distorted Worlds of Gulliver’s Travels’, Lumen 16 (1997): 179–96. Bracher, ‘The Maps in Gulliver’s Travels’, 59–74; John R. Moore, ‘The Geography of Gulliver’s Travels’, Journal of German and English Philology 40 (1941): 214–28; Arthur E. Case, Four Essays on Gulliver’s Travels (Gloucester, MA: Peter Smith, 1958).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping the New World

63

always has Gulliver provide a latitude and longitude for his position, there is no known base meridian so that the numbers are meaningless; the fact that Swift definitely did not oversee publication, which allowed errors to creep in on the part of those making the map (not only major geographic errors such as the placing of New Holland but also copying errors for place names).37 Finally, there are errors of scale, with the island of Brobdingnag in particular – which must be enormous if it is to hold giants – being represented as a small bubble off the coast of America.38 Matters are further complicated by Swift’s deliberate secrecy over the copying and submission of the manuscript of Gulliver’s Travels to the publisher (anonymously, in the night, with a letter from ‘Richard Sympson’ and with the manuscript written in another’s hand) and the strong likelihood that he therefore could not have been in a position to authorise the maps. Bracher does, however, make the important point that: There is some evidence, though of a negative kind, suggesting that Swift gave tacit approval to, if he was not responsible for, the inclusion of the maps. Most important, perhaps, is the fact that he did not strike them out of Faulkner’s edition of 1735 . . . though parts of Motte’s text were stricken out, the maps remain. (64)

As Bracher makes clear, even if Swift was not personally responsible for the addition of the maps to the volume, he chose to retain them later when he prepared a second edition himself. Thus, understanding of the role of the maps is complicated by indeterminate authorial intention: if the maps are not authorial can they still contribute to the satiric intent of the work in an integrated way? If we jump ahead forty years, we see a very different critical approach from Nicole Didicher. She offers a reading which suggests that the maps are a deliberate challenge to readerly assumptions: ‘the process of mapping and the ways that we think when we interpret maps are central to what Swift is saying in the Travels about the unreliability of supposedly reliable facts’ (181). As with More’s Utopia, then, the maps can be seen to present a double level of significance. At the first level, they seem to be simply of their time, partaking of the fashion, but at a second they are conveying a very different message to a more knowing audience. Didicher makes the point that ‘Moll, like Gulliver, insisted on his own accuracy and modernity, and

37

38

See Case, Four Essays (54–6), for cruxes in geography in relation to Lilliput; and Moore, ‘Geography of Gulliver’s Travels’, for errors in place names (229). Case states, ‘the dimensions of Brobdingnag are too large to be fitted into the map of the world as known to the eighteenth century’ (Four Essays, 57).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

64

Historicising the Fictional Map

was thus a ready target for satiric attack’ (186) so that ‘Moll’s empty posturing makes him a natural target’ (188). She points out that the choice to base the maps on Moll’s hubristic 1719 ‘New and Correct Map of the Whole World’ is also significant: ‘The Travels maps are in part traced from a source which also over-proclaims its own veracity, just as travel narratives and Gulliver do’ (190).39 Thus, Didicher concludes that: Altogether, it seems safe to say that the maps are not intended to give verisimilitude, but rather, while seeming to be trustworthy and stable, they are meant to undermine our ideas of what the world is like through puzzle and paradox. (194)

This study – which seeks to argue that all maps are inherently untrustworthy and that this element of deliberate deception of the reader is strongly present from the earliest example onwards – is clearly in agreement with the latter approach. Equally, since Swift was certainly well aware of Utopia, for which, as we have seen, the paratextual map is part of the knowing joke, it seems unlikely that he would not have attempted to incorporate any visual representations into the verbal satire. However, matters are undoubtedly complicated by the issue of intention. It seems to me that there is a fairly easy, though entirely speculative, solution: the possibility that Swift might have left (anonymous) notes or information about the maps for the publisher to follow. My reason for suggesting this is twofold. First, the problem of why an ordinary publisher or map-maker would so intelligently align the maps with Swift’s work and seek to give them a satiric nonsensical geography of their own is not easily explained.40 Second, when we consider the nature of some of the ‘errors’ concerned with the maps, they do seem to be playing geographically with concepts that are active in the text. So, for example, the land in which the humans are distortedly large is also the one on the map which is represented as far too small. Another question that has not previously been addressed, but seems to me to be highly relevant, is the existence of the detailed map of Laputa given above. Gulliver’s Travels was already standardised to the voyage narratives of the time in terms of four maps 39

40

Herman Moll, A New and Correct Map of the Whole World Shewing the Situation of Its Principal Parts . . . (London: H. Moll, T & J. Bowles, P. Overton and J. King, 1719). This map, charted on Mercator’s projection, depicts the routes of various sailors including WoodesRogers, the captain who discovered and rescued Alexander Selkirk (on whom Robinson Crusoe was modelled). Cf. Didicher: ‘More telling than Swift’s lack of objection to the plates, however, is the fact that they contain the sorts of puzzles and paradoxes beloved by Swift and unlikely to be invented by any hack engraver hired by Motte to make the book look more interesting’ (‘Mapping the Distorted Worlds’, 184).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Mapping the New World

65

(the norm) so why was this additional map detail added, especially when it presented major difficulties in terms of authentic representation? In fact, if anything, it directly undermines the very act of presenting maps in order to authenticate the imaginary worlds (since it presents an unrepresentable place in the air). Again, then, it seems highly unlikely to me that the publisher would choose to add such a map without some authorial direction or requirement, just as Swift’s extremely specific description in the text seems to lend itself to an authorial desire for visual representation. Either way, whether or not any satiric intention was there on the part of the mapcopier, when these maps are juxtaposed alongside the text they acquire a new and highly satiric meaning. Before concluding this section, I want to return briefly to Defoe. In addition to the world map added to the fourth edition of Robinson Crusoe in 1719 (mentioned above), in 1720 Defoe added a map of Crusoe’s island to his second additional volume: Serious Reflections During the Life and Surprising Adventures of Robinson Crusoe: With his Vision of the Angelick World. Written by Himself (see Fig. 2.7). This later volume responds defensively to the charge that the original volume was a work of fiction or romance, and attempts to resituate the original story as a kind of history out of which spiritual and philosophical truths can be drawn. So, for example, the first section is entitled ‘On Solitude’ and considers the importance and value of Christian fellowship. The map therefore functions as a series of vignettes relating to key scenes and sites from the earlier volume: the original shipwreck bottom centre; the establishing of a base in the centre of the island (with the parrot crying ‘Poor Robinson Crusoe’); the rescue of Friday and his father; the supplicants from the later ship and the means to return home. Ljungberg (after Blewett) tells us that the map form: represents a synoptic version of Crusoe’s verbal narrative in a manner reminiscent of the medieval Italian practice referred to as a ‘Sienese primitive’ which was often used in religious medieval art and cartography . . . It presented religious figures involved in central events compressed into one single picture, a device called simultaneous succession. (Creative Dynamics, 60)41

The publishing of the map in volume III is also significant. The image of the island is given as part of a slightly misleading attempt to connect the later

41

Ljungberg discusses Robinson Crusoe at some length in Creative Dynamics (51–66). See also David Blewett, The Illustration of Robinson Crusoe: 1719–1920 (London: Colin Smythe Ltd, 1995), 28.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

66

Historicising the Fictional Map

Fig. 2.7. ‘Map of Crusoe’s Island’ from Serious Reflections during the Life and Surprising Adventures of Robinson Crusoe

moral volume to the original story (from which it veers further and further away as it goes on). The appearance of the map in Serious Reflections means that it does not directly suffer from a tension between the simultaneity of visual representation and the sequentiality of narrative that often exists for island narratives (where the sequential unfolding of the narrative is centred upon a gradual enlargement of knowledge and possession over time which the simultaneity of the map discloses). The reader has already presumably read the Robinson Crusoe story so that this map forms part of a recollected return. This is felt most strongly in the presence of the ship bottom right. If the map were given at the start of the original narrative this would immediately resolve the crucial plot uncertainty as to whether Crusoe will be rescued or not. Perhaps this also explains why it has never been resituated in relation to the original story of volume I as we might otherwise expect.

Itinerary Maps: Paris, Ogilby, Bunyan Up to now, the chapter has considered the importance of an enlarging knowledge of global geography, and the mapping of new worlds for fictional map forms from the medieval period through to the Enlightenment. This section returns to more localised travel represented by the itinerary map, a much more everyday form, also with an ancient history, that prioritises distance between local places over all other kinds of

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Itinerary Maps: Paris, Ogilby, Bunyan

67

information. I want to consider this in relation to a post-authorial map, for Bunyan’s Pilgrim’s Progress (1678). The map focused on here is a fold-out strip map given at the front of a London edition of 1778 (presumably as a hundred-year celebration of the book’s first publication).42 The map was printed in copperplate, which remained the dominant mode of production from the sixteenth to the early nineteenth centuries (Verner, 64). As the simplest and most practical form of mapping, an itinerary map deliberately restricts itself to a primary function: that of following a particular route towards a predefined goal. In Roman and medieval times, travellers commonly followed written itineraries in the form of a list of places in the order that they occurred along a certain route with the distances between them. If they lost their way they could rely on local knowledge. This is an example, then, of verbal and written mapping that is only later translated into a graphic form. The simplicity of the itinerary structure allows for a direct correspondence between this kind of representation of an actual journey and early literary genres, reminding us of how integrated the literary narrative is with the spatial dimensions of the world it depicts and the way in which we travel through that world. The form of the itinerary map directly corresponds to a narrative for which there is a clearly defined endpoint or one which simply follows where the road leads, incorporating minor events along the way. It is applicable to pilgrimage – as the most likely reason for ordinary folk to venture away from home up until the seventeenth century. In the fourteenth century, the telling of Chaucer’s Canterbury Tales is structured in terms of a welltravelled route with clear stopping points along it. In the eighteenth century, the form of the picaresque novel is also that of a journey along a route, with regular incidents and digressions (e.g. Tom Jones; The Sentimental Journey). By the twenty-first century, the road is all that remains in dystopian visions of a future planet destroyed by man (e.g. Cormac McCarthy’s The Road). There may even be a sense in which the development of the novel itself, in its gradually enriching spatial and psychological sophistication, and expansion outwards from simple linearity, develops in parallel with the major advances of road development and increasing travel. Franco Moretti suggests as much in Atlas of the

42

The Pilgrim’s Progress . . . Complete in Two Parts. A New Edition. Embellished with a More Superb and Elegant Set of Copperplates than Was Ever Given with Any Former Edition (London: H. Trapp, 1778). The engraving was by Thomas Conder. This appears to be the earliest map of Pilgrim’s Progress. A colour version was made in 1780 and it was also turned into a jigsaw puzzle in 1790.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

68

Historicising the Fictional Map

European Novel when he comments on the connection between roads and the early novel form and gives a map of the routes used in Spanish picaresque novels.43 Two British map-makers were highly innovative in converting the early written itinerary into a graphic, visual experience. The first of these is the outstanding thirteenth-century medieval map-maker, Matthew Paris, and his itinerary map from London to Rome, presented in the three-volume chronicles Historia Anglorum (volume III: Chronica majora).44 This represented a remarkable cartographic leap, well ahead of its time and ‘probably unique in its day’ (Delano-Smith and Kain, 152).45 As Delano-Smith and Kain make clear, Paris would not have undertaken the physical journey himself and would have made the map by copying others (150–1). The map was divided into a series of strips, each corresponding to roughly a day’s travel, with two or three alternative paths represented, and with various flaps and fold-out sections added on arrival in the Holy Land (see Fig. 2.8). Such a map is strongly diagrammatic in nature and, of course, distorts the actual topographic experience (since the route is always presented as a more or less continuous straight line). In spite of the high level of detail, however, this map was emphatically not made to be used on the road. In his detailed analysis of Paris’s map, Daniel Connolly argues for it as an interactive, performative object with a ‘dynamic and participatory design’.46 He goes on: The Benedictine Brother who perused these pages understood this map primarily through its performative possibilities, as a dynamic setting, the operation of whose pages, texts, images, and appendages aided him in effecting an imagined pilgrimage. (598)

Connolly notes the paradox between a monk’s cloistered existence, locked away from the world, and the way in which ‘the cloister represented also the ideal realisation of an interior spiritual journey . . . It is in the

43

44

45 46

Moretti comments, ‘But such is precisely the secret of the modern novel (of ‘realism’, if you wish): modest episodes, with a limited narrative value’ (Atlas of the European Novel (London: Verso, 1999), 48). This is another key example of larger map-consciousness bearing directly upon the development of literary genre and form. Delano-Smith and Kain tell us: ‘The itinerary was placed, together with the maps of Britain and the Holy Land, at the beginning of each volume of, or relating to, the Chronica maiora’ (English Maps: A History , 274 n. 47). See also Suzanne Lewis, The Art of Matthew Paris in the ‘Chronica majora’ (Berkeley: University of California Press, 1987). Delano-Smith and Kain, English Maps: A History, 152. Daniel Connolly, ‘Imagined Pilgrimage in the Itinerary Maps of Matthew Paris’, Art Bulletin 81.4 (1999): 598–622; 598.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Itinerary Maps: Paris, Ogilby, Bunyan

69

Fig. 2.8. ‘Part of an itinerary from London to Jerusalem’ from Matthew Paris, Chronica majora (For the colour version, please refer to the plate section.)

cloister . . . that the monk comes to find the Heavenly Jerusalem’ (598). In such a way, this spiritual map again works analogously to a fictional map, at least in relation to its readership. The visualisation of geography is not given in order that one can follow a route to an actual place (the real city of Jerusalem) but in order to achieve the celestial city. Bunyan declares at the

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

70

Historicising the Fictional Map

start of Pilgrim’s Progress: ‘This Book will make a Travailer of thee’.47 Connolly argues that Paris’s itinerary enables just such an imaginative journey. After Matthew Paris, full development of the itinerary map had to wait for the explosion of road-books (printed forms of itineraries bound together) in the sixteenth century in response to increased numbers of travellers, and finally for the creation of the first road atlas in Europe by John Ogilby: Britannia (1675).48 This substantial work (300 pages, presenting 100 map sheets and 200 pages of text) was the first to use a uniform scale (1 inch : 1 mile, later adopted by the Ordnance Survey) and presented the seventy-three major roads out of London used as post roads in continuous parallel strips. It still combined visual and verbal forms since the maps were accompanied by full written itineraries of the route. Map sheets were divided up into strips (depicted as a series of scrolls), each sheet covering in total approximately seventy miles, probably corresponding to a day’s travel (see Fig. 2.9). The maps were made using advanced methods of surveying and triangulation covering around 7,500 miles of road. In her Norton edition of Pilgrim’s Progress, Cynthia Wall outlines a brief history of illustration of Bunyan’s popular work, through woodcuts and chapbooks to the strip map that is our focus here.49 Wall directly compares the strip map for Pilgrim’s Progress with Ogilby’s Britannia to make the point that Ogilby’s work was extremely popular and widely read. She argues, therefore, that his maps functioned as an important cultural text and context for the kind of visualised journey that Bunyan depicts, since ‘readers of Bunyan’s Pilgrims included the new travellers encouraged by Ogilby’s itineraries’ (432). Historical proximity also meant that the strip map for Pilgrim’s Progress, made only three years after Ogilby’s atlas, was cartographically right up to date (see Fig. 2.10). Turning to the Pilgrim’s Progress map, then, we see that, particularly when looking back to Paris’s much earlier map (a connection not made by Wall), the fictional map functions in a directly comparable way. The visualisation of the narrative, which is itself a journey, enables readers to

47

48

49

John Bunyan’s The Pilgrim’s Progress (1678), ed. Cynthia Wall (W.W. Norton & Company, 2009), 9. Delano-Smith and Kain note: ‘Although Matthew Paris’s name is nowhere mentioned, however, it is very likely that Ogilby drew his inspiration from Paris, for there is little doubt that Ogilby would have known about Paris’s itinerary to Rome’ (English Maps: A History, 170). For a good account of Ogilby’s maps see 168–72. John Bunyan’s The Pilgrim’s Progress, ed. Wall, 271–2, 426–32. Wall’s cover image is a coloured version of the 1778 map (London: Alexander Hogg, 1780).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Itinerary Maps: Paris, Ogilby, Bunyan

71

Fig. 2.9. The Road from London to Hith from John Ogilby, Britannia (For the colour version, please refer to the plate section.)

Fig. 2.10. Map of The Pilgrim’s Progress (1778) (For the colour version, please refer to the plate section.) Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

72

Historicising the Fictional Map

imagine themselves undertaking the same route and, ideally, achieving the same goal. The only difference is that in the case of the literary map, the route has no real-world correspondence at all and there is no possibility of it being undertaken on foot. The first two editions of Pilgrim’s Progress (1678) were published without a map or any illustrations until the third edition (1679), which contained an image of the narrator falling asleep and a figure of the pilgrim setting out on his journey behind him. Unusually in relation to this study then, Pilgrim’s Progress is a work for which any map is post-authorial (and thus, in principle, less integrated than a first edition map) despite the highly spatial nature of the text and the extended spatial metaphor at its heart.50 However, it is also a good example of a work for which a need for corresponding visual representation was strongly felt from the start. Natalie Collé Bak, in an article that argues that illustrations were an essential part of the work’s popularity, tells us: According to the advertisement included at the front of the fifth London edition, Bunyan’s first publisher Nathaniel Ponder provided the early readers of The Pilgrim’s Progress with illustrations in order to satisfy their demand for images . . . ‘And the Publisher observing that many persons desired to have it illustrated with Pictures, hath endeavoured to gratifie them therein’.51

Pilgrim’s Progress literalises the powerful spatial metaphor at the heart of Christianity: the need for the true believer to follow and remain on the right path, ‘Because strait is the gate, and narrow is the way, which leadeth unto life, and few there be that find it’ (Matthew 7:13, King James Bible). The map thus functions as a kind of allegorical road map that externalises the ideal projected spiritual journey of an individual. The particular form of the strip map, with its dominant visual linearity, is uniquely appropriate for this work, creating a remarkable conjunction between verbal and visual forms of representation. When we compare the text with the map, the dominant motif of the straight road also lies at the centre of the visualisation of the text and represents two key physical/metaphorical features of

50

51

Although Bunyan does not present an authorial map with Pilgrim’s Progress, he does clearly have a strongly spatial imagination and also produced a schematic diagram entitled ‘A Mapp Shewing the Order & Causes of Salvation & Damnation’, sold as a broadsheet in 1663–4. This fascinating spiritual map is reproduced in The Miscellaneous Works of John Bunyan, vol. XII, ed. W. R. Owens (Oxford: Clarendon Press, 1994), 417–23. Nathalie Collé Bak, ‘The Role of Illustrations in the Reception of Pilgrim’s Progress’, in Reception, Appropriation, Recollection: Bunyan’s Pilgrim’s Progress, ed. W. R. Owens and Stuart Sim (Frankfurt/Main: Peter Lang, 2007): 81–98, 84.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Itinerary Maps: Paris, Ogilby, Bunyan

73

the road: divergence from the straight route by side-routes that lead one astray, and unavoidable areas of difficulty that lie right across the path and must be encountered rather than avoided. These are depicted horizontally as areas lying across it (‘the Slough of Despond’; ‘the Hill of Difficulty’; ‘the Valley of the Shadow of Death’; ‘the River of Death). The strip map divides the narrative journey up into three main sections: 1. From the City of Destruction to the Valley of Humiliation 2. From the Valley of the Shadow of Death to Doubting Castle 3. From the Delectable Mountains to the gates of the Celestial City The strips make clear the way in which, as Christian travels further along his route, the landscape becomes increasingly less threatening, reminding us that the entire work is also intended to map onto the Christian life. The greatest challenges and upsets occur early on and dishearten the weak but, as faith and belief become stronger, so the way is less difficult because the traveller increasingly trusts in God. In the case of Pilgrim’s Progress, the ‘map’ used by Mr Great-heart as Guide in part II (described in the epigraph to this chapter) is also ‘the book’ because the entire landscape is simultaneously mapped onto, and readable through, knowledge of biblical wisdom: But he had in his Pocket a Map of all ways leading to, or from the Celestial City; wherefore he strook a Light (for he never goes also without his Tinder-box) and takes a view of his Book or Map; which bids him be careful in that place, to turn to the right-hand way. (233)

Perhaps the most interesting spatial element of the work is those points where the external landscape ceases to be universalised. Although it is generally depicted as an actual physical topography – objectively ‘there’ and experienced the same by all (as the map implies) – yet it is also somehow subjectively determined by the interior state of the person travelling across it and the degree to which they are self-reliant or recall the presence of God to help them in their journey. We see this in the Valley of the Shadow of Death, when Christian remembers to rely upon ‘the light’ and it helps him through the darkest places: [Y]et this second part which he was yet to go, was, if possible, far more dangerous: for from the place where he now stood, even to the end of the Valley, the way was all along set so full of Snares, Traps, Gins, and Nets here, and so full of Pits, Pitfalls, deep holes and shelvings down there, that had it now been dark, as it was when he came the first part of the way, had he had a thousand souls, they had in reason been

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

74

Historicising the Fictional Map

cast away, but as I said, just now the Sun was rising. Then said he, His candle shineth on my head, and by his light I go through darkness. (53)

The entire external landscape here functions as a projection of the internal state of the traveller journeying across it. The most powerful and explicit example of this occurs right at the end of the journey at the final barrier, where the depth of the river and the extent to which the ground is firm beneath one’s feet is determined by the strength of an individual’s faith: Then they asked the men, if the Waters were all of a depth. They said, No; yet they could not help them in that Case, for said they: You shall find it deeper or shallower, as you believe in the King of the place. (120)

Christian fears he will fail the final test, and so begins to drown, but is saved partly by the encouragement of Hopeful, who stands firm, and by putting his faith in God: entring, Christian began to sink, and crying out to his good Friend Hopeful; he said, I sink in deep Waters, the Billows go over my head, all his Waves go over me, Selah. Then said the other, Be of good chear, my Brother, I feel the bottom and it is good. (120–1)

At the point where the text represents a topography that is generated by the spiritual state of those moving over it, map representation reaches its limits.

Accurate Maps: Mapping the Nation All of the early fictional maps considered thus far have been either pictorial and illustrative or have maximised the uncertain limits of known geography to extend the real world into a fictional one, meeting the needs, not of the actual traveller, but of the armchair traveller enjoying reading about exotic worlds from the comfort of the fireside. All of this changes in the late eighteenth and early nineteenth centuries with the major redefinition of cartographic practices in terms of large-scale scientific activity, using trigonometry to create a grid of fixed points capable of extension over large areas and centred upon accuracy. Whereas map-making had previously been the private occupation of a few highly skilled individuals, it now demanded a significantly greater level of organisation. In Empire of Chance, Anders Engberg-Pedersen argues that from 1800 onward the cartographic revolution was also directly stimulated by the Revolutionary and Napoleonic wars and that ‘maps came to constitute a media a priori for the

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Accurate Maps: Mapping the Nation

75

planning and execution of large-scale warfare’.52 He goes so far as to argue that ‘the management of a military empire became dependent on its symbolic double’ (147) and describes the ‘paper empire’ (152) of Napoleon, who had a topographical bureau originally set up in the Tuileries Palace which then became a ‘mobile office during the campaigns’ (155).53 The mapping of Great Britain was certainly achieved in response to fear of attack, with the making of maps undertaken as a military activity designed to provide the knowledge on the ground that would give the British army the necessary advantage. William Roy spent forty years trying to initiate a major mapping project for England, but it was not until a second threat of invasion with the Napoleonic wars that first a Trignometrical Survey (1791) and then an Ordnance Survey of the nation was initiated by the Board of Ordnance.54 Instead, a French lead in mapmaking was taken by the Cassini family who (over four generations) from the late seventeenth to the late eighteenth centuries, significantly advanced the processes and the accuracy of French cartography on a systematic basis across the whole country, using a complex system of triangulation by means of a baseline and trig points on the ground. The process of triangulation by measured surveying, created a framework of accurate relative positions within which information could be represented at a uniform scale.55 In her biography of the Ordnance Survey, Map of a Nation, Rachel Hewitt explains the significance of this shift to an accurate mathematical model: triangulation promised to realise the Enlightenment ideal of perfect measurement. Because each corner of each triangle was also at the corner of further triangles, any errors could be quickly identified.56

The French survey was ‘the first survey of its kind in the world, and a model for all future national maps based on triangulation, including the

52

53

54

55

56

Anders Engberg-Pedersen, Empire of Chance: The Napoleonic War and the Disorder of Things (Cambridge, MA: Harvard University Press, 2015). See particularly chapter 5 on ‘Military Cartography and the Management of Space’, 146–83; 147. Engberg-Pedersen also describes the theories of Antoine-Henri Jomini who ‘developed a theory of war based on maps’ (159–62). Rachel Hewitt, Map of a Nation: A Biography of the Ordnance Survey (London: Granta, 2010), 69. For an excellent summary of triangulation, see Matthew Edney, Mapping an Empire: The Geographical Construction of British India 1765–1843 (Chicago, IL: University of Chicago Press, 1990), 19–21. For a full account of the French mapping project, see Hewitt, Map of a Nation, 66–92.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

76

Historicising the Fictional Map

Ordnance Survey’ (69). Four generations of the Cassini family brought it to fruition, and the third of these, Cassini de Thury, invited the British (under William Roy) to extend the triangulation from Paris to Greenwich. This project, finally begun in July 1787, was completed by August 1788 and (although Roy died in 1790) undoubtedly laid the groundwork for the future national mapping project. The Ordnance Survey formally began on 12 July 1791 under William Mudge, with triangulation undertaken from 1792 onwards using ‘the same baseline that Roy had used seven years earlier during the Paris– Greenwich triangulation’ (124).57 After a Trignometrical and then Interior Survey, the first Ordnance Survey map was finally published on 1 January 1801 at a scale of 1 inch : 1 mile, with the scale determined by the needs of the army.58 In view of fears of French invasion, this was (unsurprisingly) a map of Kent and related areas. Triangulation and accurate mapping of the entire country then followed for this First Series working from the south-east upwards and outwards, in a programme only finally completed in 1873. In the mid-century, with the introduction of the railways and after the mapping of Ireland at a larger scale, the Ordnance Survey started to develop a Second Series before it had completed the first. This County Series – mapped at a scale of 6 inches : 1 mile (1 : 10,560) – was begun in 1842 in the north of England (which had not yet been mapped for the First Series). It was then followed from 1854 by even more detailed largescale maps at 25 inches to 1 mile: ‘The 1 : 2500 scale was introduced in 1853–4 and by 1896 it covered the whole of what was considered to be the cultivated parts of Great Britain’.59 By the end of the century, Britain and Ireland had the most accurate scale maps anywhere in the world. Alongside such developments, emerging out of fear of war at home, we need also to take into account developments on the field of war itself. As Engberg-Pedersen makes clear, with ‘an urgent need for new methods of spatial management, the military map became the most advanced and

57

58

59

This baseline was ‘the line that traversed 5.19 miles of Hounslow Heath to the south-west of London’ (124). Hewitt describes this as ‘cartographically speaking, the most important spot in the British Isles. Until theodolites were replaced by satellites in the second half of the twentieth century, it was the bedrock of the national triangulation that underlay every single Ordnance Survey map’ (Map of a Nation, 126). The original scale maps of 1 inch to 1 mile (1 : 63,360) are known as the Old Series or First Series. Richard Oliver, Ordnance Survey Maps: A Concise Guide for Historians (London: The Charles Close Society, 1993), 21.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Accurate Maps: Mapping the Nation

77

efficient tool’ (147). Maps for battlefields, plans of campaigns and troop movements have an indirect influence on literary mapping, particularly in the late Victorian period when many writers had also been involved in the Boer War or written about it. John Buchan is a good example. His prolific output included a number of factual books about military campaigns, including the twenty-four-volume Nelson’s History of the War. This contained various battlefield and campaign maps so that, for him, the use of maps in fiction followed naturally. As a trained soldier, he already possessed the ‘cartographic literacy’ (165) that military mapping required.60 This major cartographic shift into accurate scientific methods for mapping has significant consequences for fictional mapping. Up to the nineteenth century, we have seen that fictional maps were quite strongly aligned to the levels of map knowledge and cartographical advances of their period. What we see from the mid-nineteenth century onward, however, is an increasing divergence in terms of historical contiguity. That is, the fictional map now increasingly seeks to differentiate itself historically from its own cartographic period and to associate itself with either a remote past (the map purports to be archaic within the text to which it corresponds) or to be presented in a self-consciously ‘amateur’ way as hand-made: a sketch map rather than a professional map. Of course, there remain notable exceptions, but on the whole, from this point on, the fictional map seeks to distinguish itself from its real-world counterpart.61 Changes to map production may also contribute. Whilst the Ordnance Survey continued to use a copperplate method well beyond the introduction of lithography in the 1820s, map-making became far more professionalised and politicised and elements such as the use of symbols on maps were standardised at major meetings.62 As map-making becomes increasingly scientific and state-owned, creative resistance to the imposition of the apparent authority and objectivity of the map (already present in Gulliver’s Travels) starts to emerge. It is striking that in the Romantic period – which corresponds directly to major cartographic advances – there are remarkably few fictional maps

60

61

62

Engberg-Pedersen states: ‘reading a military map is an operational praxis that consists in active visual manipulation that generates new knowledge’ (Empire of Chance, 165). Obvious exceptions include the maps of Jules Verne for his series Voyages extraordinaires; and the maps for Erskine Childers’s The Riddle of the Sands considered in Chapter 3 below. The Ordnance Survey only finally gave up on copperplate in the 1870s. See David Smith, Victorian Maps of the British Isles (London: B. T. Batsford Ltd, 1985), 17–21. Lithography was invented in 1796 and fully introduced to British mapping in the 1820s. See Walter W. Ristow, ‘Lithography and Maps, 1796–1850’, in Five Centuries of Map Printing, ed. Woodward, 77–112.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

78

Historicising the Fictional Map

presented alongside texts – thus implicitly suggesting such resistance and favouring other ways of understanding the world. We see this anticipated in Jean-Jacques Rousseau’s advice for the teaching of geography to children. In Émile, he directly attacks teaching through forms of cartographic representation: Vous voulez apprendre la géographie à cet enfant, et vous lui allez chercher des globes, des sphères, des cartes: que de machines! Pourquoi toutes ces réprésentations? Que ne commencez-vous par lui montrer l’objet même, afin qu’il sache au moins de quoi vous lui parlez!63 (You want to teach geography to this child and you go and get globes, cosmic spheres, and maps for him. So many devices! Why all these representations? Why do you not begin by showing him the object itself so that he will at least know what you are talking to him about!)64

Unsurprisingly, Rousseau advocates a natural model, not one that distances the child from direct learning. He also explicitly rejects accuracy – ‘il ne s’agit as qu’il sache exactement la topographie du pays’ (171) (‘the goal is not that he knows exactly the topography of the region’ (435)) – in favour of a more directly understood, experiential geography. In a similar vein, Sir Walter Scott (the major British novelist of the period and a highly spatial writer) states in a letter to an antiquarian seeking to identify locations in his Minstrelsy of the Scottish Border that: [Y]ou are to know that I am an utter stranger to geometry, surveying, and all such inflammatory branches of study, as Mrs Malaprop calls them. My education was unfortunately interrupted by a long indisposition, which occasioned my residing for about two years in the country with a good maiden aunt who permitted and encouraged me to run about the fields as wild as any buck that ever fled from the face of man. Hence my geographical knowledge is merely practical.65

Direct (Wordsworthian/Rousseauvian) engagement with nature is here prioritised (however self-deprecatingly) over the controlled rationality of map-making. An anti-Cartesian account of human spatiality also brings

63

64

65

Jean-Jacques Rousseau, Oeuvres complètes IV: Émile, ed. Bernard Gagnebin and Marcel Raymond (Paris: Editions Gallimard, 1969), 430. Jean-Jacques Rousseau, Émile or On Education, trans. Alan Bloom (Harmondsworth: Penguin Books, 1991), 168. The Letters of Sir Walter Scott, 12 vols., ed. H. J. C. Grierson (London: Constable & Co., 1832–6), I, 174. Chris Donaldson (to whom I am indebted for this reference) gives a fuller account of this experiential geography in ‘The Depth of Walter Scott’s Geography’, in Romantic Cartographies, ed. Sally Bushell, Julia Carlson and Damian Walford-Davies (Cambridge: Cambridge University Press, 2020).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Accurate Maps: Mapping the Nation

79

with it an anti-cartographic stance. Mapping as an act of geometrical abstraction in which man adopts a distanced relationship to the world is set against a more dynamic, lived alternative. Later twentieth-century ecological or phenomenological resistance to the map, and the assertion of other forms of mapping such as ‘psychogeography’ find their origins here. It is rare to find a nineteenth-century work of literature that reproduces an accurate Ordnance Survey map of a region. However, indirect use of Ordnance Survey maps – where a map is described within the text but not explicitly visually represented – does occur. Victorian pride in the superior quality of Ordnance Survey mapping (which provided a far greater level of accurate detail than maps made in other countries) is explicitly voiced in Dracula when Jonathan Harker notes: ‘I was not able to light on any map or work giving the exact location of the Castle Dracula, as there are no maps of this country as yet to compare with our own Ordnance Survey maps.’66 As we might expect, by the end of the nineteenth century, Sherlock Holmes, with his passion for rigour and scientific accuracy, draws upon such maps when the need arises.67 In The Hound of the Baskervilles (1902), a mysterious beast has killed the first owner of Baskerville Hall (Sir Charles Baskerville) and threatens his young heir (Sir Henry), who calls in Holmes and Watson to help him. Possible suspects are identified in the few isolated houses around the Hall, itself located on the wilds of Exmoor. Holmes prepares himself for the case in London by spending a day undertaking a careful study of an Ordnance Survey map: ‘Where do you think that I have been?’ ‘A fixture also.’ ‘On the contrary, I have been to Devonshire.’ ‘In spirit?’ ‘Exactly. My body has remained in this armchair. . . After you left I sent down to Stanford’s for the Ordnance map of this portion of the moor, and my spirit has hovered over it all day. I flatter myself that I could find my way about.’ ‘A large scale map, I presume?’ ‘Very large.’ He unrolled one section and held it over his knee.68

66 67

68

Bram Stoker, Dracula (London: Archibald Constable & Co., 1897), 8. Holmes uses maps less frequently than one might expect. Across the entire series of books and stories, only three diagrammatic sketch maps are given: for ‘The Adventure of the Naval Treaty’, ‘The Adventure of the Golden Pince-nez’ and ‘The Adventure of the Priory School’. Arthur Conan Doyle, ‘The Hound of the Baskervilles: Another Adventure of Sherlock Holmes’, The Strand Magazine, vol xxii, no. 129 (September 1901), 247.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

80

Historicising the Fictional Map

The exchange above, between Holmes and Watson, exemplifies the power of the later County Series maps to provide detailed knowledge without direct experience. Although Holmes has never visited the region, he has now seen it from above (‘hovered over it’). The level of detail on the largescale map has provided him with sufficient information for him to visualise and internalise the space and begin to make deductions relevant to the case. For this scale ‘the names of houses and gardens, farmsteads and other features are given’ (Delano-Smith and Kain, English Maps, 222) and the series included an extraordinary level of detail (such as ‘railway signal boxes, milestones, greenhouses, village ponds’ (222)). Clearly this is the information Holmes draws upon, as he goes on to make direct use of the map in relation to the case: ‘Here you have the particular district which concerns us. That is Baskerville Hall in the middle.’ ‘With a wood round it?’ ‘Exactly. I fancy the Yew Alley, though not marked under that name, must stretch along this line, with the moor, as you perceive, upon the right of it. This small clump of buildings here is the hamlet of Grimpen, where our friend Dr. Mortimer has his headquarters. Within a radius of five miles there are, as you see, only a very few scattered dwellings. Here is Lafter Hall, which was mentioned in the narrative.’ (247)

Holmes points out various features to Watson, making clear the extent to which geographical and spatial knowledge is a significant part of the story. The map is immediately made an active part of the investigation. Its geometrical relations easily translate into human location and then into human relations, both physically and psychologically. Clearly, then, a reproduction of this map for the reader would also be helpful. An illustration of Holmes and Watson examining the map as a large sheet was given within the text (see Fig. 2.11) in its first serial publication in The Strand Magazine (247) but not the map itself – an act of deliberate and selfconscious withholding of information from the reader. This lack of a map emphasises the extent to which the reader is intended to admire Holmes as superior, rather than attempting to emulate him, but it is also slightly frustrating. Intentionally or not, this replicates the position of the narrator, Watson, who, for much of the book, is also acting in the dark. When Watson finally tracks Holmes down (hidden in a prehistoric hut on the moors) he is at first relieved, then slightly piqued: ‘“Then you use me, and yet do not trust me!” I cried with some bitterness’ (124). The reader might make the same complaint.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Accurate Maps: Mapping Empire

81

Fig. 2.11. ‘The Hound of The Baskervilles, Chapter 3: The Problem’

Accurate Maps: Mapping Empire Accurate mapping at home worked primarily to protect the home nation by providing the army and government with detailed knowledge, superior to any owned by invading forces. Accurate mapping abroad was undertaken with a less morally justifiable aim: to take possession and assert authority and control over land and peoples acquired by the nation. In the nineteenth century, once an accurate world map and national European maps had been established, mapping as a means of asserting European claims to ‘unmapped’ regions emerged as a key colonial activity: ‘Map-making became the servant of colonial plunder, for the knowledge constituted by the map both preceded and legitimized the conquest of territory.’69 As James Akerman states: ‘A wide range of mapping practices underscored the transformation of Europe’s oceangoing empires that . . . by the end of the nineteenth century had penetrated virtually every corner of the earth.’70 Such acts of mapping then extend out to the colonised who ‘develop their own cartographic responses to imperialism’ (3) and into larger culture as ‘public forms of cartography, such as journalistic and

69

70

Anne McClintock, Imperial Leather: Race, Gender and Sexuality in the Colonial Context (London: Routledge, 1995), 27. James R. Akerman, ed., The Imperial Map (Chicago, IL, and London: University of Chicago Press, 2009), 3.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

82

Historicising the Fictional Map

literary mapping, contributed to the formation of the popular imagination of the empire at home’ (3). Imperial maps, by their very nature, seek to establish an uneven relationship between those who make and use them and the land mass and people who are subjected by, but do not have access to, that knowledge. Edward Said, in Culture and Imperialism, places geography and the possession of territory at the heart of his definition when he states: ‘At some very basic level, imperialism means thinking about, settling on, controlling land that you do not possess, that is distant, that is lived on and owned by others.’71 Said tells us that: By 1914 . . . Europe held a grand total of roughly 85 percent of the earth as colonies, protectorates, dependencies, dominions, and commonwealths. No other associated set of colonies in history was as large, none so totally dominated, none so unequal in power. (8)

Unsurprisingly, then, as Matthew Edney makes clear in the opening sentence of Mapping Empire, ‘Imperialism and mapmaking intersect in the most basic manner. Both are fundamentally concerned with territory and knowledge’.72 At almost exactly the same time that the Ordnance Survey undertook their mapping project at home, British map-makers were also busy abroad, mapping India in order to maintain control of the East India Company’s valuable colonial commodity and assert an alternative ordering of the space that challenged and devalued the claims of pre-existing systems of regional knowledge. The Great Trignometrical Survey of India can be said to have begun in 1802 (when William Lambton undertook the first triangulation with a base line at Madras) and was completed by 1878.73 As with many forms of imperial mapping, it was strongly supported by the Royal Geographical Society (RGS; established in 1831) and had as its explicit aim to pull together and subsume all prior (less ‘professional’) forms of mapping. It is easy to view this major national undertaking through a postcolonial lens as the successful assertion of cartographic ‘power as knowledge’ in which the native loses ownership even of his own geography: Imperialism after all is an act of geographical violence through which virtually every space in the world is explored, charted, and finally brought under control.

71 72

73

Edward W. Said, Culture and Imperialism (London: Chatto & Windus, 1994), 5. Matthew Edney, Mapping Empire: The Geographical Construction of British India 1765–1843 (Chicago, IL: University of Chicago Press, 1990), 1. Work towards the map preceded the naming of it as ‘The Great Trignometrical Survey’ in 1818.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Accurate Maps: Mapping Empire

83

For the native, the history of his or her colonial servitude is inaugurated by the loss to an outsider of the local place.74

However, as Edney notes, there is a contrast between the ideological and political aspirations of mapping projects and the reality on the ground. This leads him into a clear-sighted critique of Said: The problem with Edward Said’s conception of European imperialism is that it is too monolithic . . . European states and their empires could never be so totalizing. They could never be so effective . . . Even if it were epistemologically possible to construct the perfect totalizing knowledge archive, it would have been institutionally impossible actually to do so. (25)75

A text that intersects in fascinating ways with this colonial project is Rudyard Kipling’s Kim.76 As in the case of Sherlock Holmes above, direct cartographic representation of India is not provided by Kipling, but the entire text – centred upon Kim’s malleable hybrid identity – is about the tension between two different kinds of mapping taking place under the auspices of the Survey, but functioning in very different ways. In a fascinating article that draws out the complexities of imperialism as its objectives changed over time, Phillip E. Wegner makes the point that: ‘the text contains not one, but two quite different Realities’.77 Wegner notes that, unlike the earlier model of imperialism offered by works such as H. Rider Haggard’s King Solomon’s Mines (considered in the next chapter) – which limited itself to ‘those as-of-yet-uncolonized spaces’ (136) – Kim embodies the greater complexities of a ‘shift from a regional to a global model of imperialism’ (135). As a result, Wegner asserts, Kim is always played out against two ‘setting-frames’: In Kim . . . narrative space is doubled, so that the adventures of young Kimball O’Hara take place simultaneously in two different settings. The first of these setting-frames is constituted by the immediate, phenomenological space that Kim occupies – the ‘exotic landscape’ of India that continues to dazzle so many readers 74

75

76

77

Edward Said, ‘Yeats and Decolonisation’, in Nationalism, Colonialism and Literature, ed. Terry Eagleton, Frederick Jameson and Edward Said (Minneapolis: University of Minnesota Press, 1990), 69–95; 77. More specifically, Edney points out that the GTS did not employ the totalising power of triangulation correctly. It required that ‘the technology be implemented in the proper sequence . . . triangulation has to be completely computed and corrected before any detailed surveys are begun . . . The British surveys did not follow the proper sequence’ (29). Said famously wrote on Kim in Culture and Imperialism (159–96), later adapted for his introduction to a Penguin edition (Harmondsworth: Penguin, 1987). Phillip E. Wegner, ‘“Life As He Would Have It”: The Invention of India in Kipling’s Kim’, Cultural Critique 26 (Winter 1993–4): 129–59; 129–30.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

84

Historicising the Fictional Map

of Kipling’s work. As the narrative progresses, however, the reader learns that this first level of ‘reality’ is always-already contained within a larger, second context. Kipling refers to this latter frame as ‘The Great Game’ a figure that we can now read as the sign manifest in the local Indian political context of the absent presence of the global empire’s massive structure. (136)

Kim’s uniqueness (but also his difficulty in terms of self-identity) stems from his ability to harmonise with, and straddle, both contexts, as few can. In Kim doubling manifests itself cartographically in terms of two forms of nineteenth-century map-making: the Great Trigonometrical Survey (GTS) of all India, by the British Empire; and the irregular route survey mapping of the dangerous northern borders by local ‘pandits’.78 From the 1850s onward, under the leadership of Thomas Montgomerie, the GTS was extended into the remote and dangerous regions between India and Russia. However, because it would have been too dangerous for a white man to attempt to map these borders, Montgomerie trained local Indians to do this. Their accounts were reported to the RGS in the Survey’s annual reports and were greatly valued – not least because they validated the very act of colonial mapping: ‘The surveys were part and parcel of the British belief that they pursued a liberal, rational empire as much for the benefit of the Indians as for themselves’ (Edney, Mapping an Empire, 317). From the 1860s onward the stories of the exploits of these local border spies were a key feature of the Survey’s reports.79 However, as Edney makes clear, in a strongly hierarchical British mapping structure, the knowledge permitted to these natives was relatively limited: Indians could manage the basic, repetitive, and laborious tasks of the detailed topographic or cadastral survey. But the British kept the higher-level science of geodesy – informed by calculus, complex geometries and Newtonian mechanics – out of the reach of all but a very few Indians. (319)

As a result, the borders were deliberately mapped in an entirely anachronistic way by means of a route survey rather than by the more sophisticated means of triangulation. The route survey (dominant in the late eighteenth century and phased out of the main Survey by the 1840s) was faster than triangulation, and more able to be done surreptitiously, but also far less reliable since it

78

79

The distinction here between survey and route mapping anticipates the two dominant inner forms of the cognitive map that are the focus of Chapter 8. Ian J. Barrow, Making History, Drawing Territory: British Mapping in India 1756–1905 (Oxford: Oxford University Press, 2003), 130.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Accurate Maps: Mapping Empire

85

required the bringing together of many different individuals’ routes to achieve any kind of whole understanding.80 Edney provides a concise summary of it: Technically termed a ‘traverse,’ a route survey involves the measurement of the distance and direction of each leg of the route traveled. The simplest and most common instances of the route survey measured directions with a compass and distances either with a perambulator, a wheel with a counter attached to record the number of revolutions, or by time-of-travel, with an assumed speed. (92)

In centring his novel on this outdated mode of mapping by native Indians loyal to the British, Kipling was meeting the needs of his readership back home (and drawing directly on the rich material of the Survey). Kim is explicitly being trained up in this service (although he is presumably of even greater value to the Empire than the true ‘pandits’ because he is not actually Indian but merely ‘burned black as any native’).81 Ian Barrow describes the deliberate use of pilgrimage as a cover for surveying activity by the pandits: To facilitate their passage over the border passes, the surveyors disguised themselves as pilgrims and adapted religious instruments to suit their purposes. A small compass and bearings written on paper were concealed in a prayer wheel, while a rosary . . . only contained 100 [beads] with every tenth bead being slightly larger than the others. (139)

Again, we see this directly replicated in Kim who is first recruited because ‘A wandering lama with a low-caste boy-servant might attract a moment’s interest as they wandered about India, the land of pilgrims; but no one would suspect them’ (23). Later on, in his first practice mission to draw a map secretly, his actions are described in ways that directly reflect the historical description above: the Colonel ordered him to make a map of that wild walled city; and since Mohammedan horse-boys and pipe-tenders are not expected to drag Survey-chains round the capital of an independent Native State, Kim was forced to pace all his distances by means of a bead rosary. He used the compass for bearings as occasion served . . . and by the help of his little Survey paint-box . . . he achieved something not remotely unlike the city of Jeysulmir. (144)

80

81

For a reading of narrative itinerary against the form of the map, see also Sailaja Krishnamurth, ‘Reading between the Lines: Geography and Hybridity in Rudyard Kipling’s Kim’, Victorian Review 28.1 (2002): 47–65. Rudyard Kipling, Kim, ed. Zohreh T. Sullivan (New York and London: W.W. Norton & Co. Ltd, 2002), 3.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

86

Historicising the Fictional Map

At the same time, however, what literature can do, that history cannot, is to create an ambiguity around the representation of characters and subjects that may or may not be intentional (so that the imperial narrative slips knowingly or unknowingly into critique). This is exactly what happens in Kim – which in turn replicates the limits of the Survey to assert itself in the ways it intended. As Krishnamurth points out: The project of the map exists along with the project of the mediated narrative itinerary, the narrative of the frontier, which produces hybrid identities in its ambivalent space. Yet it is these hybrids who constitute the workings of the mapping machine. The British colonizers are dependent on the ability of Kim and the Babu to move between identities. (61)

Mapping on the ground, by individuals, is very different from the larger abstractions of projected cartographic control. In Kim, whilst he is ostensibly using pilgrimage as a cover for mapping in secret, that cover constantly threatens to take over and undo all the careful training up of ‘such a colt [who] knows the game by divination’ (98). Ian Baucom puts this well in his account of the Indian Survey in Kim: The wanderings, displacements and spatial uncertainties of nomadism seem, then, to articulate a direct threat to the spatial and temporal permanencies of cartography, to be the surplus that the map not only cannot represent but must suppress.82

So, for example, the metaphor of ‘the Road’ applies equally to the linear mapping of the route surveyors as part of the Great Game (‘“Lurgan Sahib and I will prepare him for the Road”’ (148)); the heart of ‘real’ India (‘the Great Road which is the backbone of all Hind’ (51)) and the Lama’s spiritual search for ‘The Way’ with Kim as his beloved chela. The fact that the narrative ceases with the conclusion of the Lama’s successful quest, and his desire to reward Kim and establish an alternative future for him (‘“Nay – he must go forth as a teacher”’ (236)) seems to me to assert itself against the dominant teleology of Kim’s cartographic masters (in ways that Said does not allow for). Not only that, but the Lama’s transcendent vision at the end sets itself against the perfect imperial cartographic vision of all India that Kim’s exploits have been contributing to: ‘At that point, exalted in contemplation, I saw all Hind from Ceylon in the sea to the Hills . . . I saw every camp and village, to the least, where we have ever rested. 82

Ian Baucom, Out of Place: Englishness, Empire, and Locations of Identity (Princeton, NJ: Princeton University Press, 1999), 94.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Ending Where We Began

87

I saw them at one time and in one place, for they were within the Soul. By this I knew the Soul had passed beyond the illusion of Time and Space and of Things. By this I knew that I was free.’ (239)

So Kim is leading a doubled double-life that vacillates between recruitment to ‘The Great Game’ and the possibility of resistance to such a future. The open-ending of the novel in particular seems to threaten to undermine all the careful planning of his British superiors, just as the Great Trigonometrical Survey can only ever be (as Edney reminds us): ‘a study of how the British represented their India. I say “their India” because they did not map the “real” India. They mapped the India that they perceived and that they governed’ (2).83

Ending Where We Began I want to conclude this not-so-brief history by looking at a fictional map that we will return to in the next chapter: the map published with the first edition of Treasure Island (see Fig. 2.12) that also gestures back to the earliest forms of medieval mapping with which this chapter started. Whilst this novel was written at the height of the British Empire, and indirectly depicts a classic act of British colonial plundering, the map emphatically presents itself as not of its time: it aligns itself not to the major cartographic leaps of the nineteenth century but to an earlier period of ‘treasurehunting’ and piracy. Of course, this is in line with the text, which makes it clear that the book is set in the mid-eighteenth century. Jim tells us at the start that ‘I take up my pen in the year of grace 17 and go back to the time when . . .’ and the map is annotated ‘Given by above J.F. to Mr W. Bones Master of ye Walrus Savannah this twenty July 1754 W B’.84 Various stylistic map elements combine to create the sense that the map is archaic. First, there is a highly decorative scale bar across the top that compares the distance of one inch on the map to ‘A Scale of 3 English Miles’. This allows the reader to envisage the size and contents of the island as the characters travel around it and it also relates something alien to something familiar (the ‘English’ mile). The map is also dominated visually by a wind or 83

84

Cf. Brotton’s account of the RGS: ‘These organisations promoted a cartography that was more of an ideological projection based on apparently objective scientific principles than an administrative reality’ (History of the World in Twelve Maps, 343). Robert Louis Stevenson, Treasure Island (London, Paris and New York: Cassell & Co. Ltd, 1883), 1.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

88

Historicising the Fictional Map

Fig. 2.12. Map from Treasure Island (For the colour version, please refer to the plate section.)

compass-rose with 32 rhumb lines radiating from it and a fleur-de-lys at the top for North. Such features make it directly comparable to the earliest sea maps of fourteenth- to sixteenth-century Europe: portolan charts that accurately mapped the interior coastline of the Mediterranean for the purposes of trade. As Tony Campbell reminds us, ‘the earliest known portolan chart is thought to be almost exactly contemporary with the Hereford world map’ but the difference between these astoundingly accurate charts and the early maps with which this chapter began is that ‘The medieval mappaemundi are the cosmographies of thinking landsmen. By contrast, the portolan charts preserve the Mediterranean sailors’ first-hand experience of their own sea.’85 85

Tony Campbell, ‘Portolan Charts from the Late Thirteenth Century to 1500’, in History of Cartography, Vol. 1. Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean, ed. J. B. Harley and David Woodward (Chicago, IL: University of Chicago Press, 1987), 372. Campbell tells us that there are around 180 surviving portolan charts (373).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Ending Where We Began

89

Fig. 2.13. Jorge de Aguiar’s Portolan chart of the Mediterranean Sea . . . as far south as Sierre Leone (For the colour version, please refer to the plate section.)

By far the most striking feature that the portolan charts had in common was their use of rhumb lines: sixteen or thirty-two intersection lines that radiated across the map and linked up with secondary centres around the edge (see Fig 2.13). These allowed for straight line navigation using a compass. At the point of intersection of these lines the design of a ‘compass rose’ was inserted. When we compare these features with the map at the front of Treasure Island, we can immediately see that the fictional map

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

90

Historicising the Fictional Map

draws upon the earlier cartographic form. In fact, this similarity extends into even more specific details that it shares with the portolan chart: ‘place names written inland at right angles to the coast . . . more important places are named in red, the rest in black . . . coastlines are shown in a generalized fashion with bays and headlands exaggerated and navigational hazards marked with crosses and dots’.86 All of these features signal that the map itself has a history preceding that given in the story and that Captain Flint must himself have inherited it from other earlier pirates. The names on Treasure Island appear to have been given by sailors or pirates regularly visiting it (‘Fore-mast Hill’, ‘Hautbowline Head’, ‘Skeleton Island’), and the name of ‘Captain Kidd’s Anchorage’ (referring to the real Captain John Kidd, 1664–1701) implies that this goes back to the late seventeenth century at least, whilst the style of the map signals that it is even older.87 In this case, then, the map is both part of and apart from the text to which it corresponds.88 In other words, not only is there a sense in which the entire narrative is the story of the map, but there is also the possibility that the map could be telling other stories from even earlier times: the map precedes and exceeds the text. This identity of the fictional map as an object that functions in a strongly dynamic and integrated way alongside the text but that also threatens to exist beyond it, becomes a core characteristic that emerges ever more strongly in the twentieth century with the iconic maps of Tolkien’s Middle-Earth (as we shall see in Chapter 6). It seems to me at least possible (if hard to prove) that a single highly influential author, or even a single map, might function as a catalyst for maps in literature within a particular culture. In France and Britain, such a role could be performed by the maps found in many of the volumes of Jules Verne’s Voyages extraordinaires series.89 Richard Phillips reminds us that, 86

87

88 89

Cartographical Innovations: An International Handbook of Mapping Terms to 1900, ed. Helen M. Wallis and Arthur H. Robinson (Tring: Map Collector Publications Ltd, 1987), 12. This is also a point where the fictional pirate world intersects with the real, since Kidd was reputed to have left an undiscovered treasure cache behind him when he was hanged for treason in 1701. In this way, as discussed in Chapter 1 above, it both is and is not a paratext. Verne’s first map is earlier than that of Treasure Island, appearing in The Voyage of Captain Hatteras (1866). Verne employed the latest cartographic methods and made the maps himself for three of the works including The Voyage of Captain Hatteras. See Terry Harpold, ‘Verne’s Cartographies’, Science Fiction Studies 32.1 (March 2005): 18–42; Richard Phillips’s ‘dominant ‘and ‘alternative’ readings of Verne in Mapping Men and Empire: A Geography of Adventure (London and New York: Routledge, 1997), 129–42; and Jörg Dunne, ‘Map Line Narratives’, in Literature and Cartography: Theories, Histories, Genres, ed. Anders Engberg-Pedersen (Cambridge, MA: MIT Press, 2017), 352–79. See also Jacob, Sovereign Map, 281, 287–8.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

Ending Where We Began

91

in part because of his ‘penchant for British heroes and British Empire settings’, Verne was extremely popular in Britain: ‘his cultural significance in Victorian and Edwardian Britain was second to none’.90 However, it seems equally likely that the map for the hugely popular Treasure Island (1883) challenges for such a role. I want to suggest that it functions as a catalyst map for the nineteenth century that opens up the presence of a map alongside the text for a far wider range of literary genres than had previously been the case. At any rate, it is certain that, from this point on, literary maps have a far greater tendency to be anachronistic, to draw upon early forms of personalised and artistic mapping and to eschew the use of accurate scientific techniques. (The most common form is one which purports to be a handdrawn sketch or map.) This means that their practical usefulness is frequently brought into question within the narrative (where they may appear to be out of date or amateurish) but it also draws attention to the fact that they are not intended to function in the same way as actual maps of the world. An experience of literary space is not the same as an experience of lived space; the mapping of the fictional world is not the same as the mapping of the real world. From the nineteenth century onward, fictional maps in their appearance and content go out of their way to tell us this.

90

Phillips, Mapping Men, 129.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.003

3

Doubleness and Silence in Adventure and Spy Fiction

We should be prepared to regard silences on maps as something more than the mere absence of something else. J. B. Harley, ‘Silences and Secrecy’

92

The disjuncture between real-world and fictional map forms that the previous chapter identified as emerging in the late nineteenth century anticipates the need for a turn towards maps within genres. When we examine fictional maps not just in terms of individual works spread out over time, but comparatively within a specific genre, new spatio-temporal meanings and patterns emerge. This chapter, then, is the first of four chapters centred upon particular genres for which the presence of the map alongside the text emerges as a core characteristic. The chapters are loosely chronological, starting in the second half of the nineteenth century and running on into the mid-twentieth century, charting the way in which the fictional map begins to spread across generic forms and to be found in a range of authors working within a single genre (rather than being viewed as a rare historical aberration). The presence of maps in adventure fiction – the generic focus of this chapter – is directly linked to the underlying ideology of Empire. At one level, historical maps showing the spread of empire across the world, or the activities of real-life explorers into Africa, make clear the levels of public interest in the acquisition of overseas territories and the linkage of models of manhood and heroism to the mapping of them. At another, the active making of maps by those explorers (for whom this was often the prime purpose of exploration) enables ongoing European control and exploitation of the colonies. Thus, when we turn from real life to fictional voyages of exploration, we can see that both the text and the map as found in works of ‘Boy’s Adventure’ occur as a natural extension of imperialism into a literary form that both partakes of and promotes that ideology. At the same time, however, it is worth heeding Matthew Edney’s qualification that: ‘When we study in detail the practices of producing and consuming maps, we find that there is actually little that distinguishes imperial from non-imperial

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Doubleness and Silence in Adventure and Spy Fiction

93

cartographies’.1 Maps have always been used by governments to ‘extend authority and transform land into territory’ (11), and ‘empire’ is simply a shared concept (creating an ‘imagined community’) that is ‘constructed through cartographic discourses that represent territory for the benefit of one group but that exclude the inhabitants of the territory represented’ (13). For literary works, the sense of an imagined community of empire is extended through imaginative fiction to the next generation.2 A second related reason for the spread of maps at this moment in history, not considered in the previous chapter but indirectly relevant, is the emergence of thematic mapping as a new way of representing spatial knowledge in the latter part of the nineteenth century. A thematic map ‘depicts the spatial distribution and variation of a chosen subject or theme which is usually invisible, such as crime, disease or poverty’.3 In To the Ends of the Earth, Jeremy Harwood describes how ‘the growth of scientifically inspired thematic cartography changed the way in which we present and understand information forever’ (141). Whilst thematic mapping started in relation to scientific disciplines such as geology or climatology, it soon began to be used to represent other forms of information in imaginative ways: ‘thematic maps developed rapidly from the early 1800s with the growth in quantitative statistical methods and public censuses’.4 Once maps are no longer concerned with merely representing geographic data but can be used to represent many forms of information, (social, cultural and historical), there is nothing to stop them also being used to spatialise the fictional lives and worlds of literature. This chapter is centred upon adventure fiction as the first genre in which the map functions as an integral element of the work but it also reads the fictional map against itself in terms of its doubleness and untrustworthy nature. It compares two iconic and highly influential examples from the late nineteenth century: Treasure Island and King Solomon’s Mines. These two Victorian examples are chosen for their generic status as early imperialist adventure novels which not only invite the presence of a (colonising) map but actively deploy the use of a map as part of the quest narrative. 1

2

3 4

Matthew H. Edney, ‘The Irony of Imperial Mapping’, in The Imperial Map, ed. James R. Akerman (Chicago, IL, and London: University of Chicago Press, 2009), 11–45; 11. See Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1983), in which he states that ‘the nation . . . is an imagined political community’ (15) and that ‘Nationalism . . . invents nations where they do not exist’ (15). Brotton, History of the World in Twelve Maps, 343. Brotton, History of the World in Twelve Maps, 343.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

94

Doubleness and Silence

The final sections of the chapter then move on to a related genre, that of spy fiction, and focus on Erskine Childers’s The Riddle of the Sands, in which the doubleness of both map(s) and text effectively brings into being a new form. Here, again, the maps form part of a particular historical moment and national state, albeit one now centred less upon confident acquisition of foreign lands and more on fear of invasion and defence of the homeland. The narrative is understood to be a ‘riddle’ rather than a quest, but one that is strongly spatial and requires maps for it to be solved.

Trusting the Fictional Map: Accuracy and Use Value This chapter will focus on three key concepts emerging from both traditional and critical cartography that are directly relevant to the mapping of represented worlds in works of fiction. These are: the accuracy of the map and the extent to which it can be trusted; the concept of cartographic silence; and the authenticity or otherwise of the map as well as its capacity for replication (doubling). As we saw in Chapter 2, map accuracy has been a fundamental element of traditional cartography from the late eighteenth century onward. At the same time, however, the very concept of ‘accuracy’ is itself brought into question by critical cartography since it involves a false naturalisation of the map as a form of representation. In The Power of Maps, Denis Wood takes Harley’s ideas further when he responds to the map as ‘a form of repression . . . a denial that anything is denied’ (77) which he describes as an inherent property of all maps that transcends any particular history or culture. Wood cites the example of Van Sant’s satellite image of the world from space, first published in National Geographic in November 1990, as ‘the acme of cartographic perfection, an image of the world so true as to render all the questions raised . . . out of date’ (49). But – of course – it is not this at all; rather, it is a highly sophisticated (and carefully selective) representation presenting itself as ‘a map that would like us not to notice the fact that it is a map’ (49). If all maps are understood to be a kind of manipulative fiction (rather than unquestioningly accepted as a neutral, factual representation of the physical world), then fictional maps cease to be a special case, and become instead a more extreme version of real-world maps with the potential to illuminate this central, deceptive, aspect of the map form. So, for example, the assumption of direct correspondence between map and world is immediately thrown into question when we consider the complex issue of map accuracy in relation to a map in a literary work. The first question that

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Trusting the Fictional Map

95

arises is ‘accurate to what?’ since there is no immediate physical place to which the map can correspond. In relation to the content of a map for a fictional world, we can envisage three main kinds of map: a map which purports to represent actual topography, with names and features directly corresponding to places in the real world (e.g. maps of the Dutch–German coastline in Erskine Childers’s The Riddle of the Sands); a map which merges actual and imaginative representations (e.g. Thomas Hardy’s Wessex map; Arthur Ransome’s maps) or nests the imaginative within the real (Trollope’s Barsetshire); a map which represents purely imaginative (even fantastic or otherworldly) topography (e.g. J. R. R. Tolkien’s Middle-Earth). When the fictional map does not have any real-world correspondence, it draws attention to the agency inherent in the visual representation itself and the fact that even real-world maps are, in some sense, ‘fictional’. Thus, we see that the fictional map functions as a more extreme, self-evident (and potentially ‘knowing’) example of the fictionality of all maps (a truth that is just much harder to see in the case of real-world maps which appear to correspond directly to the physical world). The ‘accuracy’ value of a fictional map necessarily functions in terms of the relationship between visual representation (usually at the front of the book) and verbalisation corresponding to it within the text. The usefulness of the fictional map (which is directly bound up with its ‘accuracy’) thus becomes centred upon the vital connection between map and narrative. This works at one level within the text (where the use value of the map is often put under pressure) and at another for the reader – who may use the map mimetically in the same way as the characters to move across the landscape, but is also using it in other more self-conscious/extrinsic ways in relation to plot, narrative and character (as well as being capable of analysing and critiquing the usefulness of the map at any level). In terms of imperialist forms of mapping (in which adventure fiction partakes), the doubleness that we know to be inherent in the fictional map is replicated in a doubleness of audience, as described by Matthew Edney: Imperial mapping is thus an ironic act, postulating as it does a double audience: the population in the mapped territories remains ignorant while another population is actively enabled and empowered . (‘Irony of Imperial Mapping’, 13)

The act of being able to map ‘accurately’ is bound up with this: ‘Cartographic science became, within European discourses, a crucial marker of difference between Europeans (the knowing Self ) and non-Europeans (the unknowing Other)’ (13). Edney describes this tendency as a deliberate act

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

96

Doubleness and Silence

of ‘self-empowerment that elicits an emotional response: pride, gratification . . . but also perhaps fear and anxiety’ (32). It is in this emotional register, if anywhere, that the nature of the imperial map is to be found – and we can easily locate just such emotions in King Solomon’s Mines. Even in the indirect (historicised) imperialism of Treasure Island, the difference between Jim Hawkins’s prior anticipation of the island and his response to it in actuality (and memory) reflects a division in the imperial discourse between that of the ‘naive child’ for whom ‘maps instilled a sense of glory and mighty purpose’ and the ‘cynical adult’ for whom ‘the pristine whiteness of the blank spaces. . . had been darkened . . . by the incursion of European capital’ (33). The events of the book move Jim from the first to the second position. In order to look more closely at issues of accuracy/truthfulness and use value for maps in works of fiction I want to turn to the map for Treasure Island (see Fig 2.12 in Chapter 2). It is worth paying close attention to the description of this map at the point when it first appears within the narrative: The paper had been sealed in several places with a thimble by way of seal; the very thimble, perhaps, that I had found in the captain’s pocket. The doctor opened the seals with great care, and there fell out the map of an island, with latitude and longitude, soundings, names of hills, and bays and inlets, and every particular that would be needed to bring a ship to a safe anchorage upon its shores. It was about nine miles long and five across, shaped, you might say, like a fat dragon standing up, and had two fine land-locked harbours, and a hill in the centre part marked ‘The Spy-glass’. There were several additions of a later date; but, above all, three crosses of red ink – two on the north part of the island, one in the south-west, and, beside this last, in the same red ink, and in a small, neat hand, very different from the captain’s tottery characters, these words: – ‘Bulk of treasure here.’5

The map has three distinct use values as described here, as well as being an aesthetically pleasing object with a strong material presence. Although frequently referred to as a ‘map’, its primary identity is as a chart with a navigational value providing depth readings, latitude and longitude coordinates for the island (withheld from the reader) and information on harbours and safe places to land. It is also geocentric, a topographical map of what is on the island, (its physical features) with a scale supposedly measuring the extent of the island by the familiar ‘English mile’. Thirdly, it is egocentric, a map reappropriated by a particular individual to locate

5

Robert Louis Stevenson, Treasure Island (London: Cassell & Co. Ltd, 1883), 51.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Trusting the Fictional Map

97

something personal on the island (requiring place-specific bearings and marks in a different colour ink). In this respect, it is Captain Flint’s map, with the status almost of a will and a value that exceeds its informational content. This map can thus be used in three ways: to get you to the island; to help you situate yourself once there; to locate a specific, personal (highly significant) point upon it. If we consider the integration of the map in relation to the narrative then we see that these three attributes of the map shape and determine each section. At the start of the book, key events and characters’ actions are driven by the desire to maintain or acquire possession of the map. The story begins in a remote backwater somewhere on the south-west coast of England where the narrator, Jim Hawkins, lives with his parents. However, the quiet privacy and containment of this world is shattered by the arrival first of ‘the old buccaneer’ (Billy Bones), seeking to hide because he has the map, and then of the pirates in search of the map’s current owner. We first hear of its existence within the text in a remarkably indirect way, when Bones tells Jim, amid other ramblings, that: ‘I’m the on’y one as knows the place. He gave it me at Savannah, when he lay a-dying’ (22). In this first phase of its existence the map is both hidden and objectified. It is repeatedly referred to as ‘it’ – emphasising the double worth ascribed to it in terms of its own materiality as well as what it represents. Soon afterwards, as the other pirates ransack the inn in search of the map, their conversation reads as follows: ‘Is it there?’ roared Pew. ‘The money’s there.’ The blind man cursed the money. ‘Flint’s fist, I mean,’ he cried. ‘We don’t see it here nohow,’ returned the man. ‘Here, you below there, is it on Bill?’ cried the blind man again. (38)

The description of the map as ‘Flint’s fist’ is never directly explained, but points to the power of the map acquired by association with its maker.6 Even once it has come into Jim Hawkins’s possession it continues to be described as ‘the thing in my breast pocket’ (43), ‘the thing that they were after’ (47) and, when the bundle containing it is finally opened, as ‘a sealed

6

A fist is made either to strike a blow or to grasp at something, and perhaps both these senses are in play in relation to the map. More straightforwardly (though a meaning no longer current), ‘fist’ could simply be used of a person’s handwriting (OED).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

98

Doubleness and Silence

paper’ (49). The object has an apparently dangerous power, as though it cannot be named directly. The middle section of the narrative shifts emphasis from acquisition of the map as a unique and valuable object, to practical use of it as a chart to reach the island and anchor there. At this stage, carelessness with the map, or with knowledge of it, is particularly dangerous. The Squire’s foolish inability to keep quiet allows the pirates to infiltrate the crew of his hired ship, the Hispaniola, and puts them all in danger. The way in which the map’s different meanings are palimpsestically lain on top of one another, with the egocentric value exceeding the navigational or geographical, is strongly felt when Captain Smollett shows Long John Silver a piece of paper that appears to be the map but is in fact a copy of it, with Flint’s vital information removed: Long John’s eyes burned in his head as he took the chart; but, by the fresh look of the paper, I knew he was doomed to disappointment. This was not the map we found in Billy Bones’s chest, but an accurate copy, complete in all things – names and heights and soundings – with the single exception of the red crosses and the written notes. (95)

Long John’s response makes the distinction between the geocentric and the egocentric map absolutely clear. The pirates already know how to get to the island, where to land upon it and what all its features are called. The identical copy of the ‘map’ is of no value because its value does not lie in neutral geographical information. Again, there is a sense of the map as something which exceeds its informational content, and this is felt once more towards the end of the book when Long John Silver gives the map to his crew, to keep them loyal to him. Jim describes how the appearance of the chart was incredible to the surviving mutineers. They leaped upon it like cats upon a mouse. It went from hand to hand, one tearing it from another; and by the oaths and the cries and the childish laughter with which they accompanied their examination, you would have thought, not only they were fingering the very gold, but were at sea with it, besides, in safety. (243–4)

There appears to be a confusion here between the map’s material and referential values. The pirates assert the map’s authenticity on the basis of signature – ‘that’s Flint, sure enough. J. F., and a score below, with a clove hitch to it; so he done ever’ (344) – but they then take this as an absolute assurance that the map will ‘work’. The pirates are clearly extremely naive map readers, but at the same time the map does seem to operate as if it has a power and ‘truth’ of its own. It functions as a kind of dark talisman,

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Trusting the Fictional Map

99

offering false promises and leading only to mislead. The fictional map seems to possess active agency (as a form of representation that sows confusion across its own coded layers of meaning) in ways that anticipate the more recent findings of critical cartography. The final section of the narrative concerns activities on the island itself, as the two opposing camps (Pirates/Gentlemen) openly battle it out for the treasure. At this point, the map not only changes hands; it also changes sides in a way that mirrors the actions of the two main characters in the book, Silver and Jim. At the same point in the narrative that Jim is unwillingly made to change sides (when he returns to the stockade in the dark and is captured by the pirates), the map is also found to have done so, to the confusion of both characters. Silver asks: ‘“And, talking o’ trouble, why did that doctor give me the chart, Jim?” My face expressed a wonder so unaffected that he saw the needlessness of further questions’ (237). Equally, Silver, who has proved himself entirely duplicitous, uses Jim and the map to save himself and repeatedly switch sides in the unstable endgame of the search for treasure. Once the map is finally put to use as a treasure map by the pirates, of course it turns out to be valueless since the treasure has already been removed from its original location by Ben Gunn. The map thus shows itself to be ‘accurate’ in that it leads the ship to the island and the men to the spot where the treasure was buried, but this ‘truth’ turns out to be hollow. It proves in the end to be as duplicitous as those who fight over it. The map of Treasure Island also seems to possess agency in a way that is highly appropriate when we recall the creative process underpinning this literary work. In his autobiographical essay ‘My First Book’, Stevenson famously gives a full account of the role of the authorial map within the creative process.7 He recalls himself drawing with his stepson: On one of these occasions, I made the map of an island; it was elaborately and (I thought) beautifully coloured; the shape of it took my fancy beyond expression; it contained harbours that pleased me like sonnets; and with the unconsciousness of the predestined, I ticketed my performance, ‘Treasure Island.’. . . as I paused upon my map of ‘Treasure Island,’ the future character of the book began to appear there visibly among imaginary woods; and their brown faces and bright weapons 7

There are two accounts of the map’s origins. Lloyd Osbourne, Stevenson’s stepson, describes how ‘I happened to be tinting the map of an island I had drawn. Stevenson came in as I was finishing it, and with his affectionate interest in everything I was doing, leaned over my shoulder and was soon elaborating the map and naming it’ (Osbourne, ‘Note’, The Works of Robert Louis Stevenson, Tusitala Edition (London: William Heinemann Ltd., 1923), II: xviii). In this account, the boy draws the map and Stevenson only annotates it.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

100

Doubleness and Silence peeped out upon me from unexpected quarters, as they passed to and fro, fighting and hunting treasure, on these few square inches of a flat projection. The next thing I knew I had some papers before me and was writing out a list of chapters.8

Stevenson’s account displays the strongly visual nature of his imagination as well as his heightened awareness of spatial representation. Here the geometric representation brings into being its own referential ‘reality’ in a way that once again draws attention to the fact that all maps do this to some extent (however naturalistic they purport to be).9 The ‘flat projection’ of imagined abstract space is translated directly into written embodiment of it. Later in the same essay Stevenson reinforces this when he describes how ‘I had written it up to the map. The map was the chief part of my plot’ (‘My First Book’, 127). The prior visualisation continues to directly inform and shape the writing of the narrative: For instance, I had called an islet ‘Skeleton Island,’ not knowing what I meant . . . and it was to justify this name that I broke into the gallery of Mr. Poe and stole Flint’s pointer. And in the same way, it was because I had made two harbours that the Hispaniola was sent on her wanderings with Israel Hands. (127–8)

It is interesting to compare this authorial account of the relative priority of map and text with a cartographical critique of Harley’s use of Foucault. In ‘Images of Power’ Barbara Belyea argues that, while Harley’s methods fully contextualise the map, he still treats it essentially as a fixed object capable of holding intentional and unintentional meanings behind it, rather than going further and removing individual agency to reveal the map as intrinsically part of a discourse of power/knowledge beyond the context of the intentions of its maker(s). Belyea makes the point that: ‘Foucault insists that far from passively inscribing a pre-existent object, the énoncé forms that object. Hence maps determine our perception of the capes and bays rather than the reverse.’10 When a literary map forms part of the creative process it also functions in just such a way in relation to the fictional world – asserting power over it rather than being subject to it. Furthermore, this also appears to be true for writer as map-maker – even though that world would not exist at all without his or her imagination. 8

9

10

R. L. Stevenson, ‘My First Book: “Treasure Island”’, Essays in The Art of Writing (London: Chatto & Windus, 1919), 111–34; 118. For a more sceptical reading of Stevenson’s autobiographical account here, see Laura Eidam, ‘Re-examining Illustration’s Role in Treasure Island: Do Images Pirate Texts?’, English Literature in Transition 1880–1920 55.1 (2012): 45–68. She interprets the map as a radical example of ‘proleptic’ illustration. Barbara Belyea, ‘Images of Power: Derrida/Foucault/Harley’, Cartographica 29.2 (1992): 1–9; 6.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Trusting the Fictional Map

101

A second example of a Victorian adventure map will allow further consideration of these issues of accuracy and trustworthiness. King Solomon’s Mines was written in direct response to Treasure Island, so we might initially expect the maps to work in similar ways.11 Both narratives also concern a search for treasure that is constructed as a quest (involving an outward journey into the unknown and a safe return home), and such a structure would seem naturally to require the presence of a map as guide.12 As with Treasure Island, then, the map for King Solomon’s Mines (see Fig. 3.1) performs a vital motivating function in relation to the quest narrative, working as a catalyst to initiate a sequence of events. First, it creates a connection between the major characters, then it motivates the journey which forms the core of the narrative and finally it is actively used to lead the characters across hostile terrain. However, unlike in Stevenson’s book, where everyone is after the same thing (the map as a displaced metonym for the treasure), three separate motives inform this journey from the outset, only one of which is directly linked to the map. The primary narrative concerns Sir Henry Curtis’s search for his lost brother who was last seen heading off in the destination marked on the map. Their search leads them to the old hunter, Quatermain, who possesses a copy of the map. His adventure is primarily motivated by a search for wealth in the form of the diamonds in the mines, but he has possessed the map for years without using it and does not expect to be successful. The third, subsidiary narrative concerns the native Umbopa who volunteers to go with them because he is searching for a lost homeland of which he is the prince. The first motive – the search for a lost brother – dominates the action since Sir Henry is funding the expedition, but all three quests are extremely openended, with none of the participants really believing in the likelihood of success. So, although the map is basically a treasure map, it is not really being used as a treasure map.

11

12

In a biography of her father, Lilias Rider Haggard relates that ‘Travelling up to London with one of his brothers they started discussing Treasure Island, just then making a great success. Rider said he didn’t think it was so very remarkable, whereupon his brother replied, rather indignantly: “Well, I’d like to see you write anything half as good – bet you a bob you can’t.” “Done,” said Rider, and when the day’s work was over, promptly sat down in the dining-room at Gunterstone Road to try’ (Lilias Rider Haggard, The Cloak That I Left: A Biography of the Author Henry Rider Haggard (Ipswich: Boydell Press, 1951), 121–2. D. S. Higgins gives a slightly less Boy’s Own account, from Haggard’s The Days of My Life, in The Great Storyteller (London: Cassell, 1981), 70. For comparison of these texts as quests, see Robert Fraser, Victorian Quest Romance: Stevenson, Haggard, Kipling, and Conan Doyle (Plymouth: Northcote House, 1998).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

102

Doubleness and Silence

Fig 3.1. Map from King Solomon’s Mines

Again, in a way similar to Treasure Island, by far the least successful part of the adventure is the visit to the mines themselves: the final destination to which the map leads and which supposedly determines the need for it. Whilst the map itself remains true and does not trick its users (the mines are still full of diamonds), the men are almost buried alive by the witch doctor, Gagool. The treasure is devalued by this horrific experience, so that it is only by force of habit that Quatermain fills his pockets before he leaves, telling the reader: ‘I can assure you that if you had passed some twentyeight hours with next to nothing to eat and drink in that place, you would not have cared to cumber yourself with diamonds.’13 Meanwhile, the discovery of Curtis’s brother (the ostensible reason for the whole adventure) occurs by chance on the way home at a point when the characters no longer need the map since they follow an alternative 13

Henry Rider Haggard, King Solomon’s Mines (London: Cassell & Co. Ltd, 1885), 278.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Trusting the Fictional Map

103

native route ‘which is clearly a better route than that marked out in old Dom Silvestra’s plan’ (301). Thus, the map is ultimately shown to be redundant in terms of the primary objective of the quest and the book even offers a subtle critique on colonial prejudice. (Europeans rely upon European knowledge even though it is three hundred years old, rather than stopping to ask the locals.) Here, Matthew Edney’s account of an ‘ironic discourse’ (‘Irony of Imperial Mapping’, 32) for the imperial map again comes into play. We have seen how the map in Treasure Island was repeatedly overvalued, with its worth unquestioned until the end of the book. The most striking difference in the map for King Solomon’s Mines is the lack of trust that the characters initially have in it. This means that, whereas in Treasure Island the map is gradually revealed as untrustworthy, in King Solomon’s Mines the opposite is true, as it slowly proves its worth. As in Treasure Island, the map is handed over at the point of death, wrapped in a package, but there is scepticism about the usefulness of the information it contains from the start. This is because it has been returned to Europe, brought back to Africa, and passed down from (European) hand to (European) hand. Naturally, it is assumed to be long out of date. Initially, characters set out with little hope of success, with Quatermain describing theirs as a ‘wild quest’ (44) and pointing to the distant mountain range to declare ‘God knows if we shall ever climb it’ (57). As they get halfway across the desert and are beginning to suffer from lack of water, Quatermain points out that ‘If we can trust to the old Don’s map there should be some about’, but ‘nobody seemed to derive much satisfaction from that remark. It was so evident that no great faith could be put in the map’ (75–6). Nevertheless, stopping points marked on the map do save the characters twice over. On the first occasion the map lets the characters know that there is the possibility of water nearby (‘pan bad water’ (72)) so that they manage to find it. The second time, as the travellers are about to die of cold on the slopes of ‘Sheba’s Breasts’ the map saves them. In an almost exact repetition of the first experience, Quatermain declares ‘If we don’t find it before dark we are dead men’ (86) and, soon afterwards, the cave (mentioned in Da Silvestra’s writing) is spotted. The map then further authenticates itself when the travellers discover the preserved frozen body of its maker in the cave that shelters them under Sheba’s Breast: ‘Who on earth can it be?’ said I. ‘Can’t you guess?’ asked Good. I shook my head. ‘Why, the old Dom, José da Silvestra, of course – who else?’

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

104

Doubleness and Silence ‘Impossible,’ I gasped, ‘he died three hundred years ago.’ . . . ‘Look here,’ he went on, stooping down and picking up a queer shaped bone scraped at the end into a sharp point, ‘here is the ‘cleft bone’ that he used to draw the map with.’ We gazed astonished for a moment, forgetting our own miseries in the extraordinary and, as it seemed to us, semi-miraculous sight. ‘Ay,’ said Sir Henry, ‘and here is where he got his ink from,’ and he pointed to a small wound on the dead man’s left arm. ‘Did ever man see such a thing before?’ (89–90)

The two primary cartographic dimensions of any map – the making of it and the use to which it is put – are usually kept apart, but here the users of the map come face to face with the preserved remains of the map-maker and the desperate physical means by which the map was produced. They are first disbelieving, then ‘astonished’ and even ‘appalled’ (65) by the encounter. The body of the dead map-maker seems to be possessed of even greater authority than the map itself: ‘The thing overpowered us’ (65). Perhaps Quatermain’s reaction is coloured by guilt at his previous distrust of the object that cost its maker so much personal effort and pain to produce. Within the fictional framework of King Solomon’s Mines, the journey makes the map ‘real’ – of active use – just as the map enables the journey to occur. At the same time, the map is also implicitly being used to ground the adventure for the reader outside the text. Just as the narrative is initially far-fetched and hard to believe for the reader, so the map is initially viewed as far-fetched by the characters. In other words, their relationship to the map mirrors ours towards the narrative and the two work together to ‘authenticate’ each other within and beyond the text. If Treasure Island showed the map to be part of a past world, sealed off and no longer accessible to the reader, then the map for King Solomon’s Mines asserts the power of the past to remain true in the present for both character and reader.

Cartographic Silence Having considered the ways in which the meanings of map and text are integrated and the extent to which the fictional map, like all maps, manipulates both characters and readers, I want now to turn to the second cartographic concept under consideration here, that of cartographic silence. In his influential essay ‘Silences and Secrecy’ Brian Harley is concerned with

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Cartographic Silence

105

establishing a theory of cartographic silence in terms of ‘the intentional or unintentional suppression of knowledge in maps’ and in validating the claim that ‘that which is absent from maps is as much a proper field for enquiry as that which is present’.14 This involves a conceptual shift from the prior visual metaphor of blank spaces to the verbal metaphor of ‘silence’. Harley states: I am deliberately insisting on the term silences in the context of maps, rather than the somewhat negative blank spaces of the older literature, for the reason that silence should be seen as an ‘active human performance’. Silence can reveal as much as it conceals. (‘Silences’, 86)

This shift is important for Harley because it allows extension beyond the visual representation of the map itself, as a single object, into the larger context of historically contingent intentions and human acts involved in the making of it. So, instead of understanding maps as ‘neutral’ or ‘value free’ (86) they can be viewed as ‘socially constructed perspectives on the world’ (86). Whilst Harley’s concern is specifically with political silences, he acknowledges that there are many different kinds of silence present at all stages of the map-making process: ‘Silences are contributed by many agents in the map-making process, through the stages of data gathering to those of compilation, editing, drafting, printing, and publication’ 85). Such a comment allows for the extension of this concept to fictional maps and, as we shall see, to fictional maps at different stages of existence. If cartographic silence re-inscribes subjective intentions into the apparently neutral object then it can also work in this way for fictional maps. Indeed, when fictional maps mislead this is often attributable to the dubious motives of the fictional maker of the map and the question of whether or not his or her intentions can be trusted. Equally, though, because the fictional map does not stand alone, but is always presented in an expository way alongside a text, any ‘silences’ on the map have the potential to be ‘filled’ by the text to which the map corresponds. The most notable example of cartographic silence in Treasure Island is the fact that the map is literally and metaphorically ‘two-faced’; the reader is only provided with a visual image of the front of the map but the text makes clear that it contains information on the front and the back. When the map makes its first appearance within the world of the novel, we are

14

J. B. Harley, ‘Silences and Secrecy: The Hidden Agenda of Cartography in Early Modern Europe’ [1988], in The New Nature of Maps, ed. Paul Laxton, intro. J. H. Andrews (Baltimore, MD: Johns Hopkins University Press, 1997), 84–107; 84, 86.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

106

Doubleness and Silence

given a full description of both sides (this quotation follows on from that given earlier): Over on the back the same hand had written this further information: – ‘Tall tree, Spy-glass shoulder, bearing a point to the N. of N. N. E. ‘Skeleton Island E. S. E. and by E. ‘Ten feet. ‘The bar silver is in the north cache; you can find it by the trend of the east hummock, ten fathoms south of the black crag with the face on it. ‘The arms are easy found, in the sand hill, N. point of north inlet cape, bearing E. and a quarter N. ‘J. F.’ (51)

Here, then, initial visualisation to the reader (the map at the front of the book) silently withholds vital information that the later verbal account reveals. The information on the back of the sheet is solely in Flint’s hand and gives detailed directions as to what to do once on the island to recover the treasure. This is, in fact, the most important piece of information that the map contains. We could go even further and note that what is truly ‘not on the map’ is the human price paid for it. In order to bury his cache securely in a place that only he would know, the owner of the map killed six of his own men. Ben Gunn tells Jim that: ‘I were in Flint’s ship when he buried the treasure; he and six along – six strong seamen . . . One fine day up went the signal, and here come Flint by himself in a little boat, and his head done up in a blue scarf. The sun was getting up, and mortal white he looked about the cut-water. But, there he was, you mind, and the six all dead – dead and buried.’ (81–2)

In subsequent searches to find the treasure, one man has been cruelly marooned, and a further seventeen have been killed. Ultimately, then, the map’s uniqueness, and what renders it so valuable, are the acts of betrayal and the literal ‘silencing’ of any other human sources of knowledge of the treasure’s location. More explicitly than for Treasure Island (where the setting in a remote past distorts this aspect) the map for King Solomon’s Mines functions in terms of European knowledge-as-power enabling penetration into unknown territory and the removal of valuable native commodities for personal gain. Since the most direct application of ‘cartographic silence’ to literary studies might be thought to occur in postcolonial studies it is not surprising that this map has received an unusually high level of critical attention. However, when considered closely, it emerges that such attention is not centred upon

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Cartographic Silence

107

a silence on the map but, conversely, in terms of a silence within the text that the map reveals. Anne McClintock opens her monumental work Imperial Leather with an analysis of Rider Haggard’s map, using it to establish the core themes of her study: Haggard’s map assembles in miniature three of the governing themes of Western Imperialism: the transmission of white, male power through control of colonized women; the emergence of a new global order of cultural knowledge; and the imperial command of commodity capital.15

As she also points out, ‘What sets Haggard’s map apart from the scores of treasure maps that emblazon colonial narratives is that his is explicitly sexualised’ (3). King Solomon’s Mines is a very masculine tale. Quatermain, the narrator, openly celebrates the fact that ‘there is not a petticoat in the whole history’ (3), and the map is made by men, for men. However, as Rebecca Stott makes clear (in a paper that pre-dates McClintock): ‘King Solomon’s Mines does have its female body. That body is that of Africa herself, waiting to be explored.’16 The gendered nature of the landscape, particularly the dominant physical feature of ‘Sheba’s Breasts’, is striking. Quatermain’s description of the mountains within the text is emphatically sexual: These mountains placed thus, like the pillars of a gigantic gateway, are shaped after the fashion of a woman’s breasts . . . Their bases swelled gently up from the plain, looking, at that distance perfectly round and smooth; and on the top of each was a vast round hillock covered with snow, exactly corresponding to the nipple on the female breast. (77)

In her work on Rider Haggard’s African landscapes, Lindy Stiebel emphasises that ‘Haggard did not consciously intend to create these highly sexualised landscapes’ and that ‘his readers seemed by and large to be unaware of what today appears as obvious landscape sexualisation’ .17 The map also strongly partakes of this unconscious sexuality. Firstly, it adopts a dominant male (and imperialist) perspective in its use of an aerial viewpoint. Stott comments of such perspectives that they: operate as part of an imperialist structure: a group of explorers look down onto the land that they are about to enter. The land stretches out before them invitingly . . . 15 16

17

McClintock, Imperial Leather, 1, 3. Rebecca Stott, ‘The Dark Continent: Africa as Female Body in Haggard’s Adventure Fiction’, Feminist Review 32 (1989): 69–89; 77. Lindy Stiebel, Imagining Africa: Landscape in H. Rider Haggard’s African Romances (Westport, CT: Greenwood, 2001), 80.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

108

Doubleness and Silence

such contemplation involves temptation: the imperialist temptation of the desire to possess and conquer. (78–9)

The travellers gaze down on the land to be penetrated, with obvious sexual implications: ‘the imperialist gaze (the aerial view in Rider Haggard) is also the gaze of the voyeur’ (84). Secondly, these critics interpreting the map have pointed out that, in a quite extraordinary way, the map symbolises the female form. So, McClintock (and others) state that ‘if the map is inverted it reveals at once the diagram of a female body’ (3) with Sheba’s breasts leading to triangular caves below. The fantasy, as Stott points out, is of Africa as a passive female waiting to be possessed, but also bound up with the potential danger of the powerful female taking control herself (as when the men almost become buried alive in those caves): ‘The fear is of possible incorporation (absorption into the body)’ (87). When read in this way, it is clear that the map does unconsciously represent the repressed sexual anxieties of the novel. Thus, the map as a masculine product of a masculine form provides a classic example of the ways in which latent silences within the text can be given voice through the more blatant visualisation of the map. Visual expression unconsciously gives voice to what the verbal text withholds. Cartographic silence, then, would appear to work in both directions for literary maps. If the ideological silence of an actual historical map is concerned with ways in which real places in the world can be misrepresented in the interests of one party or another, the silences of the literary map concern both the deliberate withholding of knowledge from character, or reader, and the unintentional revealing of slippages between text and map, between what is intended and what is actually communicated.

The Authenticity of the Fictional Map Finally, we can turn to the third cartographic concept under consideration here and the question of authenticity in relation to the fictional map. What constitutes ‘authenticity’ for a map? Historically, prior to the invention of printing, medieval maps were entirely unique, hand-made objects, as were the portolan charts discussed in Chapter 2 which often included signature, date and place of making. In such cases, there was a single unique original. With the introduction of printing, printed maps from woodcuts also often contained an imprint with information about the date and place of publication and the name of the printer, partially enabling authentication to

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

The Authenticity of the Fictional Map

109

occur. However, this practice was open to abuse, so that many sixteenthcentury mapmakers ‘included imprints on their maps, many of which were copies of other men’s work’ (Wallis and Robinson, 251).18 As a result, authenticity for a geographical map developed in terms of provenance and proof of historicity and consists of determining that the map does correspond to the date at which it purports to have been made. It is established by considering a number of the map’s attributes: lineage in relation to other source maps; content; hand and inscription or nature of printing; paper; ink; physical condition.19 Maps also project authority and authenticity through their formal elements (scale bars; compass points; frames etc.). However, early maps can be reprinted from an original plate, or an uncoloured map can be coloured in the remote past, and such items can still be considered valuable antiques. If we turn from actual maps in the world to the fictional map, then it rapidly becomes clear that two different levels of authentication have to be in play because of the primary doubling created by having one map that faces outwards to the reader, and another that is represented as being used to negotiate place by characters within the narrative. Within the fictional world, as we have seen, the map has to authenticate itself in order to be used by the characters. So for example, in King Solomon’s Mines a detailed history of the map’s line of descent is provided and Quatermain reassures Sir Henry that the copy he has shown him will be authenticated: ‘You shall see the original map and writing when we reach Durban’ (23). Beyond the fictional world, in the map that is presented to the reader, authenticity is reinforced by means of internal visual evidence: the appearance and style of the map needs to confirm the account of its line of descent within the text. Thus, it seeks to authenticate itself as a map and as a mimetic object connecting the reading and use of the map by the reader to the reading and use of it in the narrative by the characters. Frequently, the map that the readers look at purports to be a copy, but the presence of a supposedly authentic original within the text that we do not see, and have no access to, works paradoxically. On the one hand it imbues this object (the Ur-Map) with power and authority and a desire to 18 19

Wallis and Robinson, Cartographical Innovations, 251. For an extended debate over map authenticity (from which these features are largely derived) see The Geographical Journal 140.2 (1974) which reproduces a series of papers from a symposium on The Vinland Map. Defending it (in ‘The Matter of Authenticity’, The Geographical Journal 140.2 (1974): 191–4), G. D. Painter comments that: ‘Its genuineness rests on the interlocking consensus of all its features, both physical and intellectual, the intricate regularity with which all its details fit their historical context’ (191).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

110

Doubleness and Silence

attain it. On the other it draws attention to the absoluteness of its own nonexistence. Duplication within the narrative thus works in a deeply ‘twofaced’ way. It encourages the readers to assume that they are seeing what the characters see and thus ‘sharing’ the map with them, in an imaginative extension of the use of the inner map to the outer, but the fact that we can only ever look upon a ‘copy’ ultimately reminds us that there can be no ‘original’ (since the map is only ever a symbolic fictional representation of a fictional space). Relevant here is Walter Benjamin’s famous essay ‘The Work of Art in the Age of Mechanical Reproduction’.20 In this piece Benjamin points out that works of art have always been able to be replicated but, in previous eras, this did not undermine the value and status of the original: ‘Even the most perfect reproduction of a work of art is lacking in one element: its presence in time and space, its unique existence at the place where it happens to be’ (214). By contrast, ‘that which withers in the age of mechanical reproduction is the aura of the work of art’ (215). A similar sense of ‘aura’ surrounds the ideal map of Victorian adventure fiction. The way the map is treated, both at a dramatic level by characters within the text, and in terms of its material presentation within the book and in the eyes of the author, strongly resembles the account of the ritualistic cult object by Benjamin. But, the literary map is really playing its own game of origins, presenting itself as if there is such an original when in fact it is lacking in exactly these elements of authenticity – a ‘presence in time and space’.21 Thus the map reminds us that the literary work of which it is a part also shares such anxieties over originality and the question of what is lost or gained in endless reproduction. In this way text and map are self-protectively linked, as the map makes explicit an anxiety that concerns the larger literary work as well. When we start to consider this aspect of the map (its fear of erasure leading to duplication) we find that it extends from duplication within the text out into to the real world beyond, at least in both of the examples considered here. For the first edition of King Solomon’s Mines, two versions of the map were given: the sketch map presented within the text, and a fold-out map at the front of the volume. The latter was created by Aggie Barber, Rider

20

21

Walter Benjamin, ‘The Work of Art in the Age of Mechanical Reproduction’, in Illuminations, ed. Hannah Arendt, trans. Harry Zohn (London: Collins/Fontana, 1973), 211–35. Benjamin states ‘For aura is tied to his presence; there can be no replica of it’ (‘The Work of Art’, 223).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

The Authenticity of the Fictional Map

111

Fig. 3.2. Fold-out map from King Solomon’s Mines (For the colour version, please refer to the plate section.)

Haggard’s close intellectual companion (and later sister-in-law).22 A photograph of this material ‘replica’ of the map was presented at the front of the 1885 edition (see Fig. 3.2). Here then, Rider Haggard attempts to 22

Higgins, The Great Storyteller, 71. Lindy Stiebel also gives a good account of the map’s origins: ‘Together with his sister-in-law he physically made a map for King Solomon’s Mines, artificially aged it and had it inked’ (Imagining Africa, 99).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

112

Doubleness and Silence

overcome the representational problem of a fictional map for a world that does not exist, by creating a material object that exists outside the text and can only be reproduced within it. Thus, the map is duplicated within and outside the narrative. Ironically enough, although the sketch map within the text is always given in a modern edition, the photographic reproduction is not – so that fear of erasure by the map ‘original’ is proven justified in this instance. Approaching the fictional map in this way also sheds new light on a minor spat between Stevenson and his publishers over the map for Treasure Island. In ‘My First Book’ Stevenson appears to relate how he sent his original map to the publisher only for it to be lost. He recounts: The time came when it was decided to republish [the serialised story], and I sent in my manuscript, and the map along with it, to Messrs. Cassell. The proofs came, they were corrected, but I heard nothing of the map. I wrote and asked; was told it had never been received, and sat aghast. It is one thing to draw a map at random, set a scale in one corner of it at a venture, and write up a story to the measurements. It is quite another to have to examine a whole book, make an inventory of all the allusions contained in it, and with a pair of compasses, painfully design a map to suit the data. I did it; and the map was drawn again in my father’s office . . . my father himself brought into service a knack he had of various writing and elaborately forged the signature of Captain Flint, and the sailing directions of Billy Bones. But somehow it was never Treasure Island to me. (128)

It has already been noted that the account of the map’s private origins acquired two slightly conflicting versions, and now so does the account of its first public appearance. Stevenson’s account strongly implies that what he sent to the publishers was his own authorial original of the map but, because they lost it, an alternative had to be made (‘the map was drawn again in my father’s office’). However, such an account does not tally with evidence in the Letters. There is some confusion around maps, but it is absolutely clear that his father’s office made the map for publication from the authorial original with the author’s full approval and with the expectation that this would be the version to appear when the work was published in book form. In a letter of 25 August 1883, Stevenson writes: I think the chart should be got underway. If it is not already done add a few more soundings about the mouth of Capt. Kidd’s anchorage and round the south-eastern corner of the Island.23

23

The Letters of Robert Louis Stevenson, ed. Bradford A. Booth and Ernest Mehew (New Haven, CT, and London: Yale University Press, 1994), 4: 148.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

The Authenticity of the Fictional Map

113

Then, again, in a letter of 30 August, Stevenson writes ‘I have just sent off the title sheet of Treasure Island, which only waits the map: marche-t-elle?’ (Letters, 4: 149). In September, this first map made for publication was mistakenly sent back to Stevenson who sent it on to his publishers with a note: ‘P.S. Pray preserve for me the original and so do me a great favour’ (Letters, 4: 151). At this point a map is lost, but Stevenson’s letter makes it absolutely clear that the lost map is the one made for publication in his father’s office: ‘They have never acknowledged the second map’ (Letters, 4: 164) – the publishers already have the original. That authorial original is safely returned two months later: ‘Many thanks for the map come duly to hand’ (Letters, 4: 204). None of this matters hugely, but it is of interest in relation to authorial myth-making.24 Stevenson here appears to create a narrative of a literally ‘lost original’ in which the publishers are to blame, but this serves really to disguise and protect the authorial original (which presumably has survived but was never intended to be made public). The preciousness of the private map is felt in Stevenson’s disingenuous final comment about the published version that ‘somehow it was never Treasure Island to me’.25 The withholding of the authorial map, used as part of the creative process, from the reader represents a final silent assertion of literary cartographic power over the only possible ‘original’ that the fictional map can be. In a way that an actual map cannot, and does not need to, the fictional map multiplies and duplicates itself as if in an endless attempt to escape its own fictionality. Thus the map fatally becomes the agent of its own undoing and potentially that of literature itself, or at least of spatial meaning within the literary text. The more we try to ‘get at’ the place within the text that the map represents, the more we are forced to acknowledge the impossibility of doing so.26 Although the fictional map exists to further convince the reader of the ‘real’ nature of the fictional world and involve him or her in it, ultimately it works in a precisely opposite way. The literary map is always a lost original, one that knows it does not exist but refuses to accept it.

24

25

26

Laura Eidam critiques Stevenson’s account of the map’s origins in ‘My First Book’, but accepts the account of the map’s loss at face value (‘Re-examining Illustration’s Role in Treasure Island ’, 55). Christian Jacob briefly discusses the loss of the map (The Sovereign Map, 283). Of course, if Lloyd Osbourne’s account of the original map is correct (rather than Stevenson’s) then there is no single authorial original – since he drew the island and Stevenson named it. See further discussion of referential inversion between world and object and of map as simulation at the start of Chapter 7.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

114

Doubleness and Silence

From Adventure Story to Spy Fiction A succinct examination of maps in early spy fiction in the final sections of this chapter acts as a bridge between Chapters 3 and 4. Although this genre is more commonly defined in relation to detective fiction rather than adventure fiction, early models seem to me to strongly partake of both genres, particularly in terms of spatial aspects of meaning and relationships between acts of mapping and narration. In a helpful attempt to define the genre, David Seed describes ‘Spy Fiction’ as: ‘a close but distinct variation on the tale of detection with the difference that there is no discrete crime involved but rather a covert action which . . . transgresses conventional, moral, or legal boundaries’.27 The key phrase here is ‘covert action’: the genre is strongly identified by the need for doubleness, for masks, deception and ‘cover stories’ that conceal the truth, often at multiple levels. Much early spy fiction plays with the question of whether or not a crime has been committed and this is certainly true for the first novel in the new genre, considered below – The Riddle of the Sands – and for many of the works of another key author in relation to spying and maps, John Buchan. In Spy Thrillers from Buchan to Le Carré, Clive Bloom describes how ‘the spy genre like the world it depicts is a form attempting to exist in disguise’.28 More importantly, for the purposes of locating it here, he also asserts that: ‘The formal origins of the spy genre lay hazily within an amalgamation of the imperial adventure tale and the detective novel’ (1). John Buchan’s work (with maps in six of his novels) exemplifies the doubled generic origins.29 This sense of crossover between spy and adventure fiction is strongly felt in Buchan’s first bestseller, Prester John (1910). Like Rider Haggard, Buchan had been in South Africa and worked for the high commissioner immediately after the second Boer War (1902). His novel draws directly upon his experience, being set in Africa and involving a native uprising foiled by the ingenuity and determination of the narrator. In Prester John the map of ‘The Various Journeys of Mr David Crawfurd’ itself also bridges the two genres of adventure and spy fiction. On the one hand, it works in an imperialistic way to locate the British hero, Crawfurd, within a hostile 27

28 29

David Seed, ‘Spy Fiction’, in The Cambridge Companion to Spy Fiction, ed. Martin Priestman (Cambridge: Cambridge University Press, 2003), 115–34; 115. Clive Bloom, Spy Thrillers from Buchan to Le Carré (Basingstoke: Palgrave Macmillan, 1990), 1. Maps are provided for: Prester John (1910); Huntingtower (1922); The Three Hostages (1924); The Dancing Floor (1926); The Courts of the Morning (1929); The Island of Sheep (1936). Buchan also used maps of the battlefields in his work for the War Office, which resulted in Nelson’s History of the War in 24 volumes with numerous maps and plans.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

From Adventure Story to Spy Fiction

115

Fig. 3.3. Map from Prester John (1910)

colonial landscape (see Fig. 3.3). The kraal (mountains) that run down the left-hand side are associated with safety, order and British control, whilst the plains below (within which most of the action takes place) are held within the latent power of the natives: ‘Once on the plateau I would be inside the white man’s lines. Down here in the plains I was in the country of my enemies’.30 The map thus comes into full use for the central ‘adventure’ action of the novel, on a colonial backdrop, but the focus of action is centred upon the highly politicised and racialised geography that turns Crawfurd into an eavesdropper and covert agent (for the Europeans against the natives) on whose ability to escape capture everything depends. During the course of the novel, native power comes to fruition under the leadership of the charismatic ‘Laputa’ who draws upon the legend of Prester John to motivate his followers. The map thus ultimately works to trace the (colonial) hero’s actions in thwarting the (native) rebellion by acting rapidly and alone on the basis of acquired information. 30

John Buchan, Prester John (London: Thomas Nelson & Sons Ltd, 1910), 226.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

116

Doubleness and Silence

One reason that the spy thriller lends itself so strongly to the presentation of a map is because of the causal relationship between hero and plot, between personal and political, which creates a structure in which a single individual has the capacity to intervene and disrupt events, with potentially massive consequences for great numbers of others. Buchan actively uses maps to link spatial meaning to the interplay between personal actions and larger consequences. He does this primarily through a recurring element of pursuit that characterises his novels and comes to a climax in a particular location, employed as a theatre on which dramatic and intense events can be played out. Buchan’s heroes are frequently placed in this position and directly articulate their feelings about it. So, David Crawfurd in Prester John describes how I had a terrible feeling that I alone could do something to ward it off, and just what that something was I could not tell. I was horribly afraid, not only of an unknown death, but of my impotence to play any manly part. (109; my italics)

Where the hero is reluctant, amateur and accidentally pulled into action (as is often the case in Buchan’s novels), this effect is exacerbated. A personalised hunt at one level thus also has national or universal importance at another (in a way that develops out of but is not true for the adventure story for boys), and this comes to a head at a particular site (which the map represents).

Integrative Cartography in The Riddle of the Sands We need to turn back from Buchan to Erskine Childers, his predecessor, who anticipates him in various ways in his singular work, The Riddle of the Sands. This classic work of ‘invasion-scare’ literature offers a model of literary mapping that is unique in the extraordinary level of integration of literature and cartography that it presents. In an article on ‘The Birth of the British Spy Novel’, David Stafford makes a direct connection between the larger context of anxiety and the emergence of a new literary form: Awareness of these revolutionary changes and concern with national vulnerability, invasion, and hostile espionage made their mark on the popular literature of these years . . . It is here that the roots of the spy story are to be found. The invasion scares had given rise to a whole genre of fiction depicting some imaginary invasion of Britain.31 31

David A. T. Stafford, ‘Spies and Gentlemen: The Birth of the British Spy Novel, 1893–1914’, Victorian Studies 24.4 (1981): 489–509; 496.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Integrative Cartography in The Riddle of the Sands

117

At the start of the twentieth century, Britain was feeling isolated as a result of the second Boer War, German nationalism was on the rise and there was a sense of shifting power in Europe. From 1898 onwards, a series of German naval bills had enabled the creation of a powerful navy and by spring 1902, ‘the British Admiralty was advised by their naval attaché in Berlin that the rapidly growing German High Seas Fleet was of such a design that it could only be intended for war against Britain’.32 As a result, ‘The Admiralty secretly began to plan a naval base at Rosyth as part of their reply to that German threat’ (73). The larger mood of the time was a desire for the government to act. Leonard Piper tells us: In March, just before the publication of the book, a Bill was placed before Parliament to do precisely that. Fact was rapidly catching up with fiction. Erskine added a postscript to the book acknowledging the measures being taken by the government. (74)

The Riddle of the Sands thus has a remarkable and unique relationship to the nuances of its own historical period which also causes its identity to shift from being a fictional work of warning and persuasion to becoming a propaganda tool in the hands of the government. The book gains much of its tension from a specific historical state of mind: that of uncertainty as to the enemy’s intentions, which may themselves not yet be definite. Piper describes an atmosphere of ‘mutual spying and the drawing up of contingency plans’ (139) from the turn of the century onwards. Such a mood bears directly upon the atmosphere of The Riddle of the Sands and gives it much of its power as the first spy thriller. There is thus a doubled temporal urgency built into this work of literary mapping: the heightened value of mapping in periods of tension between nations, and the value of making these maps relating to this coastline at exactly this time. In How to Lie with Maps, Mark Monmonier considers why governments guard maps, leak false information and even deliberately falsify geography through acts of ‘cartographic disinformation’ (118). He reminds us that ‘mapped information often must be guarded. If knowledge is power, an enemy’s knowledge of your weaknesses and strengths is a threat’ (113). As a result: ‘What is mapped, as well as the maps themselves, must be kept confidential, for to reveal an interest in a particular area or features is to reveal one’s plans’ (115). This is exactly the scenario for The Riddle of the 32

Leonard Piper, Dangerous Waters: The Life and Death of Erskine Childers (Hambledon: Continuum, 2003), 73.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

118

Doubleness and Silence

Sands. Carruthers notes Davies’s annoyance at the state of the English charts: ‘I suppose those waters are only used by small local craft,’ I put in; ‘that would account for inaccuracies.’ Did Davies think that Admiralties had time to waste on smoothing the road for such quixotic little craft as his, in all its inquisitive ramblings? But he fired up. ‘That’s all very well,’ he said, ‘but think what folly it is.’33

In contrast to the British government, Davies has been spending his time creating his own accurate maps of a region that presents particular challenges of navigation because of the sands at low tide and the need to know where the deeper water lies: ‘“The charts were shocking, but I worried out most of the channels”’ (40). In wartime, or under threat of invasion, detailed cartographic knowledge of a region of strategic significance takes on a value that it would not ordinarily hold and simple factual information about communication, transport, rivers and tides becomes loaded with meaning. Davies thus puts himself at risk simply by the act of trying to make an accurate map. In the larger context of fictional mapping, the book is highly unusual in three ways: in being a realist novel that also contains maps; in containing maps that are historically accurate and of their time; in containing multiple maps (four at the front as well as a sketch map within the text). The sense of the novel as somehow operating beyond ordinary fictional bounds or on the boundary between fact, fiction and future reality extends into the narration and frame, as well as bearing upon the nature of the maps. The semi-fictional narrator, who shares the same initials as the author, is presented as ‘editing’ material provided by first-hand accounts from two men who urgently want their story to be told and acted upon. The elaborate device of a reconstructive editor mainly exists in order to allow different written and visual records of the trip (diary, log, map, chart, oral account) to be pulled together. Other critics have described such a combination of fictional documents as a ‘doceme’ and considered that this kind of structure occurs quite frequently in relation to the fictional map (Utopia, for example, works in a similar way).34 The verisimilitude of the gathered notes and materials enables the maps to be directly referred to throughout,

33

34

Erskine Childers, The Riddle of the Sands: A Record of Secret Service Recently Achieved (London: Smith, Elder & Co., 1903), 41. See Ekman, Here Be Dragons, 22.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Integrative Cartography in The Riddle of the Sands

119

without undermining the realism of the novel. So, a note in the first edition tells us that: The fragments of charts shown at pages 69 and 219 are reproductions, on a slightly reduced scale, and omitting some confusing and irrelevant details, of British and German Admiralty Charts. Space precludes the insertion of those bulky engravings in full; but the reader who wishes for fuller information is referred to Charts Nos 406 and 407 of the British series and to No. 64 of the German series. (vii)

The maps present accurate coastal outlines with real place names and depth soundings on the charts and are far more realist in appearance than most literary maps. At the same time, though, they also contain literary elements that strongly connect them to the narrative, allowing the reader to mimetically follow the route of the yacht or locate key events. The two thus work together to reinforce each other, as Matthew Graves makes clear: Verisimilitude is heightened by the narrative’s inscription in a referential geography which the novel’s Maps and Charts validate graphically. The maps’ fixed coordinates and political borderlines serve to anchor the narrative in the real.35

In the popular paperback version of the text, the reader is presented with two pairs of map and chart at the front (Map A/Chart A; Map B/Chart B) in which each pair works together. However, in the first edition the ‘General Map’ is given as a fold-out frontispiece map at the start, with the other maps integrated within the pages of the book (see Fig. 3.4).36 This integrates the maps far more with the points within the text at which they are to be used (encouraging active reading and mapping on the part of the reader) as well as giving the General Map higher status.37 The novel follows the adventures of two friends, Davies and Carruthers, sailing along the Baltic and Frisian coastlines at the end of summer. An apparently innocent holiday invitation from Davies to sail with him on his boat, the Dulcibella, turns into something far more sinister as Davies slowly reveals prior events to Carruthers and they both become involved in spying upon a supposed German, Dollman, (who is in fact British) and uncovering German military plans for the region. The General Map sets the scene for 35

36

37

Matthew Graves, ‘Maps and Texts – The Case of The Riddle of the Sands’, Trans 16.9.6 (July 2006), Internet-Zeitschrift für Kulturwissenschaften (Knowledge Networking in Cultural Studies): www.inst.at/trans/16Nr/09_6/graves16.htm. In the first edition, the ‘Chart to illustrate the stranding of the Dulcibella’ is given opposite p. 69; the ‘Map of East Friesland’ opposite p. 141; and the ‘Chart of Juist, Memmert and parts of Nordeney’ opposite p. 219. As such, this is also an example of the vulnerability of the fictional map over time, with the requirements of paperback publication bearing upon paratextual position and meaning.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

120

Doubleness and Silence

Fig. 3.4. ‘General Map’ from The Riddle of the Sands

the entire book, but also shows the proximity of England to the area of Holland and Germany around which the two men are sailing and so relates to the novel’s larger defensive purpose. The second map situates the larger narrative in relation to a key event that has occurred before Carruthers arrived – the ‘stranding of the Dulcibella’ (see Fig. 3.5). In a way characteristic of the fictional map’s expository nature, it is directly linked to the narrative by the presence of key events described on the map itself (e.g. ‘Point where Medusa hove to and the short cut was decided upon’) but, equally, those events are able to be mapped because they involve highly skilled spatial manoeuvres at sea. In order to use these maps effectively, the reader is required to interpret symbols and words carefully on the map as well as to move actively between two scales of map and text. This is quite demanding, as one contemporary reviewer in the Times Literary Supplement noted disparagingly: ‘The book can only be read by the aid of large maps, railway guides and special information about things nautical’.38 The next map (not given here) is a land map of the Frisian Islands with the railway lines also marked on the East Friesland mainland that relate directly to Carruthers’s final train journey around the region (as he pretends to return home) during which the solution to the mystery is found. The second chart (also not reproduced) is a detailed one of three of the islands and presents a complex double layer of literary map information.

38

Piper, Dangerous Waters, 75.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Integrative Cartography in The Riddle of the Sands

121

Fig. 3.5. ‘Chart Illustrating the Stranding of the Dulcibella’ from The Riddle of the Sands

There are depth soundings and detailed markings of routes across the sands but it is also provided explicitly ‘to illustrate Chapter xxi, xxii and xxvii’ (and in the first edition is given within the text a few pages into chapter xxi, opposite p. 219). It has marked upon it a long and exciting boat row through fog from Norderney harbour to Memmert and back, made by Carruthers and Davies in their dinghy, to spy on the four Germans. This links it to the final map given within the book which is Carruthers’s handdrawn sketch map made during that trip as ‘a rough sketch of the scene, as I partly saw and chiefly imagined it’ (233). This provides spatial understanding at a different scale for his cautious movements whilst spying on the Germans in the fog (see Fig. 3.6). In a stimulating article on ‘Maps and Texts’ in this novel, Matthew Graves describes the effect of this series of maps as inviting the reader to ‘envision the novel as a triptych of geographies unfolding along an east– west axis’ (6) but he also goes further, claiming that: We might thus envisage the map’s function in The Riddle of the Sands as that of a chronotopic surface, organising the relations between time and space (and by extension character) in the narrative. (7)

The maps certainly have a strongly performative function for the reader, to which the ‘editor’ draws our attention. At the same time, they have a primary function within the narrative of enabling navigation on the voyage

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

122

Doubleness and Silence

Fig. 3.6. ‘Sketch Map for Journey in the Fog’ from The Riddle of the Sands

for Davies and Carruthers and allowing negotiation of specific routes across the sands during the adventure. Continuous, active use of the maps within the narrative then strongly encourages reconstruction of routes for both characters (retrospectively revisiting earlier intense experiences) within, and readers outside, the novel. The verisimilitude of the charts encourages the reader to assume that these correspond to what Davies is using on the yacht and thus to engage in a visual reconstruction of the detailed routes and events described within the book. Carruthers stands in for the reader: a novice struggling to make sense of complex cartographic information, but recognising the importance of doing so. At times he also explicitly asks for past narratives to be told in relation to the charts, thus actively connecting the acts of mapping, narrating and reading. So, when he finally persuades Davies to explain what happened with Dollman, he requires the maps to be present, and consulted, alongside the retelling of the narrative: ‘Wait a bit, let’s have the chart,’ I interrupted. (65) ... ‘Here it is,’ said Davies (see Frontispiece) and I looked with a new and strange interest at the long string of slender islands. (65)

The factual geography of the area looked at in cold objectivity is re-nterpreted in the light of the dramatic events that took place off those islands, imbuing it with human meaning. The maps and charts work in exactly the same way for the reader, in relation to the story, as the narrative brings the meaning of the maps to life. We are not, after all, exploring the Frisian Islands in a boat, so

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Double Meanings; Double Intentions

123

that the chart has no practical purpose or value (as it supposedly does for the characters). Only when it starts to be loaded with actual and potential human meanings does it become of more than background interest. This is of course true for all literary maps, but The Riddle of the Sands enacts this situation knowingly through Carruthers’s response.

Double Meanings; Double Intentions What the atmosphere of ‘mutual spying’ creates in this singular novel is a work for which all codes – the visual codes of the map, the hermeneutic and proairetic codes of the text, the unifying symbolic meanings within the story and the external cultural context in which it is written – become capable of bearing a double meaning. Moreover, as in the earlier examples from Victorian fiction, the maps themselves are not only doubled but also deliberately duplicated by the characters. The ‘authentic’ maps (the hidden originals) are valuable twice over: positively for the cartographic information they encode, and negatively in terms of what their very existence signifies if they are discovered (that the two men are not what they seem). In How to Lie with Maps, Monmonier describes the Soviet Union’s ‘systematic falsification of geographic location’ (118) which must result in two sets of maps, one (inaccurate) to be presented to the outside world in order to mislead them, and a second (accurate) to be held as secret information. He comments that: ‘Maintaining a second, secure set of accurate maps is expensive’ (118). In a comparable way, Davies and Carruthers also adopt a procedure of double mapping: We had two sets of charts, German and English. The former we decided to use in practice, and to hide, together with the log, if occasion demanded. My diary, I resolved, should never leave my person. (Riddle of the Sands, 140)

The English charts are already inaccurate and thus of little use. Their presence on the shelf suggests that they are only relatively amateur sailors. By contrast, the accurate German charts, made even more accurate by Davies’s annotations, are to be hidden. Of course, even if found, they only reveal that the two men have been mapping the area, not that they are necessarily ‘spying on a spy’ (as they are in fact doing). But as we have seen, accurate mapping becomes a danger in itself in any pre-war period.39 39

Ironically what ultimately ‘gives them away’ is an object on the shelf, but not the charts. This is a book written by a younger Dollman that proves that he is British (and which they do not even realise is important until it is recognised by his daughter).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

124

Doubleness and Silence

The extraordinary power and distinctiveness of The Riddle of the Sands lie in its hermeneutic play, which ties together all forms of coded meaning, visual and verbal, into the solving of a ‘Riddle’. This riddle is as much spatial as verbal, and has the further complication that it may not actually exist at all. It is also played out at every level within the novel: plot, narrative and character. Actions bear double motivations that only those who undertake them can truly know, because intentions are masked. Characters appear to be one kind of person but may turn out to be another. (Dollman presents himself as German but is really English; Dollman’s daughter may or may not be innocent; Carruthers and Davies increasingly cease to be what they at first appeared to be.) Conversations become games based upon mutual uncertainty because both sides have the potential to be, or not be, what they seem. What, then, is the object of all this double-meaning? For The Riddle of the Sands, the solution to the ‘Riddle’ is essentially a cartographic (or hydrographic) one. Ultimately, what Carruthers and Davies are trying to determine is why this particular geographical area is of value to the Germans and what is motivating their attempts to protect knowledge of it. As we might by now have come to expect, they come up with two theories. Davies’s hypothesis concerns the natural harbour that is created by the line of islands: ‘The space behind them is like an immense tidal harbour, thirty miles by five, and they screen it impenetrably. It’s absolutely made for shallow war-boats under skilled pilotage. They can nip in and out of the gaps, and dodge about from end to end . . . It’s a perfect base for torpedo-craft.’ (143)

For him, the value of the region lies in its strategic importance as Germany’s only coastline, from which he concludes that there must be some defensive secret attached to the channels, which allow rapid access to the German mainland. For Carruthers, however, this theory ‘doesn’t supply sufficient motive’ (105). He therefore interprets the mystery differently, deciding that the key lies in a particular place, Memmert, where there is a cover venture of dredging up buried treasure (in which all the conspirators are involved) which must hide a second activity. Clues provided by overhearing German conversation further lead the two men to try and identify the significance of seven geographical items, which could concern islands, villages or canals in the region.40

40

The kind of clue given here – the knowledge of a number without knowing what it refers to – also strongly anticipates John Buchan’s The Thirty-Nine Steps (as does the ambiguity over whether the villains are villains in the novel and the protagonist’s uncertainty in reading them).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

Double Meanings; Double Intentions

125

As it turns out, however, both of them have entirely misinterpreted, not the map or the geography of the area but the reason it is of wartime value. It is only when Carruthers has temporarily left the region, pretending to return to England, and when he compares a larger-scale Ordnance Survey land map with the sailing charts, that he is able to finally reach the correct interpretation of the Germans’ actions: At the bookstall at Emden I bought a pocket ordnance map of Friesland, on a much larger scale than anything I had used before . . . Charts are apt to ignore the geography of the mainland, except in so far as it offers sea-marks to mariners. On the chart this stream had been shown as a rough little corkscrew, like a suckingpig’s tail. On the ordnance map it was marked with a dark blue line, was labelled ‘Benser Tief’, and was given a more resolute course . . . For the first time that day there came to me a sense of genuine inspiration. Those shallow depths and short distances, fractions of metres and kilometres, which I had overheard from Böhme’s lips at Memmert, and which Davies had attributed to the outside channels – did they refer to a canal? (283, 285)

Crucially, it is the difference between two kinds of mapped representation of the same area – the land map and the sea charts they have been using – that enables Carruthers to understand the spatial mystery. He realises that what looked like insignificant narrow land waterways on the charts might in fact be man-made channels down which a troop boat could pass, drawing upon rail communications in the region. Rather than a German scheme of coastal defence, the region is in fact to be used to launch an invasion of the south of England ‘in seven ordered fleets . . . from seven shallow outlets’ (309). As Davies notes, his reading of the natural features and their importance was correct, but not his assumption as to what use they might be put to: ‘I was right – only upside down,’ he murmured more than once. ‘Always really right – those channels are the key to the whole concern’. (318)

This of course renders their information even more valuable and important for the British government within the fictional narrative whilst, in real life, as the frame narrative reminds us in the epilogue: [O]ur primary purpose is to reach everyone; and there may be many who, in spite of able and authoritative warnings . . . are still prone to treat the German danger as an idle ‘bogey’. (326)

At a level of content and action, The Riddle of the Sands offers a remarkable example of the power of maps and the way in which, in wartime, they rapidly cease to be innocent and turn into dangerous objects

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

126

Doubleness and Silence

to possess. Nevertheless, from a readerly point of view, all the information necessary to follow the narrative and solve the riddle is presented on the maps given at the front of the book, yet the crucial larger-scale map that actually reveals the truth to Carruthers is not provided. A practical explanation is offered for this in terms of scale, but the effect is to encourage the readers to adopt the same blind spot as the obsessive sea-lover Davies. The maps and charts privilege the margins of the land over the land itself, which is where the solution actually lies.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.004

4

Mapping Murder

‘Such duplicity is terrible,’ said M. Bouc, ‘But it seems to please you.’ Agatha Christie, Murder on the Orient Express [A]fter twenty chapters stinking with red herrings, it turns out that the man with the gun did it after all. Dorothy L. Sayers, Five Red Herrings

This second genre chapter is centred upon another, newly emerging, popular genre in the mid to late nineteenth century in which maps proliferate: detective fiction. The larger argument here, as with the previous chapter, is concerned with making clear the importance of the map as a core generic element that signals the importance of spatial meaning for works of detection. In the case of detective fiction, the genre is intrinsically spatial yet there is remarkably little critical and theoretical exploration of it in such terms. This chapter seeks to clarify the critical reasons for this as well as to redress the imbalance. Conceptually, in relation to a genre that is strongly focused on the search for truth (and must find a clear way forward despite manipulation and deception) the chapter reveals the extent to which the fictional map works against the text it supposedly supports. The map is presented (denotatively) as if it is a straightforward part of the scientific, objective and logical superstructure of the detective’s method, yet it frequently turns out (connotatively) to participate in human acts of deception against detective or reader, or both. In a way that has not previously been attempted, the first section of the chapter seeks to locate the origins of the house plan in the earliest illustrated newspapers (the Illustrated London News and the Illustrated Police News). It identifies the first examples of maps for murder cases in newspapers, and then in fictional works. The second section establishes a theoretical context for the genre in terms of self-referentiality and considers the role and function of the map in detective fiction of the 1920s and 30s as an item that appears to be trustworthy and to form part of an empirical search for truth, but that may well function counter to this. The third and final section opens up the function of the map in new ways to consider its

127

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

128

Mapping Murder

role as part of ‘human geometry’ – a network of communal, social and psychological relations with particular reference to the interwar village and a female interpreter (in the form of Miss Marple).

The Origins of Maps in Detective Fiction I want to begin by exploring the complementary relationship between real and fictional acts of detection, between journalistic and novelistic writing about crime that is found at the historical moment when detective fiction emerges as a genre in the second half of the nineteenth century. Understanding the origins of the map’s presence alongside the narrative, first in journalistic reporting and then in fictional detection, also helps in the understanding of its function and in relation to questions around the neutrality and objectivity of the map that will be considered further in the second part of the chapter. Critics of detective fiction as it emerges in the nineteenth century have long argued for a complementary relationship between fact and fiction, as well as for the shared nature of ‘sensational’ reporting of real crimes in newspapers and of fictional crimes in the novel. In an article on sensational journalism, Dallas Liddle makes clear the ways in which different forms of writing were bound together from the start: The three major genres of writing labelled ‘sensational’ in Great Britain in the 1860s – the sensation novel, the sensation drama, and sensational newspaper journalism – are now usually considered parallel and complementary projects, even as variations on the same theme.1

Haia Shpayer-Makov in her excellent study, The Ascent of the Detective, goes further, making a convincing case for a symbiotic relationship between the press and the newly emerging detective forces.2 She argues that they adopted similar methods, had overlapping roles and developed alongside each other: ‘Not only were both occupations alike, they also evolved in parallel. Each had existed before the nineteenth century, but for both the 1840s constituted a turning point’ (157). In consequence, ‘the two occupations developed a certain mutuality and even interdependence, 1

2

Dallas Liddle, ‘“Anatomy of a Nine Days’ Wonder”: Sensation Journalism in the Decade of the Sensation Novel’, inVictorian Crime, Madness and Sensation, ed. Andrew Maunder and Grace Moore (Aldershot: Ashgate, 2016), 89–104; 89. Haia Shpayer-Makov, The Ascent of the Detective: Police Sleuths in Victorian and Edwardian England (Oxford: Oxford University Press, 2011).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

The Origins of Maps in Detective Fiction

129

as the links between them were potentially advantageous to both’ (158). The police needed positive representation in the press, journalists needed material about crime to satisfy their growing readership. The result was ‘[a] web of mutual dependencies between detectives and journalists interested in crime . . . propelled by a dynamic relationship moulded by the internal changes taking place in both occupations’ (160). It is also necessary to bear in mind the close historical proximity between the founding of a detective force in France and England and the development of fictional versions of the detective. It is widely accepted that detective fiction as a literary genre takes as its formal starting point Edgar Allan Poe’s story ‘The Murders in the Rue Morgue’ which first appeared in Graham’s American Magazine in 1841. However, the setting of his stories in France, by Poe, in its turn reflects the early establishment of a Brigade de Sûreté (Security Brigade) by Francois Vidocq in Paris from 1811 onwards – the first detective force in Europe.3 Vidocq was a criminal and police informer whose skills and knowledge enabled him to turn detective. A larger-than-life figure, Vidocq himself straddles another liminal boundary for the form – that between criminal and detective – as Julian Symons makes clear: Vidocq’s importance rested in his nature as the archetypal ambiguous figure of the criminal who is also a hero. The interpenetration of police with criminals, and the doubt about whether a particular character is hero or villain, is an essential feature of the crime story, and Vidocq embodied it in his own person.4

The mid-nineteenth century is a point at which narrative sympathy (in fact and fiction) shifts from the romanticised figure of the criminal, to the detective as heroic force for good, a shift that Vidocq embodies and enacts. As Howard Brown notes, in an article exploring the relationship between police work and its representations: This shift is not to be found either solely in the realm of quotidian police practices, or principally in the area of literary representations; rather, it is the interplay of these two relationships between practices and perceptions, that defined the new era.5

3

4

5

In 1833, after leaving the Brigade de Sûreté, Vidocq also established the first private detective agency in France. Julian Symons, Bloody Murder: From the Detective Story to the Crime Novel: A History (London: Faber and Faber, 1972), 29–30. Howard Brown, ‘Tips, Traps and Tropes: Catching Thieves in Post-Revolutionary Paris’. In Police Detectives in History, 1750–1950, ed. Clive Emsley and Haia Shpayer Makov (Aldershot: Ashgate, 2006), 33–60; 34.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

130

Mapping Murder

After he retired, Vidocq ghost-wrote a collection of cases, entitled Mémoires de Vidocq, (1828–9) which further functioned to provide a ‘bridge between the traditional literature of roguery stretching back to the Spanish picaro and the modern crime novel usually centred on a brilliant detective’ (Brown, 38). The supposedly autobiographical nature of the (ghostwritten) Mémoires meant that the first two volumes charted the shift from criminal to detective in Vidocq’s own life, before the last two established a highly influential episodic ‘case study’ model, recounting various cases led by him as director of the Brigade de Sûreté. Thus, criminals became detectives, and detectives became their own, semi-fictionalised heroes. In Britain, the Metropolitan Police Force was established slightly later than the Paris Police, in 1829, with the Detective Branch established in 1842 (the same year as the Illustrated London News started). As is well known, both Charles Dickens and Wilkie Collins based their literary detectives on real-life examples (Inspector Field and Inspector Whicher respectively) and drew heavily on the journalism of the day for plots and themes.6 Meanwhile, in 1849, a similar series of supposed case-studies to that of Vidocq was begun by Chambers Journal, entitled Recollections of a Detective Police Officer, with the author ‘Waters’.7 ‘Waters’ was not, however, (as the title might lead one to assume) a retired policeman drawing on his past experiences, rather this was a pseudonym for a journalist, William Russell. As with Vidocq, then, the boundaries between actual and fictional detection on both sides continued to be blurred for this emerging form.

Mapping Crime Scenes Having established, however briefly, a fluid exchange across real and fictional forms of written crime representation, I want now to focus on two early, illustrated newspapers, the Illustrated London News (1842–1901) and the Illustrated Police News, Law Courts and Weekly Record (1864–1938) in order to identify the earliest use of maps to visualise murder cases in newspaper

6

7

Dickens wrote a number of articles including: ‘The Modern Science of Thief-Taking’, Household Words (13 July 1850); ‘A Detective Police Party’, Household Words (27 July and 10 August 1850); ‘Three “Detective” Anecdotes’, Household Words (14 September 1850); and ‘On Duty with Inspector Field’, Household Words (14 June 1851). These were extremely influential in shaping public opinion about the detective force. See a full account in Makov, Ascent of the Detective, ch. 5. William Russell, The Recollections of a Detective Police Officer by ‘Waters’ (London: W. Kent & Co., 1849).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Mapping Crime Scenes

131

reporting.8 The Illustrated London News was the first illustrated news magazine in the world. It started in May 1842 and was published on a weekly basis, costing sixpence. The paper covered a number of regular items: ‘Royal Visits and Events’, ‘National Sports’, ‘Churches of the Metropolis’, ‘Theatres and Fashion’, to name a few. A range of maps and plans were used as illustrations from early on, celebrating achievements in wars or colonial lands or being used to illustrate technical or architectural advances (particularly in relation to the City of London).9 From 1843 onward, impressive large maps that anticipated new developments in the capital were provided, such as the plan for the new Houses of Parliament or for the proposed building of Embankment (30 September 1843 and 30 December 1843). These maps included complex street plans and interiors and so perhaps paved the way for the presentation of maps in relation to real murder cases – and finally for their appearance within fictional works. Although the Illustrated London News included a regular police and court section, it did not adopt a sensationalist approach to crime reporting and this is reflected in the simple factual form of the maps it provides. The first time a map is given in relation to a murder case occurs on 12 April 1845 for ‘The Hampstead Murder’ (Fig. 4.1). In this case, Thomas Henry Hocker, in an act of highway robbery, stole from, and then assaulted, James Delarue, causing his death. Eyewitness accounts by a nearby baker and policeman make much use of street names and physical positioning – the witness delivering bread in Haverstock Terrace ‘heard the cry of murder six or eight times’ but both he and the policeman were not immediately able to locate the wounded man: ‘I went up Haverstock Terrace, but neither saw nor heard anything. I then went down Belsize lane and got into the Hampstead Road again’. The map is provided as an aide to the reader: ‘We annex a plan of the neighbourhood in which this atrocious murder was perpetrated; and in illustration of the evidence, it will be serviceable to those who are unacquainted with the locality.’10 The map thus ostensibly helps the reader to situate events but, since neither the movements of the witnesses nor the location of the body in a corner of the field are marked on the map, it is only of quite limited use.

8

9

10

My thanks to Kate Ingle for her assistance in locating the earliest examples of these maps and commenting on this chapter. The first map of any kind appears in June 1842 in relation to the Afghan War (‘Map of the Affghan Country’). From 1842 onward there are maps of battles in India, or territories in China and cross-section plans of boats and railways. Illustrated London News, 12 April 1845, 230.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

132

Mapping Murder

Fig. 4.1. ‘Map of the area in Hampstead where Thomas Henry Hocker robbed and murdered James Delarue’ from Illustrated London News, 12 April 1845

The first interior plan to be provided as an illustration in relation to a murder occurs in the Illustrated London News for 8 September 1849, in relation to ‘The Bermondsey Murder’ (Fig. 4.2). An earlier newspaper account (1 September 1849) had provided the fairly standard exterior illustration of the house at 3 Miniver Place, Bermondsey, in which the murder of Patrick O’Connor by a married couple (the Mannings) took place. A later account of the examination of the prisoners (once they had been located and caught) then further provided the plan view as a replica of material evidence used in the case: ‘copied from the Plan prepared by direction of the solicitor for the prosecution, and produced in the court’. ‘The Bermondsey Murder’ was a notorious case, in which husband and wife together lured the victim to their house, murdered him and buried him under the flagstones of the kitchen floor. However, you would not gather any of this from looking at the map, which is presented without any additional information to connect it to the dramatic and grotesque events that occurred in that ordinary suburban location. Nevertheless, since the case was extremely well known, it is safe to assume that any contemporary reader would have been able to superimpose such knowledge onto the plan. In that case, its very objectivity renders the map sinister since beneath the clean surface that it represents something terrible lies hidden (literally). In this way, early newspaper maps begin to anticipate the paradoxical doubleness of later maps in detective fiction.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Mapping Crime Scenes

133

Fig. 4.2. ‘Plan of the basement of the house at no.3 Miniver Place, Bermondsey, where the body of Patrick O’Connor was discovered under flag stones’ from Illustrated London News, 8 September 1849

I want now to turn to a second example, the Illustrated Police News. This was a four-page weekly newspaper costing a penny and specialising in sensational and bizarre stories from the police courts in Britain and abroad. A key element in the success of the paper was its dramatic front page, presented as a full-page, black-and-white illustration from November 1867. This page often consisted of a montage, depicting the various events to be reported within the paper. When there was a significant story, particularly a crime of some kind, the montage would feature different stages of the crime, portraits of the criminal or victim and images of key settings. Somewhat surprisingly, although there are frequent illustrations of the outside of a house or of the interior of a room relating to a dramatic crime, the first use of an actual plan in relation to murder does not occur until

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

134

Mapping Murder

Fig. 4.3. ‘Tragedy at Reading’ from Illustrated Police News, 18 January 1879

1879 (see Fig. 4.3). This is given in relation to the ‘Tragedy at Reading – A Policeman’s Wife Shot Dead’.11 Whilst the headline suggests some kind of revenge killing, the story is in fact a terrible accident. A postmaster, Albert Bullock, whose business was not doing well, took out his anger by shooting off bullets: ‘He afterwards went out in to his yard and commenced firing off his pistol, it is said, at a blackboard, but in a direct line to a room in which were Police Constable Clifford, his wife and child.’ As a result, one bullet was shot into the open window of his neighbour, killing the woman. The text attached to the front-page story states that ‘By referring to the plan printed on the top of our front page; the reader will be able to understand more clearly the direction of the shot, and how the accident occurred.’ Here the plan is explicitly introduced because the need for spatial understanding in relation to an (extraordinary and unlikely) event is acknowledged. When we look more closely at it we can immediately see that the Illustrated Police News takes the use of maps for murder further forward than those for the Illustrated London News. The map forms part of the total visual display and unites the other visual elements – the different houses and the focal dramatic scene inside the front room. The plan view is given in order to provide an alternative perspective ‘Showing the House and Line of Shot’ alongside these. It also begins to anticipate the later form of a literary map since it includes on it verbal information that ties it into the narrative – such as the ‘course of the bullet’. What is striking to the modern eye is the effect created by representing real-life dramas and tragedies not by means of photography but through illustration.12 This immediately distances and stylistically fictionalises realworld events. In the case of the maps themselves, the hand-drawn sketch, and the nature of the writing upon it, also have a kind of fictional quality. 11 12

Illustrated Police News, 18 January 1879, issue 779. Photographic reproduction was only able to be widely used from the 1880s onward.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Mapping Crime Scenes

135

Here, then, the nature of the visual reproduction of events itself blurs the boundaries between fact and fiction in a way that is highly representative of the closely integrated relationship between real-life crime and its fictional counterpart at the historical moment when the two developed together. When we turn from plans for real murders to their fictional counterparts this sense of interchange between fact and fiction is strongly felt once more. We have seen that the first plans for murder in British newspapers occur in the mid-1840s (soon after the establishment of Scotland Yard), with maps that anticipate and resemble fictional maps appearing by the 1870s. The presence of a house plan in a work of fictional detection first appears between these two dates as the genre of detective fiction develops. The Notting Hill Mystery by Charles Warren Adams (actually Charles Felix) was serialised in Once a Week from 1862 to 1863, and published in book form in 1865.13 Claims have been made for this book as the first detective novel and, whether this is true or not, it is certainly the first fictional work of detection to contain an explicit map.14 The Notting Hill Mystery is constructed as a reassembling of evidence from different sources by an insurance man looking into a life insurance claim. It concerns a weird case of manipulation through mesmerism in which twin sisters, separated at birth, are susceptible to the health of each other. The murderer (a mesmerist Baron R), realising this, marries the stronger of the two sisters and poisons her to a degree that almost kills her, but does actually kill her wealthy twin by means of the sympathetic connection. The twin thus dies with all the apparent symptoms of poisoning but no physical evidence of it – allowing the Baron to inherit the family money via his wife. The map (see Fig. 4.4) provided at the start, of the ‘Basement Floor of Baron R**’s House in Russell Place’, does not relate to the primary murder but is given in relation to the secondary murder of the second sister (the Baron’s unfortunate wife) later in the narrative. As a result, it is explicitly linked to a particular section of the narrative (‘Vide Section vii. 8’) in which she is hypnotised into going downstairs into the laboratory and drinking acid. Although the Baron claims that his wife was sleepwalking down to the kitchen, witnesses confirm the self-revealing positions of both husband and wife: ‘Madame 13 14

Charles Felix, The Notting Hill Mystery (London: Saunders & Otley, 1865). Julian Symons makes this claim in Bloody Murder: ‘There is no doubt that the first detective novel, preceding Collins and Gaboriau, was The Notting Hill Mystery’ (53). A closely competing claim certainly could be made by Emile Gaboriau with L’Affaire Lerouge, (Paris: Dupont, 1866). Personally, it has always seemed to me ungenerous not to credit Bleak House (1853) as the first detective novel.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

136

Mapping Murder

Fig. 4.4. Map from The Notting Hill Mystery

R** never went into the kitchen at all’; ‘she went straight into the laboratory’; ‘the Baron watched her as she came out’ (224). The map is not evidence in itself but it works to ground and authorise the spatial elements crucial to understanding the significance of the verbal testimony. As a result, this first fictional house plan works in a way directly comparable to that in the newspapers; making spatially explicit the full implications of the evidence supplied. It also introduces and anticipates a distinctive temporality for maps in later detective fiction which often presents the map within the pages of the text and in relation to a specific stage of the investigation rather than for the work as a whole. Rather

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Mapping Crime Scenes

137

Fig. 4.5. Map from Monsieur Lecoq

frustratingly, however, although the evidence is incontrovertible and extremely damaging to the murderer (the Baron), the overall ‘doceme’ nature of the larger narrative – an insurance man reporting to his bosses and merely accumulating different forms of data – ultimately leaves it unresolved. Thus the full power of the map to support and illuminate visual evidence and spoken testimony for a crime is present, but left untapped. A second early example of a fictional map occurs in the work of Émile Gaboriau, the first French novelist to write detective fiction. The plan of the murder site and movements around it (see Fig. 4.5) is given in his novel Monsieur Lecoq (1869), the last in a series of five novels featuring this detective.15 The novel is highly spatial, both in terms of the setting in which the murder takes place and the murder scene itself. After three men are found dead in a tavern, shot by a fourth who proclaims himself innocent, Lecoq sets out to prove (against the over-rapid assumptions of his chief, Gevrol) that this was not a mere tavern brawl. Working with the far less able ‘Father Absinthe’, he reconstructs the movements of individuals around the murder site, uncovering the presence of two other women and a man and explicitly visualising the evidence: ‘The young man [Lecoq] seated himself at the table, and, with the view of making his recital as intelligible as possible, he began by sketching a plan of the scene of the murder’ (44). The map itself draws awed (ironised) respect from Father Absinthe: ‘The plan amazed that worthy man. He had seen a great deal; but he had always supposed that it was necessary to be an engineer, an architect, or, at least, a carpenter, to execute such work’ (46). The map is

15

Émile Gaboriau, Monseieur Lecoq (Paris: Dupont, 1869). The story first appeared in Le Petit Journal (a Paris daily newspaper) between May and December 1868. It was revised for book publication (2 vols.) in 1869.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

138

Mapping Murder

the first to appear in Gaboriau’s work and it immediately poses a number of questions concerning provenance and usefulness that maps in detective fiction need to raise: Why is the map there? Who made it? For what purpose? The rationale for the map is fully in line with the maps found in newspapers as discussed above: it is given in order to provide an alternative visual/spatial understanding of the scene for which such information is highly relevant. However, the narrative continues: It will be seen that in the memoranda appended to this explanatory diagram, Lecoq had not once written his own name . . . he referred to himself simply as one of the police. This was not so much modesty as calculation. By hiding one’s self on wellchosen occasions, one gains greater notoriety when one emerges from the shade. It was also through cunning that he gave Gevrol such a prominent position. (44)

This map, then, immediately presents itself as having a double meaning in the way I suggested at the start of this chapter. Whilst it has the explicit denotative meaning that we would expect – illuminating spatial understanding of the murder scene – it also strongly bears a connotative one – concerned with self-advancement on the one hand, and the discrediting of a superior on the other. The latter meaning, however, is encoded in such a way that it is protected by the former: the creator of the map is able to deny such an intention. In this way then, Gaboriau’s map could be said to stand as the first true map of detective fiction because it displays the classic doubleness (the Janus-facedness) of the literary map that, for this genre, is associated with manipulation of hidden knowledge that others seek. It has a logical, intellectual reason for being there, and presents itself as simply there to help, but the true human motives behind it are far less innocent.

Theorising Detection Thus far I have shown that maps for detective fiction emerge at the same period that maps for crime scenes begin to appear in newspapers and partake of the kinds of slippage and fluidity of the form itself. I want now to consider the emerging self-consciousness of detective fiction as a genre, and how this, too, might bear upon the explicit presence of maps or the importance of spatial meaning within the text. A metatextual approach is to some extent invited by a form that, from the start, concerns itself with the nature of the method it displays, showing

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Theorising Detection

139

what Heta Pyrhönen calls ‘a self-reflexive understanding of its own ingredients’.16 In the case of major influential examples – by Poe and Conan Doyle, but others as well – the figure of the detective is strongly concerned with developing and proving a rigorous ‘scientific’ method of detection in order to authorise his own actions and give credibility to the new profession. This scientific mode is frequently questioned by others, who are initially sceptical of it, as well as by the nature of the cases themselves (which may appear to be supernatural but are proved not to be). So, Poe’s ‘The Murders in the Rue Morgue’ opens with an extended account of the nature of the analytical power as ‘the higher powers of the reflective intellect’ (141) before rapidly moving into a display of the detective Dupin’s talents that has the narrator exclaiming ‘Tell me, for Heaven’s sake . . . the method – if method there is’ (145).17 The detective also defines his own approach in opposition to what has gone before – Dupin notes of the police that ‘There is no method in their proceedings, beyond the method of the moment’ (151), whilst Holmes, in his first appearance in A Study in Scarlet, actively dismisses both Dupin and Lecoq as his fictional predecessors (‘showy and superficial’ and ‘a miserable bungler’) in favour of his own powers of deduction based upon a bank of prior knowledge.18 Such methodological self-consciousness also bears upon the presence of a map as a supposedly objective and positivistic form of knowledge in relation to the crime scene. Turning to early twentieth-century critical attempts to demarcate the characteristics of the genre, we find that they follow the detective’s own self-projections to some extent, dwelling upon the formal properties of the narrative as an intellectual exercise comparable to other popular puzzles: ‘The pages of every magazine and newspaper swarm with cross-words, mathematical tricks, puzzle-pictures, enigmas, acrostics, and detective stories.’19 If such fiction is understood to be primarily concerned with the solving of a puzzle by a superior intellect (detective or reader-as-detective) then more literary or complex elements are not to be welcomed since they 16

17

18

19

Heta Pyrhönen, ‘Criticism and Theory’, In A Companion to Crime Fiction, ed. Charles J. Rzepka and L. Horsley (Chichester: Wiley-Blackwell, 2010), 43–56; 43. Edgar Allan Poe, The Complete Tales and Poems, ed. Hervey Allen, (New York: Random House, 1938). Gaboriau’s young detective, Lecoq, works in the same way, and this is also true of Gaston Leroux’s The Mystery of The Yellow Room (Le Mystére de la chambre jaune, 1908) where there is a direct conflict between two detectives (Larsan/Rouletabille) working two different systems to solve the crime. D. L. Sayers, ‘Introduction’, The Omnibus of Crime (New York: Payson and Clarke Ltd, 1929), 72.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

140

Mapping Murder

distract from this. S. S. Van Dine’s ‘Twenty Rules for Writing Detective Stories’ (1928) and Ronald Knox’s ‘Ten Commandments for Detection’ (1929) confirm such a view, describing it as ‘a kind of intellectual game’ for which ‘the method of the murder and the means of detecting it must be rational and scientific’ before proceeding to define the rules of that game in terms that, again, reduce expectations for the form as a literary genre capable of bearing detailed analysis.20 In fact, Van Dine makes exactly this point with his sixteenth rule: A detective novel should contain no long descriptive passages, no literary dallying with side-issues, no subtly worked-out character analyses, no ‘atmospheric’ preoccupations. Such matters have no vital place in a record of crime and deduction. They hold up the action and introduce issues irrelevant to the main purpose, which is to state a problem, analyze it, and bring it to a successful conclusion. (Haycraft, 191)

The fact that successful major writers themselves openly expressed unease about their chosen form paved the way for the fullest critical expression of such concerns in Edmund Wilson’s 1945 article for the New York Times entitled ‘Who Cares Who Killed Roger Ackroyd?’21 Wilson wittily mocks the concept of detective fiction as a game for which the reader is required to know the rules. He concludes: ‘with so many fine books to be read, so much to be studied and known, there is no need to bore ourselves with this rubbish’ (397). How to analyse such texts is not even on the table, when what is at issue is the question of why anyone should waste their time reading them. The genre appears to be rescued from its sense of low literary self-worth in the second half of the twentieth century when it is appropriated by the high priests of French psychoanalytic and deconstructive theory: Lacan and Derrida.22 On the one hand, this serves to elevate the genre of detective fiction by virtue of its self-referentiality. A form of fiction concerned with 20

21

22

Ronald A. Knox, ‘Ten Commandments of Detection’, in his introduction to The Best English Detective Stories of 1928 (London: Faber, 1929); S. S. Van Dine, ‘Twenty Rules for Writing Detective Stories’, The American Magazine, September 1928. Both pieces are reproduced in Howard Haycraft, The Art of the Mystery Story: A Collection of Critical Essays (New York: Carroll & Graf, 1974), 189, 191. Wilson wrote three pieces attacking the genre of detective fiction: ‘Why Do People Read Detective Stories?’, New Yorker, 14 October 1944; ‘Who Cares Who Killed Roger Ackroyd?: A Second Report on Detective Fiction’, New Yorker, 20 January 1945; ‘Mr. Holmes, They Were the Footprints of a Gigantic Hound’, New Yorker, 17 February 1945. The second article is reprinted in Haycraft, Art of the Mystery Story, 390–7. See Jacques Lacan, ‘Seminar on “The Purloined Letter”’, Yale French Studies 48 (1972): 39–72; and Jacques Derrida et al., ‘The Purveyor of Truth’, Yale French Studies 52 (1975): 31–113.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Theorising Detection

141

the search for a particular truth through self-consciously rigorous method proves easily able to bear a weight of hermeneutic complexity centred upon the reading and misreading of signs. On the other, the result of this is that the form itself is effectively commandeered for a meta-critical debate, centred on the nature of theoretical illumination rather than on the understanding of the particular genre. Thus, Lacan’s reading of Poe’s short story ‘The Purloined Letter’ seeks to explore the nature of intersubjectivity through two (repeatedly redetermined) triadic relationships between the three main characters and the detective: ‘to illustrate that it is the symbolic order which is constitutive for the subject by demonstrating in a story the decisive orientation which the subject receives from the itinerary of a signifier’ (40).23 In the narrative, a letter containing compromising information has been stolen from a lady by a blackmailing government minister. It is this that the detective, Dupin, must locate and retrieve. The letter itself functions as ‘pure signifier’(45): a material object the content of which is never revealed and which is in any case far less significant than the power it wields in terms of who possesses it: ‘with the letter’s use its power disappears’ (62). Against this, Derrida critiques the kind of reading Lacan offers, arguing that he uses the text illustratively to try and prove a psychoanalytic truth but that, in doing so, he deliberately writes out the act of narration that forms the fourth element in a relationship that is not triadic at all. For Derrida, ‘The insistence of the Lacanian letter is the sublation of writing in the system of speech’ (85). Finally, Barbara Johnson critiques Derrida’s critique to show that Derrida’s own reading is subject to exactly the same criticisms he brings to bear upon Lacan.24 Whilst these articles undoubtedly contain illuminating readings of the play of meaning and power in the text, their primary focus is their interest in each other so that, as Johnson concludes her essay: ‘the true otherness of the purloined letter of literature has perhaps still in no way been accounted for’ (505). At the risk of my own act of critical appropriation and reduction, it is worth noting that two core characteristics of detective fiction as a spatial genre vitally inform and underpin these readings. The first is a structure of doubling and return inherent from the start in Lacan’s primary focus on this particular story as an example of ‘intersubjective repetition’ (45) but

23

24

Lacan continues at this point: ‘It is that truth, let us note, which makes the very existence of fiction possible’ (40). Barbara Johnson, ‘The Frame of Reference: Poe, Lacan, Derrida’, Yale French Studies 55/56 (1977): 457–505.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

142

Mapping Murder

also of central importance to a form concerned with reading, misreading, manipulation and misdirection. The second is the geometric patterning of meaning inherent in the form at a level of both character and plot.25 Lacan undertakes a brilliant micro-reading of the kind of intimate ‘human geometry’ that I will be discussing later in the chapter in his neat summation of the multiplicity of ‘the glance’ in ‘The Purloined Letter’: The first is a glance that sees nothing: the King and the police. The second, a glance which sees that the first sees nothing and deludes itself as to the secrecy of what it hides: the Queen, then the Minister. The third sees that the first two glances leave what should be hidden exposed to whomever would seize it: the Minister, and finally Dupin. (44)

Here, the stolen letter functions as an intersubjective prism with angles of sight centred (or not) upon it. Thus, as a heavily overloaded ordinary object it makes explicit the centrality of geometric patterning to both story and genre at every level: plot, action, motivation, character, solution. At its crudest for detective fiction in a larger sense, this patterning concerns only physical elements relating to the investigation, such as angle of shot, movement of characters at a certain time and place etc. but it is also capable of being interwoven with greater levels of complexity in relation to human interaction and motivation and in relation to both map and text. One other highly influential theoretical model is offered by Tzvetan Todorov in his essay ‘The Typology of Detective Fiction’.26 He celebrates the normative influence of early codification on the grounds that ‘the masterpiece of popular literature is precisely the book which best fits its genre’ (43) but also draws out a central structural duality, ‘This novel contains not one but two stories: the story of the crime and the story of the investigation’ (44).27 These two stories are temporally separated (‘the first story, that of the crime, ends before the second begins’ (44)) but spatially connected, with the result that ‘the whodunit thus tends toward a purely geometric architecture’ (45). The second structure, in particular, involves a necessary ‘mapping out’ of spatial and human relations in order 25

26

27

Johnson notes: ‘“The Purloined Letter” is traversed by an uncanny capacity for doubling and subdividing. The narrator and Dupin are doubles of each other, and Dupin himself is first introduced as a “Bi-Part Soul” . . . the Minister and Dupin become doubles of each other through the fact of their both being already double’ (‘Frame of Reference’, 470). Tzvetan Todorov, ‘The Typology of Detective Fiction’, in The Poetics of Prose, trans. R. Howard (Oxford: Blackwell, 1977): 42–52. Todorov maps this core structure – ‘what really happened’ vs ‘how the reader (or the narrator) has come to know about it’ (45) – onto the universal Russian formalist distinction between fabula (story in chronological order) and syuzhet (plot).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Trusting and Not Trusting the Map

143

to reconstruct the first and so resolve the narrative. Only once everything is ‘in place’ can truth emerge. In her article on ‘Criticism and Theory’, Heta Pyrönhen further extends the influential narratological argument made by Todorov, whilst incorporating the genre’s high level of self-referentiality to outline a doubled readership inherent in detective criticism that effectively allows a work to be read in two ways: When treated as mass-produced literature, detective fiction is assumed to represent a paradigmatic case of easy readability. In contrast, when the genre’s literary selfawareness forms the starting point of analysis, the conclusions emphasize the readership’s appreciation of form. (45)

She argues that, while, on the one hand, the form does rely on ‘codes and conventions that make detective stories easy to read’ (54), there is also inherent within it the capacity for a ‘re-reading approach’ that is ‘based on treating the genre as a self-reflexive textual enigma’ (54). This context fits well with the Lacanian account given above in which the first, second and third glances represented within the text as different levels of nuanced reading of the situation also correspond to greater interpretative perspicacity on each readerly return. Such an approach is of interest here, in part because it strongly applies to a work such as The Murder of Roger Ackroyd (considered below) where double meanings are built into the text in ways that only a second reading can uncover, but also because the doubled nature of readership clearly applies to the reading of maps, and of map and text together. In other words: what is true for the genre is also true for the map. The interior or exterior plan ostensibly serves a supportive function in relation to the highly formal pattern of the investigation – involving the reader in a secondary reconstructive act from that of the narrative itself. Thus, at a primary level it seems to partake of the cold intellectual game-playing inherent in the structure. But at the same time the map is also a human form, made by someone, for someone else to read, and with all kinds of (often disturbing) motivations also held within it.

Trusting and Not Trusting the Map Maps in the form of house or village plans are found in a significant number of major writers of detective fiction in the first half of the twentieth century. They are used on more than one occasion by Dorothy L. Sayers; Margery Allingham; Ngaio Marsh; and Edmund Crispin, to name a few key

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

144

Mapping Murder

figures.28 However, the Queen of Crime is also the Queen of Maps. Agatha Christie presents maps within nine books, but she also frequently presents more than one map at different scales so that her body of work contains twelve maps in total.29 Unsurprisingly then, Christie’s first novel of 1921, The Mysterious Affair at Styles, contains a map.30 This is also Poirot’s first case, introducing the famous Belgian detective. The story is narrated by Poirot’s rather limited British assistant, Colonel Hastings, who, recovering after the war, has gone to stay with a family friend, John Cavendish. Whilst he is there, John’s mother is taken ill in the night and dies, with suspicion of murder falling upon her new husband, Alfred Inglethorp. Poirot happens to be settled in the village as a Belgian refugee and is called in by Hastings to help solve the case quietly. Suspicion falls on various characters until it is finally revealed that the first suspect (the husband) was in fact the murderer all along. The book makes use of an interior plan twice over, presenting two maps at different scales: a floor plan of all the bedrooms and then a detailed plan of the room in which the murder takes place (see Figs. 4.6 and 4.7). In The Mysterious Affair at Styles it is Colonel Hastings who provides readers with a map showing the sleeping arrangements of the family on the night that Mrs Inglethorp dies (poisoned, in her bed). The house plan is presented immediately prior to Hastings’s account of being woken in the night and his eyewitness description of various members of the family trying to gain access to Mrs Inglethorp. It is introduced in an explicitly integrated way by the narrator to his readership: ‘To make this part of my story clear, I append the following plan of the first floor of Styles’ (42). The situating of the map within the text rather than at the front of the book, and the presentation of it immediately before the key event takes place, all encourage readers actively to read and interpret at a specific point within the text. In the ensuing chapter, a second larger-scale map is provided as a detailed plan of the room in which the murder took place and this is given immediately prior to Poirot’s detailed examination of that room in which 28

29

30

Five works by D. L. Sayers contain a map: Lord Peter Views the Body; The Nine Taylors; Five Red Herrings; Unnatural Death; Clouds of Witness. Five works by Marjorie Allingham contain a map: Cargo of Eagles; More Work for the Undertaker; Sweet Danger; Dancers in Mourning; Mystery Mile. Agatha Christie’s use of maps may owe much to Gaston Leroux’s locked room novel The Yellow Room (1908) which provided two detailed plans. (See An Autobiography (London: W. Collins Sons & Co. Ltd, 1977), 216, 263). Maps occur in: The Mysterious Affair at Styles; Murder at the Vicarage; Murder on the Orient Express; Death in the Clouds; Evil Under the Sun; Towards Zero; They Do It With Mirrors; The Murder of Roger Ackroyd; Poirot Investigates. Agatha Christie, The Mysterious Affair at Styles (London: W. Collins Sons & Co. Ltd, 1921).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Trusting and Not Trusting the Map

145

Fig. 4.6. House plan from The Mysterious Affair at Styles

Fig. 4.7. Room plan from The Mysterious Affair at Styles

he uncovers ‘six points of interest’. The maps are thus fully integrated within the narrative structure materially and conceptually since (in Todorov’s terms) the first map relates to the final event of the first narrative (the ‘story of the crime’) and the second map to the first event of the second narrative (the ‘story of the investigation’). These maps appear to ground events by presenting an objective, empirical account that will contribute to the search for truth. However, they also partly distract, since the information they provide – in this novel at least – ultimately proves to be irrelevant. The murder has actually been committed by preparation of a bromide in advance, so that the location of

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

146

Mapping Murder

suspects given in the floor plan at the time of the death has no bearing upon the final solution. The potential of the map to mislead is further reinforced when it turns out that the murder was committed by someone not even in the house at the time (the victim’s husband, Alfred Inglethorp). In fact, the absence of an individual from the building – itself a potentially suspicious act – is in danger of being overlooked as a result of the map’s focus upon the movements of those present.31 The reader trusting the authority and apparent objectivity of the map naturally assumes that it is selectively representing the most relevant and useful information, but this is not necessarily the case. The second, more detailed, plan in the narrative has an immediate spatialising function in locating and fixing the various clues that Poirot finds in the room (a piece of fabric in the door, a smashed coffee cup on the floor by the bed, etc.). However, ‘the final proof’ (275) which does exist in that room (a piece of paper hidden amongst spills in a vase on the mantelpiece) is overlooked by Poirot until much later in the book. At the time that the reader is given the room plan it is accounted for only in the following generalised description: ‘He rose from his knees, and walked slowly across to the mantelpiece, where he stood abstractedly fingering the ornaments, and straightening them’ (63). The reader is not told either that there is a vase, or that there are pieces of paper in it. Thus, the apparent specificity and helpfulness of the map in situating ‘clues’ within the room again works only to deceive. The limits of the two maps should also serve to remind us of their fictional origins. They are provided for the reader by Hastings, not Poirot, and Hastings is very clearly an unreliable narrator – made clear from his lack of intelligence, his bias against Alfred Inglethorp (husband of the victim), his friendship with the victim’s son, John Cavendish, and his partiality for Mary Cavendish. If the narrator is unreliable, then so are the maps he provides. From the start, then, Christie’s maps signal a need to read actively, not passively, and to reconsider and revisit events, encouraging the ‘advanced’ readerly dynamic for the genre as described by Pyrhönen. The best example of this occurs in Agatha Christie’s controversial 1926 novel The Murder of Roger Ackroyd (controversial because it does not adhere to the rules of ‘fair play’, most notably Knox’s first rule: ‘The criminal must not be anyone whose thoughts the reader has been allowed to follow’).32 The story is narrated in the 31

32

The narrator, Col. Hastings, does note the suspicious nature of the absence: ‘Where was Alfred Inglethorp? His absence was strange and inexplicable’ (49). However, because his judgement cannot be trusted, we overlook this. Haycraft, Art of the Mystery Story, 194. In An Autobiography, Christie notes: ‘Of course a lot of people say that The Murder of Roger Ackroyd is cheating; but if they read it carefully they will

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Trusting and Not Trusting the Map

147

first person by the reassuring figure of Dr Sheppard – the family doctor, who has access to many secrets and intimacies within the village. A first death – that of Mrs Ferrars – is followed by the death of the squire, Roger Ackroyd, who is shot in his study. It emerges that Mrs Ferrars took her own life after killing her husband, but was being blackmailed by someone who knew she had done this, so that the sending of a letter to Roger Ackroyd just before her death (revealing all to him) then leads to his murder. The doctor rushes to the Hall following an unattributed phone call to say that the killing of Ackroyd has occurred, and so finds the body. Poirot, as a mysterious neighbour, gradually becomes involved, with the doctor occupying Hastings’s role as his amateur assistant. After considerable misdirection, and with suspicion falling on Roger Ackroyd’s stepson, the apparently reliable first-person narrator is ultimately revealed to be the murderer. Throughout, the novel plays with readerly assumptions about narrative authority, appearing to authorise the narrator, not only through Poirot’s approval but also by his social status as doctor. Like Hastings, Dr Sheppard, conforms to Todorov’s description of the self-consciously literary narrator and he is also explicitly compared by Poirot to his former sidekick: ‘“You must indeed have been sent from the good God to replace my friend Hastings,” he said, with a twinkle. “I observe that you do not quit my side”.’33 Since generic conventions lead us to place our faith in the detective’s assistant, this lulls us into trusting him. In this case, however, even for the self-aware reader, there is a further temporality built into the text since it is impossible to appreciate the doubleness of the text unless you already know the solution. In other words, only on a second reading can one read the narrator against himself. Dual intentions are attributable to many of his comments only once one knows that he is the murderer. This feature of narration is present from very early on in the book, as the following conversation between the narrator and his sister about the possible suicide of Mrs Ferrars illustrates: ‘Ten to one she’s left a letter confessing everything.’ ‘She didn’t leave a letter of any kind,’ I said sharply, and not seeing where the admission was going to land me. ‘Oh!’ said Caroline. ‘So you did inquire about that, did you?’ (6)

33

see that they are wrong. Such little lapses of time as there have to be are nicely concealed in an ambiguous sentence’ (352–3). Agatha Christie, The Murder of Roger Ackroyd (London: W. Collins and Sons Co. Ltd, 1926), 107.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

148

Mapping Murder

In this dialogue, Caroline, without any evidence, intuitively (and correctly) guesses that the dead woman committed suicide and left some kind of final note. The doctor replies ‘sharply’ – an emotional response that is interpreted by his sister as confirmation of her own opinion. However, once one knows that the doctor is the murderer, and that the letter under discussion (which has been posted to Roger Ackroyd) will incriminate him as the dead woman’s blackmailer and cause him to murder its recipient, then his ‘sharpness’ is attributable to a very different cause – not one of uncovering a suicide but of covering up his own actions. A further element of verbal misdirection occurs in the comment ‘not seeing where the admission was going to land me’ which appears only to have immediate meaning for the dialogue taking place, but, of course, also has far more sinister resonance in the light of the narrator’s future actions. A second example of doubled intentions occurs when the doctor describes his actions in relation to Ackroyd. The latter receives the incriminating letter from Mrs Ferrars whilst he is there but is reluctant to open it: But for some reason, obscure to myself, I continued to urge him. ‘At least read the name of the man,’ I said. Now Ackroyd is essentially pig-headed. The more you urge him to do a thing, the more determined he is not to do it. (63)

Here, Sheppard at first appears to want to bring his fate onto himself (since it is his name in the letter) but – as he informs the reader – his knowledge of Ackroyd’s character means that he knows that by doing this the opposite will occur. So a strategy of apparent openness – to Ackroyd and the reader – is actually an act of deliberate calculation. This contradictory mode of behaviour characterises Sheppard throughout. Elsewhere, simple statements of fact that appear to relate to his role as doctor become highly loaded once we are aware that Sheppard is the agent of destruction: ‘Ackroyd was sitting as I had left him in the arm-chair before the fire.’ (70) ‘Parker hurried away, still wiping his perspiring brow. I did what little had to be done.’ (71)

Later, when he has revealed his true identity at the end of the book, Sheppard himself glories in his own acts of omission and elision: ‘What a judicious use of words: “I did what little had to be done”. It was quite little – just to shove the dictaphone into my bag and push the chair against the wall in its proper place.’ (367)

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Trusting and Not Trusting the Map

149

Fig. 4.8. House plan from The Murder of Roger Ackroyd

However, as events unfold and power begins to shift from murderer to detective, Poirot plays this game of doubled-rhetoric back onto the murderer. At times this leads him close to revealing that he has seen through the Doctor’s many deceptions: ‘Everyone concerned in them has something to hide.’ ‘Have I?’ I asked, smiling. Poirot looked at me attentively. ‘I think you have,’ he said quietly. (114)

At other times the detective, like the doctor, makes an apparently throwaway comment that is heavily loaded: ‘“Not at all,” cried Poirot heartily. “You and I, M. Le docteur, we investigate this affair side by side. Without you I should be lost”’ (122; my italics). At every level of plot, character and dialogue, as well as in relation to the narrator and the detective, the novel provides dual intentions. When we turn from the verbal to the visual ‘text’, therefore, it stands to reason that, if Hastings as map-maker was not quite to be trusted, this is far more true for Dr Sheppard. Two maps at different scales are once again provided by the narrator (Figs. 4.8 and 4.9). The first is a plan of the house given soon after the finding of the body when he tells us: ‘To make things clear and explain the position, I have appended a rough sketch of the righthand wing of the house’ (62). As in The Mysterious Affair at Styles, the first map provides largely irrelevant information regarding the movements of the murderer. In fact, it goes further, since it deliberately misleads, encouraging the reader to consider the movements of those in other parts of the Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

150

Mapping Murder

Fig. 4.9. Room plan from The Murder of Roger Ackroyd

house, when we have been ‘beside’ the murderer all along. It also represents the immediate environs around the house as well as the inside, strongly leading the reader to assume that the murderer came from outside, when this is not the case. The second, more detailed, large-scale map is of the room in which the murder took place. The narrative at the point when the map is given within the text is concerned with discussion of a chair which the butler affirms was pulled out to face the door in an odd position, but which has now been pushed back into place. Poirot asks him: ‘Anything else?’ ‘Yes, sir, this chair was drawn out a little more.’ He indicated a big grandfather chair to the left of the door between it and the window. I append a plan of the room with the chair in question marked with an X. ‘Just show me,’ said Poirot. The butler drew the chair in question out a good two feet from the wall, turning it so that the seat faced the door. ‘Voilà ce qui est curieux,’ murmured Poirot. ‘No one would want to sit in a chair in such a position, I fancy.’ (91–2)

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Human Geometry

151

However, in the map provided within the text (around which this dialogue takes place on the page) Dr Sheppard (as map-maker) presents the chair placed pushed back, in the upper right-hand corner, a fact to which Poirot later draws the reader’s attention: ‘In your manuscript you have drawn a neat little plan of the study. If you had it with you this minute you would see that – the chair being drawn out in the position indicated by Parker – it would stand in a direct line between the door and the window.’ (298)

The movement of the chair by the murderer proves highly significant when Poirot finally reveals that the table had contained a dictaphone, with the murderer using a recording of the dead man’s voice to misdetermine the time of death. The map as given in the book by the narrator is thus actively misleading for the reader. Again, like the manuscript in which it is drawn, it partakes of a double game on the part of the murderer, one of apparently providing innocent and helpful responses to aid Poirot (and the reader) while he is actually omitting or eliding vital information as a deceptive act of self-protection. The narrator presents the room as he claims to ‘remember’ it, apparently not seeing the significance of the shifted chair, when in fact it was he who moved it in the first place. As these two examples show, the ability of literary maps to ‘lie’ and our willingness to grant them an authority beyond themselves, is brought into play in Agatha Christie’s use of the house plan in two ways. Their presence leads us to assume that they are trustworthy not least because they are expressly ‘readerly’ in terms of how they are presented. They function as if they are tangible clues to which the reader has direct access, whilst at a narrative level they seem to exist to help us to interpret the text at points where spatial knowledge bears upon character and event. However, they provide no real help and obfuscate rather than illuminate. The gullibility of the reader, and the lure of apparently ‘empirical’ evidence takes us in.

Human Geometry Detective fiction often pushes spatial and temporal contiguity to its limits. Christie’s novels provide extreme examples of this: situations in which a number of individuals who knowingly or unknowingly share a common element are brought together in a particular place at a particular time. So, for example, in Murder on the Orient Express and Ten Little Nigger Boys (retitled And Then There Were None) a spatial network has to be uncovered – in the

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

152

Mapping Murder

latter by participants who are unwittingly brought together but have all committed murder in a way that ordinary forms of justice cannot touch; in the former, by the detective who must first realise that there is such a network and then find the common point of interest located in the past (the death of a child).34 In both these cases as well, the physical situation is deliberately isolated – an island cut off from the mainland, a train stuck in the middle of a blizzard – so that normal structures of civilisation and authority cannot be drawn upon. In such examples, the interplay between physical space, the act of murder, and past narratives that provide human motivation for present acts creates a complex social network at the core of which lies the solution. Of course, such social networks are present in all forms of fiction (as in real life) but in detective fiction, where there is a strong need to determine a particular meaning emerging from that web, the network is brought to the fore. One way in which one can analyse spatial meaning in such texts further, then, is by means of what might loosely be called ‘human geometry’ a geometry of everyday life that the form of the map makes explicit. In the final part of this chapter, then, I want to consider the ways in which the presence of the map also draws attention to such social and communicative networks, particularly in the traditional ‘Golden Age’ setting of the village. In a helpful article on ‘Interwar Detective Fiction and the Village Community’, K. D. M. Snell notes: The house-by-house cartography forms a grid for the plot in a way rarely found in nineteenth century fiction. The sense of place has become very minute indeed. One could explore a generic cartography in the stereotypical village layouts of these novels, and these divulge the hope or reality of certain kinds of community life.35

There is surely also a gendered dimension to human geometry when it relates to the space of an interwar village. In an article that argues that the figure of the spinster covertly questions power and gender structures whilst upholding the status quo, Kathy Mezei reminds us that for much of the late nineteenth and early twentieth centuries there was a disproportion between male and female survivors of wars, creating a generation of single women.36

34

35

36

A map of the train carriage is provided for Murder on the Orient Express (London: William Collins & Sons Co. Ltd, 1934) and Christie uses the same device as in Styles when the map fails to provide the crucial physical detail within a carriage that unravels the mystery. K. D. M. Snell, ‘“A Drop of Water from a Stagnant Pool?”: Inter-War Detective Fiction and the Rural Community’, Social History 35.1 (2010): 21–50; 26. Kathy Mezei, ‘Spinsters, Surveillance, and Speech: The Case of Miss Marple, Miss Mole and Miss Jekyll’, Journal of Modern Literature 30 (2007): 103–20. Mezei records ‘the 1851 census

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Human Geometry

153

Christie indirectly refers to this in her Autobiography when she describes Miss Marple as partly modelled on ‘old ladies whom I have met in so many villages where I have gone to stay as a girl’ (449). Female involvement in village life enables a certain kind of knowledge and connectedness to come into being, with the figure of the spinster placed in a unique, if paradoxical, position: situated as an instrument of surveillance precisely because of her marginal and indeterminate position. Her narrative function, in representing the dialectic between seeing and being seen, omniscience and invisibility, often mirrors the ambiguous and hidden role of the author/narrator.37

This kind of position is, of course, exemplified by Christie’s second famous detective, Miss Marple, who first appears in a book-length work in The Murder at the Vicarage (1930).38 In An Autobiography Christie states: I think it is possible that Miss Marple arose from the pleasure I had taken in portraying Dr Sheppard’s sister in The Murder of Roger Ackroyd. She had been my favourite character in the book – an acidulated spinster, full of curiosity, knowing everything, hearing everything: the complete detective service in the home . . . I liked the part she played in village life: and I liked the idea of village life reflected through the life of the doctor and his masterful sister. (448)

A geometric form is concerned with relative positions and the dynamic between each of them as well as relations between part and whole. The way in which ‘village life’ is ‘reflected through’ the spinster, who is also a part of it (but marginalised and overlooked) is particularly relevant here since it makes clear the highly spatial and communal nature of Miss Marple’s method. Her intimate knowledge of the relations between individuals within a community enables her to make larger connections and deductions out in the wider world for those she does not know. This method is a direct consequence of her social situation. Certain forms of behaviour that ultimately prove useful to a detective result from her social marginalisation: Extreme forms of focalization – surveillance, spying and voyeurism – and of narration – eavesdropping (the verbal equivalent of spying) – foster misdirection

37 38

which listed 405,000 more women than men and . . . the post-World War One situation in which there were 1,098 women for every 1,000 men’ (104). Mezei, ‘Spinsters, Surveillance, and Speech’, 104. Miss Marple actually first appears in a short story ‘The Tuesday Night Club’ in The Sketch Magazine (1926). This story became the first of a sequence later published together as The Thirteen Problems (1932) told by different members of the village community (and all of which Miss Marple quietly solves).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

154

Mapping Murder

and misrecognition; they represent not only devices to further the plot, but also modalities of control for those without power.39

One striking thing about The Murder at the Vicarage as the first Miss Marple novel is the negative way in which she is presented.40 She is introduced indirectly, and unflatteringly, by the Vicar’s wife as one of many old ladies: ‘She’s the worst cat in the village,’ said Griselda. ‘And she always knows every single thing that happens – and draws the worst inferences from it’ (10). Even Miss Marple’s method is subject to unsubstantiated prejudice. After questioning her, the Chief Constable in charge of the case comments that: ‘I really believe that wizened-up old maid thinks she knows everything there is to know. And hardly been out of this village all her life. Preposterous. What can she know of life?’ I said mildly that though doubtless Miss Marple knew next to nothing of Life with a capital L., she knew practically everything that went on in St Mary Mead. (76)

To some extent, of course, this portrait is true; Miss Marple is used to spending her time carefully observing and analysing the characters and actions of those around her in a way that is not entirely healthy. It is the act of murder as a significant event that renders such skills invaluable, validates her natural inclinations and makes a virtue of them. Miss Marple is an expert in human geometry. She understands individual and connective forces and motivations so well that she can successfully extrapolate from the particular to the universal. In her own account of her ‘hobby’, she herself acknowledges this: ‘You see,’ she began at last, ‘living alone, as I do, in a rather out-of-the-way part of the world, one has to have a hobby . . . my hobby is – and always has been – Human Nature. So varied – and so very fascinating. And of course, in a small village, with nothing to distract one, one has such ample opportunity for becoming what I might call proficient in one’s study. One begins to class people, quite definitely, just as though they were birds or flowers, group so-and-so, genus this, species that. Sometimes, of course, one makes mistakes, but less and less as time goes on . . . But I have always wondered whether, if some day a really big mystery came along, I should be able to do the same thing. I mean – just solve it correctly. Logically, it ought to be exactly the same thing. After all, a tiny working model of a torpedo is just the same as a real torpedo.’ (218–19)

Miss Marple’s method is thus in many ways a feminised version of earlier male detectives’ deductive powers. When presented only with the end

39 40

Mezei, ‘Spinsters, Surveillance, and Speech’, 107–8. Agatha Christie, The Murder at the Vicarage (London: W. Collins and Sons Co. Ltd, 1930).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Human Geometry

155

result, individuals endow Holmes with extraordinary powers, although these are in fact the logical outcome of a causal line of reasoning based on detailed observation. In a similar way, Miss Marple is sometimes treated as if she has almost preternatural abilities, when all she is doing is being absolutely systematic in her thinking, in her knowledge of different psychological motivations and in her attention to detail. As the Vicar acknowledges: ‘There is always some perfectly good and reasonable explanation for Miss Marple’s omniscience’ (235). Like Holmes, she draws upon an encyclopaedic knowledge of human nature which is frequently referenced back to personal quirks of particular members of the community, but is never inaccurate. The murder itself is that of Colonel Protheroe, shot in the Vicar’s study. Events unfold through the relatively limited perspective of the Vicar as narrator who, whilst intelligent and capable of making logical inferences, has certain blind spots (such as an anxiety about his wife’s feelings for him) that affect his judgement. This novel also contains no less than three maps and I want to suggest that this is by no means a coincidence. Mezei notes that ‘houses serve as symbols and settings of the community and play central roles in the novels’ (105) whilst Snell points out that ‘In the course of the murder and its detection, the internal structure of the village becomes important, with its vantage points and angles of perception’ (26). The house and village plans in this book provide us with an excellent example of the way in which literary maps can function to support character as much as narrative and draw attention to social, verbal and visual patterns of connection (see Figs. 4.10–4.12). These maps are less to do with plot or solution (although they do contribute to these) and far more to do with the introduction of Miss Marple herself and her physical and social place within the community. This proves to be crucial to both the staging and the solving of the murder. In Murder at the Vicarage, the first two maps are explicitly presented to us by the Vicar as his own. The first map shows the main murder site of the Vicarage and its environs; the second gives a more detailed plan of the room in which the murder occurred (the study). For the first map, the Vicar tells us ‘I append a rough sketch here which will be useful in the light of after happenings, only sketching in such details as are necessary’ (26) and, for the second, ‘For the convenience of my readers, I append a second plan of the room’ (43). Characteristically, then, the maps are made by the narrator, not the detective, and for the reader. PLAN A is ostensibly provided to aid the reader’s understanding of movements in relation to the murder. Key information on the map concerns the position of the study

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

156

Mapping Murder

Fig. 4.10. Plan A from Murder at the Vicarage

Fig. 4.11. Plan B from Murder at the Vicarage

in relation to the studio in the garden (which is occupied by an artist and visited by various females) and in relation to front and back entrances to the Vicarage. But of as much significance is the location of the Vicarage next door to Miss Marple’s house. Her position within her own garden allows her to monitor the back lane and the path from Old Hall. The first

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Human Geometry

157

Fig. 4.12. Plan C from Murder at the Vicarage

map therefore bears upon the key event of the murder itself, but also strongly upon Miss Marple’s initial status within the book as a major witness and, more than this, upon the way in which she works as an amateur sleuth and expert in ‘the art of seeing without being seen’ (189). From Christie’s point of view, the book might as well have been called ‘Murder in the house next door to Miss Marple’. PLAN B is a fairly standard Christie device that occurs when she gives two maps at different scales. In this case it has significance in two ways. First, it explains how Mrs Protheroe could claim to have glanced in the room through the window (top left) and not seen her dead husband at the writing table (although this is later revealed to be a lie) and, second, it contains the important, though apparently minor, detail of ‘Tall stand with pot on it’ (which supposedly obscures the view). In a way that echoes Poirot’s revelation about crucial papers hidden on the mantelpiece in The Mysterious Affair at Styles, it is through absent-mindedly fiddling with the Vicar’s plant and reminding him to water it, that Miss Marple realises that the murder weapon must have been hidden in the plant pot by an accomplice, then taken out by Mrs Protheroe and used to shoot her husband through the window. Apparently insignificant objects on the room plan turn out to have a highly significant function in helping to solve the crime. PLAN C is immediately of interest because of the way that it is presented silently (without attribution) within the text. It is not given to us by the narrator but, by implication, by the author (although the plain and neutral

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

158

Mapping Murder

style remains the same).41 This third map presents the village of Mary Mead with all the key characters in their relative positions.42 In terms of the murder, it adds little information beyond that given in the first map. Thus, its primary role is to present neighbourly relationships and proximities but, again, in relation to Miss Marple. It functions to situate her not just alongside the Vicarage, but at the heart of the village network. Both the first and the third maps thus also make explicit the way in which angles of vision are of great importance to village life, as Snell signals:43 In this predestined process the morally apposite role of nucleated village structures seems to be indicated, where people necessarily overlook each other in a matrix of house and garden viewpoints, highly conducive to gossip, effective observation and inevitable detection.44

The map presents spatial and geometric patterns of buildings but these in turn are vitally connected to the sight and vision of characters whose actual and social positions predispose them to be surveillant and then to translate what has been seen into positive or negative verbal forms (such as rumour, gossip or, in written form, as poison pen letters). This map is given within the text at the same time that Miss Marple is responding to the inspector’s questions and it makes clear that, in spite of her social marginalisation, she is central to the murder inquiry. He opens his meeting with her by stating that: ‘“it seems possible that owing to the position of your house and garden, you may have been able to tell us something we want to know about yesterday evening”’ (70). In response, Miss Marple freely admits: ‘“As a matter of fact, I was in my little garden from five o’clock onwards yesterday, and, of course, from there – well, one simply cannot help seeing anything that is going on next door”’ (70). At a physical and spatial level, she is the individual with the best ‘line of sight’ because of the location of her house, but that vision applies both literally and metaphorically. When we pay really close attention to lines of vision in relation to Miss Marple, two significant, related, points emerge. The first is the fact that she possesses an active desire to be all-seeing and all-knowing to such a degree

41

42 43

44

Representational slippage in terms of who purports to make the literary map – the author or a character – is not uncommon. (See the next chapter for more discussion of this in relation to Milne and Ransome.) An interactive version of this map can be found on the internet at www.poirot.us/map.html. In her Autobiography, Christie flags the importance of the village for this book: ‘The village is as real to me as it could be – and indeed there are several villages remarkably like it even in these days’ (449). Snell, ‘“A Drop of Water”’, 27.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Human Geometry

159

that everyone else is aware of it. The Vicar notes: ‘Miss Marple always sees everything. Gardening is as good as a smoke screen, and the habit of observing birds through powerful glasses can always be turned to account’ (19). He also describes returning home and passing ‘the danger point of Miss Marple’s garden’ (26) created by the proximity of her cottage to the Vicarage. The second point, far less obvious and barely explicit within the text, is that this desire to be in command of knowledge as a result of her physical location in the village and her natural disposition is actively manipulated by the murderers and forms an essential part of their plan.45 In other words, they seek to use her all-seeing eye against her. This becomes clear as the account of movements into and around the Vicarage at the time of the murder are repeatedly replayed, with Miss Marple as a key witness. The two murderers (Lawrence Redding and Mrs Protheroe) are fully aware that: ‘“If there was anything to be seen yesterday evening Miss Marple saw it”’ (133); and for this very reason, the murderess walks past her in full sight without carrying a handbag. Miss Marple’s own witness statement then becomes part of the alibi. She affirms the apparent innocence of Mrs Protheroe because she asserts that she could not have been carrying a gun with which to shoot the victim: ‘“Well that she didn’t do,’ said Miss Marple, with unexpected decision. ‘I can swear to that. She’d no such thing with her . . . She wasn’t carrying a handbag.”’ (74). Later, when Miss Marple sees the plant pot in which the murder weapon was hidden she suddenly realises how she has been manipulated: “Oh dear,” she muttered to herself. ‘“I have been stupid. So that was it. Perfectly possible all the time.”’ (224). The murderers attempt to manipulate her in one way (in terms of her visual acuity) but significantly underestimate her in another (inner vision and ability to ‘read between the lines’). At a narrative level, Miss Marple is introduced into the narrative as if she is a tangential figure when in fact, as the investigation unfolds, it rapidly becomes clear that she is absolutely central: the only person really capable of making sense of all the facts and with the abilities required to solve the crime. The maps in The Murder at the Vicarage thus function almost as an objective correlative for Miss Marple herself – appearing to be marginal but actually being central (and able to be central in part as a result of that marginalisation).

45

Miss Marple does realise this and comments on it when she solves the crime: ‘“[S]he passes my garden and stops and speaks, so as to give me every opportunity of noticing that she has no weapon with her and also that she is quite her normal self. They realized, you see, that I am a noticing kind of person”’ (242).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

160

Mapping Murder

Fig. 4.13. Street plan from More Work for the Undertaker

I want to conclude by setting alongside this analysis a second example of the role of ‘human geometry’ in detective fiction in the work of Margery Allingham. Allingham uses maps less frequently than Christie (in five of her twenty-one Campion novels) but to great effect, as we can consider here. In More Work for the Undertaker the detective, Albert Campion, is invited to investigate ‘The Palinode Case’ in which one member of a decayed, genteel, eccentric, family (Miss Ruth) has been murdered and the others are either the murderers or may be under threat themselves.46 The map (see Fig. 4.13) is a street map showing a small part of London around the Palinode house. It appears to be a straightforward local plan with the names of different businesses helpfully marked upon it. However, as the story develops, it becomes clear that all of these businesses are bound by strong ties to the family. When the Palinodes were wealthier, they supported their community by always employing locals as their solicitors, lawyers etc. so that, in the present, even though they no longer have any status or money, they still command considerable respect from those around them and function as a kind of feudal ‘squirearchy’ (114). Miss Jessica tells Campion that: ‘“My father believed in

46

Marjorie Allingham, More Work for the Undertaker (London: William Heinemann, 1948).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Human Geometry

161

employing local people, they may not be the best, he said, but they are one’s own”’ (114). When Campion smiles at this and remarks, ‘“I was thinking it was rather a large town to be parochial in’, she replies that ‘London is made up of many villages”’ (114). The neutral statement of anonymised place that the map at first appears to present is, in fact, nothing of the kind. Instead it actually represents a community in which all kinds of tension of ownership, debt and favour are in play. This is given a sinister twist by the presence of an undertaker using the Palinode basement as a storage place for his coffins and clearly running some kind of smuggling business on the side. In this book a sense of unseen connections, just out of reach, is manipulated by Allingham to create a plot in which highly unpredictable events repeatedly occur that no one – reader, policeman or detective – is immediately able to understand. For instance, the police detective goes to talk to the chemist as part of a routine inquiry but, before he can ask any questions, the chemist takes a substance to kill himself, in what appears to be a troublingly disproportionate response. This is because all of the local characters, who appear only to have a literal physical spatial connection to each other, are in fact vitally connected through criminal activity or through the Palinode family. The spatial configurations that are always in play for a murder are here also played out again at a secondary level that may or may not have anything to do with the case. Allingham thus interweaves rich and complex layers of connection and at times shifts between these layers in a self-conscious way: Campion walked up Barrow Road. It was the first break he had had in which to give his mind to the various brightly coloured threads that made up the puzzle. He walked for a long time considering each strand in the tangle, following each loose thread as far as it would take him. He was still a long way from the solution when he turned aside to plunge into the web of small roads which must take him back to Portminster Lodge. As he stepped out from the pavement to cross over a narrow side street a ramshackle truck . . . was advancing towards him . . . Its sudden murderous swerve towards him astounded him, even as his instinctive leap saved his life. (123)

As the detective moves within the physical structure of the streets (‘the web of small roads’) with his mind abstracted by the intellectual structure of interrelated motives and movements (the ‘various brightly coloured threads’ that recall the origins of ‘clew’), he is almost taken out of play because he has not yet realised that the ordinary working men and women all around him are the threat and that the ‘web’ of streets is itself the

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

162

Mapping Murder

solution to the crime.47 Meanwhile, the subplot of the book’s title – that of the undertaker smuggling criminals out of the country in coffins – provides a negative manifestation of the network, centred on the literal/metaphorical concept of ‘Going Up Apron Street’ (being smuggled out in a coffin as if dead/a metaphor for death itself ). The novel thus gradually uncovers a complex structure of physical, psychological and emotionally loaded spatial relations and, ultimately, it is exactly this structure that provides the clue to the murder. The murder victim, Miss Ruth Palinode, had willed apparently valueless shares to a large number of local people, including the bank manager. When he realises that some of these shares are valuable, he murders her and then, when he thinks he has been discovered, he uses the coffin route to escape. The climax of the book involves a crazy chase to stop the coffin wagon from ‘Going Up Apron Street’ in order to literally uncover the first network and intellectually solve the second. So human geometry – a hidden network of loyalties and indebtedness (once healthy and positive but increasingly subject to manipulation and corruption) underpins the straightforward geometry of the street plan and ultimately both enables and solves the murder (as with The Murder in the Vicarage). ‘Human geometry’ is shown to function as a powerful spatial element in crime and detection, not only in the capital cities of early detective fiction but even more powerfully in the internecine spaces of village life. Those who can see through the mists of banality must either have superior powers or, in the case of Miss Marple, be both part of that world and able to stand apart from it. The actuality of crime and detection comes down to everyday spaces – the kitchen, the bedroom – and small objects (the plot plant, the vase). But that very ordinariness becomes part of the cover used by the warped psyche, relying on dullness to mislead the ordinary mind. The map in detective fiction is able to manipulate through its perceived authority, neutrality and objectivism. The primary message it projects is that you can trust it. In a literary form that is fundamentally concerned with who can or cannot be trusted, and with the search for truth amidst many lies, this is no small claim. At the same time, however, whilst its function may vary, the map always possesses a doubled identity. This may manifest itself in terms of which way it faces (a doubled audience within and beyond the text) or in terms of the layers of meaning it conveys, or

47

The clue (or ‘clew’) finds its origins in the classical tale of Theseus and the Minotaur as a literal, physical object (the thread used by Ariadne) that enables safe negotiation through unknown territory in the dark.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

Human Geometry

163

both (a doubled doubleness). The map’s primary ‘surface’ meaning is selfevident but its secondary connotative meanings vary considerably and, as I hope this chapter has illustrated, form a fully integrated element of interpretation that can deepen and enrich understanding of spatial meaning for the genre.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.005

5

Playspace: Spatialising Children’s Fiction

Qu’il fasse lui-même la carte de tout cela. (Let him make a map of all of these things himself, a very simple map.) Jean-Jacques Rousseau, Émile

The previous two chapters on adventure, spy and detective fiction have argued that the presence of the map is an integral part of newly emerging genres in the latter half of the nineteenth century and that approaching a generic form spatially, by means of the explicit visualisation that a literary map provides, can help illuminate both critical debate and interpretation of theme and content, particularly in relation to spatial meaning. This chapter adopts a slightly different approach because of the nature of the genre involved. Rather than being centred directly on the function of the map– text relationship, it offers a broader approach to spatialising the genre. The identity and emergence of ‘children’s fiction’ as a discrete entity is far less determined historically and generically than for other genres. As always, the map functions to bring spatial meaning to the fore and is fully integrated with the meaning of the work. However, in a way that is not true for other popular genres considered in this book, the spatial can also function in a larger sense as a generically determining element. In other words, one way in which we could determine a work as belonging to the category of children’s fiction is by means of a particular spatial dynamic between adult and child and the nature of place and space that develops out of it. This chapter therefore sets out to articulate a new concept of ‘playspace’ for the genre. Where maps occur, this space is represented by the map but it also extends beyond it to determine the larger spatial nature of the genre or – it could be argued – to redetermine the genre as spatial.1 In so doing, playspace partly sets itself against a still-dominant deconstructive critical mode that postulates that the entire generic form is a kind

1

164

For other recent spatial interpretations of children’s fiction, see: Maria Sachuko-Cecire, Hannah Field, Kavita Mudan Finn and Malini Roy, Space and Place in Children’s Literature, 1789 to the Present (Aldershot: Ashgate, 2015); Joseph. L. Zornado, Inventing the Child: Culture, Ideology and the Story of Childhood (New York and London: Garland, 2007).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Spatialising Children’s Fiction

165

of willing adult delusion; the ‘child’ a projection based on adult need. Playspace takes its origins from the same Rousseauvian account as that of the sceptical Freudian reading (see below), but finds instead a positive alternative in the necessary framing of child experience by the adult when that relationship is understood spatially (and embodied in a shared map form). Playspace is therefore an abstract form that the map visualises and embodies but it is also a dynamic and co-created space shared by adults and children over time. After a brief consideration of the importance of place, space and maps to the generic identity of children’s fiction, the chapter turns to an exemplary case study in the form of Arthur Ransome’s Lake District Series, which allows us to understand playspace in action across a series of books set in the same location for which the repeated image of the map is central. The second half of the chapter then builds on this account to argue against the dominant sceptical reading of children’s literature and in favour of a safe, shared space created by a positive dynamic between adult and child as a vital constituent of the genre. The final part of the chapter considers the equally powerful effect of a loss, or negation, of playspace and the relation of this to the act of mapping.

Spatialising Children’s Fiction Children’s fiction has always had a problematic identity in terms of genre because it does not contain obvious innate generic characteristics, but is defined primarily by readership. Is it to be understood as books that can be read to children; books written specifically for children, or should it consist only of books written by children? Various attempts to create a distinct identity for the genre have been made.2 However, in 1984 Jacqueline Rose’s

2

A distinction between works read to and by children is convincingly made by Felicity A. Hughes in ‘children’s literature: Theory and Practice,’ ELH 45 (Autumn 1978): 542–61. She argues that the move from a larger category of ‘children’s literature’ (that would include Robinson Crusoe and Gulliver’s Travels, say) to that of ‘writing for children’ is caused by the disassociation of the novel from family reading in the late Victorian period. Barbara Wall, in The Narratior’s Voice: The Dilemma of Children’s Literature (Basingstoke: Palgrave Macmillan, 1991), determines children’s fiction in terms of: single address directed solely at the child reader; a double address involving a self-conscious awareness of adult and child separately; a rare dual address managing to address both adult and child together. Crucially, a work can attract a dual readership without having a dual address (36); see also ch. 2, 20–36. For a thorough account of different positions see Karin Lesnik-Oberstein, Children’s Literature: Criticism and the Fictional Child (Oxford: Clarendon Press, 1994), ch. 5, 131–64.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

166

Playspace: Spatialising Children’s Fiction

deconstructive argument in The Case of Peter Pan: The Impossibility of Children’s Fiction took this core problem and essentially exploded the claim to any generic identity when she argued that ‘children’s fiction is impossible’.3 For Rose, ‘There is no child behind the category “children’s fiction” other than the one which the category itself sets in place, the one which it needs to believe is there for its own purposes’ (10). At the core of her critique is the idea that the fictional child is treated as if it is something separate and self-contained, rather than held in language and subject to it. Fiction for children is seen as a product of adult need, projected onto the figure of the child as holder of ‘a primitive or lost state to which the child has special access’ (9). Ultimately, this is because of the nature of representation and the medium through which such fiction must be communicated: ‘language is not innocent’ (73). Looking back to Rousseau (the origin point for a false origin point) Rose claims: [T]his notion of cultural decay produces two related assumptions. First, that the child – if we get to him or her soon enough (or keep away long enough) is where this sensibility is still to be found (childhood is therefore defined as something which exists outside the culture in which it is produced). (43–4)

The false assumption here is that the innocence of the child can save culture by and through a return to nature. The child is given a weight of responsibility for adult needs. This then works at a larger level for literature itself when it is understood as a means of giving back ‘something innocent and precious which we have destroyed’ (45). Like Rose, this chapter also explores a Post-Romantic model of childhood with origins in Rousseau and also draws on Peter Pan. However, in all cases it does so to spatialise the adult–child relationship, in an attempt to partly counter the sceptical account above. This is achieved by shifting the focus from representation of the child (by the adult, for the adult) to the shared space of childhood play. Of course, this is still represented by the adult writer through language, but it reveals that children’s fiction constructs a self-conscious, bounded space for the child in a way that is not unconscious but explicitly knowing (by the adult within the text and outside it). The common presence of the map in many canonical texts of Children’s Fiction signals the importance of this space for the genre. Not only that, but a core issue at the heart of critical debates – the relationship 3

Jacqueline Rose, The Case of Peter Pan or The Impossibility of Children’s Fiction (Philadelphia: University of Pennsylvania Press, 1984), 1.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Spatialising Children’s Fiction

167

between child and adult reader, child and adult writer – can be interpreted in terms of a spatio-temporal dynamic that the map often makes explicit.4 When we focus exclusively on spatial elements, three concepts are consistently present. First, children’s fiction is frequently set in a rural or pastoral setting that looks back to a Wordsworthian (or Rousseauvian) ideal. Second, there is an almost omnipresent motif of the journey, often constructed as both physical and metaphorical (leading to growth) and frequently involving a clear end goal. This in turn brings certain narrative structures into play: linearly, the structures of the road and the river, but more commonly a circular structure that determines a safe return home at the resolution of the narrative (‘There . . . and back again’) and on completion of a quest. In relation to the life journey, since the characters are still young, mapping itself becomes a powerful tool for development and growth as Carol Billman testifies: ‘It is in coming to terms with place – that is, by mapping – that they [children] not only accept the universe created in the story but also master it.’5 This results in what Peter Hunt calls ‘an active setting’, that is: ‘The meanings of the landscapes themselves provide a subtext for the journeys: places mean’.6 Third, there is an emotional core which is also strongly spatial in nature: the desire to find, return or positively redetermine the concept of home. The greatest unspoken fear of children’s fiction is that home will no longer exist or that one will be unable to return to it. To some extent, then, I accept the deconstructive account of children’s literature as offering an idealised construction of the child and childhood, fed by underlying ideologies that are not fully recognised by author or reader, and I would accept that such idealisation partly extends into representations of place and its occupation. However, it is notable that Rose’s argument dwells on a particular aspect of the adult–child dynamic in order to problematise it: ‘A rupture almost, between writer and addressee. Children’s fiction sets up the child as an outsider to its own processes’ 4

5

6

Playspace could also be interpreted as a Bakhtinian chronotope or ‘time-space’. It determines children’s fiction as a shared chronotope between adult and child and one which is never truly left (as children becomes adults and have children of their own). See Mikhail Bakhtin, ‘Forms of Time and of the Chronotope in the Novel’, in The Dialogic Imagination: Four Essays by M. M. Bakhtin, ed. Michael Holquist, trans. Caryl Emerson and Michael Holquist (Austin: University of Texas Press, 1981), 84–258. Carol Billman, ‘Reading and Mapping: Directions in Children’s Fantasy’, Children’s Literature Association Quarterly 7 (1982): 40–6; 43. See also Deirdre F. Baker, ‘What We Found on our Journey through Fantasy Land’, Children’s Literature and Education 32 (2006): 237–51. See Peter Hunt, ‘Landscapes and Journeys, Metaphors and Maps: The Distinctive Feature of English Fantasy,’, Children’s Literature Association Quarterly 12.1 (1987): 11—14; 11.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

168

Playspace: Spatialising Children’s Fiction

(The Case of Peter Pan, 2). The focus on adult writer and child reader leaves out other aspects of the adult–child dynamic that bear upon this relationship in more positive and shared ways over time. In contrast to Rose, then, I am interested in the shared past/future space that is created between children in the narrative and the adult writer and between child readers/ adult readers and the text, or the map. In a useful article that aims to define a ‘literary children’s geography’ Jenny Bavidge states that this has to start from the basic point that children’s literature does not, of course, represent a child’s view of the world at all. What it does represent is a privileged space in which we witness the operations of adult dialogues with children.7

Bavidge also notes, following Rose, that ‘children’s literature can never be an innocent interaction’ but that children’s geography ‘offers us versions of the world which, at most, we might claim coincide with children’s own readings of their world’ (321). I want to go further and suggest that multiple spaces in the past and present coexist for adult and child readers and that it is necessary that they do so. The focus for this chapter, then, is on the space and place of children’s fiction as a shared or doubled space for both adult and child, inside and outside the narrative. ‘Playspace’ is an environment requiring the right level of adult distance and proximity to enable exploration and growth, but also to ensure the safe return home. In terms of readership, it allows for the presence of an adult at the margins as either creator (writer and implicit narrator) or mediator (reader) and is likely to map onto Barbara Wall’s concept of ‘double addressee’. At a level of representation within the fictional world, playspace is experiential, releasing the children held within it into a state of self-sufficiency that enables growth without endangerment. Also essential to the nature of playspace, then, is a concept of intergenerationality and hope for the future. Adult and child occupy the same space but at different times concurrently, as the child of one generation becomes adult for the next generation. In this way, a nostalgic return to imaginary worlds also comes to form part of a larger cultural indoctrination via the ‘classics’ of the canon and works to perpetuate a certain model of Englishness in and through shared literary culture (as children grow up and reread the literature as adults to their children, and so on). This is particularly true for places in which fictional texts and their maps are 7

Jenny Bavidge, ‘Stories in Space, The Geographies of Children’s Literature’, Children’s Geographies 4 (2006): 319–30; 321.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

The Lakes As Playspace

169

projected out of, and then back onto, a landscape (e.g. ‘The Hundred Acre Wood’ of Winnie-the-Pooh) and create a strong sense of national identity or an idealised model of English childhood.8

The Lakes As Playspace I want to illustrate the concept of playspace at a level of fictional representation as embodied in stories of English childhood holidays in Arthur Ransome’s Lake District series. A controlling spatial conceit is central to the way in which both Ransome and his child characters create an imagined space within a real place, but also to the linkage of map to text, which extends across the series as we shall see below (with different adventures played out around the same mapped space). The landscape of the lake and the dynamic between children and adults allows and enables a certain kind of spatial experience. As Peter Hunt puts it: The hills that ring the lake . . . are a secure boundary, rather like the hills that enclose the Thames valley in The Wind in the Willows . . . Consequently, the children can play and grow within known limits.9

The books are set in a semi-imaginary place, itself nested within an area of England that is both an iconic and a nostalgic landscape (fixed in time by virtue of being a national park) and one that is famous for being a contained region that nonetheless holds points of danger within it (i.e. you can die by falling off Striding Edge). Hunt observes: There is a restricted learning area, and codified games, and, as with many other children’s books, the landscape acts as an encompassing and finite background . . . Swallows and Amazons thus provides both immediately recognizable incidents and wish-fulfillment within quite a subtle framework of protected freedom.10

Hunt succinctly describes a state in which ‘the children securely “live apart”’ (27) that strongly recalls the Rousseauvian model (considered in greater detail below) and I want to explore this concept further here at a level of represented place and space. 8

9 10

The map and many elements within the book correspond to a real landscape, Five Hundred Acre Wood in Ashdown Forest in East Sussex, close to where A. A. Milne had a summer home at Cotchford Farm. On the map, however, it is clear that ‘One Hundred Acre Wood’ does not apply to the whole area but only to a specific part in which Owl lives. Peter Hunt, Approaching Arthur Ransome (London: Jonathan Cape, 1992), 90. Peter Hunt, ‘Ransome Revisited: A Structural and Developmental Approach’, Children’s Literature in Education 12.1 (1981): 24–33; 26, 27.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

170

Playspace: Spatialising Children’s Fiction

In terms of narrative and plot structure, the first book, Swallows and Amazons, establishes a template that is then replicated with variations across the series. Two (or three) groups of children are released within the circumscribed landscape around the lake; adults are present yet peripheral; a base camp is established at a suitable distance from the adults but within reach; there is a playing-out of a meta-narrative (‘arc narrative’) that they all participate in; and finally there is the possibility of real danger as a consequence of independence (which problematises the structure). The focus of Ransome’s particular ‘playspace’ is actually always topographical – centred upon the lake itself with different kinds of action occurring on or around it and a new arc-narrative constructed in relation to it, according to season and other factors. Ransome also often cleverly makes good use of a slightly decentred temporality within the holidays so that books are rarely situated directly in the middle of the vacation period (when the children can be assumed to spend all their time sailing on the lake, simply repeating the experience of the first book), but either represents an unexpected extension at the end of the holiday period (Winter Holiday) or anticipates the holiday to come (as in Swallowdale and The Picts and the Martyrs). The latter ends with one of the adults looking forward to a quiet life and Nancy retorting ‘“Well, you won’t exactly have one . . . Not with the Swallows coming, and Uncle Jim, and five whole weeks of the holiday still to go”.’11 In each case the children within the book also themselves determine the arc-narrative for play that allows them to work together towards a clear goal that is in, and of, the landscape and to act out assigned roles in order to do so. In the first book when these conventions are established, the primary identity is as sailors/voyagers. However, for Swallowdale they become explorers and mountaineers; for Pigeon Post, prospectors seeking out a mine; for Winter Holiday, Arctic explorers racing for the North Pole. The structures and key elements remain the same across the books and are essentially repeated with variations. Ransome creates a child-led environment in which adult authority is within reach, but not inhibiting or surveillant, enabling a sense of place to exist that easily slips between the actual and the imaginary or literary. Thus, in a way that is also highly characteristic of children’s fiction more broadly, the doubled perspective of adult and child reader are both in play within the represented world of the text, with the latter nested within the former. On the one hand, adults seem to be pushed to the physical margins. 11

Arthur Ransome, The Picts and the Martyrs: or Not Welcome at All (London: Jonathan Cape, 1943), 305.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

The Lakes As Playspace

171

This is essential for the children’s play to function as they want it to, with the necessary level of independence and freedom. On the other hand, the adults hand, the adults are always close by, intervene if necessary, and sometimes act as negative catalysts for the creation of the playspace (as with the Great Aunt in The Picts and the Martyrs). The imaginary world of the arc-narrative is made accessible to only a few chosen adults. But always it is the integration of closeness to/distance from adults, both physical and emotional, that allows the right kind of adult-free play. This is felt most strongly when it is denied. In Pigeon Post, for example, the limited spatial (and therefore imaginative) parameters are made clear when the children arrive at Beckfoot: ‘You know we’re camping in the garden . . .’ ‘In the garden?’ said Titty, rather sadly. ‘Only till your mother comes to Holly Howe. You won’t have Swallow till then, and we can’t all eight of us cram into Amazon. So Wild Cat Island’s no use. And anyhow, while Mother’s the only parent she wants to have us all within reach. She says she’s too busy with paperhangers and plasterers to keep an eye on us if we camp too far from the house.’12

In a chapter entitled ‘Two Kinds of Camping Places’, the children are then allowed to move up the valley and camp out behind a local farm with a rather controlling farmer’s wife in charge. Here they are still too close to adult supervision (the boundaries of playspace are set too tightly): To be camped within hearing of the house and its natives, no matter how friendly . . . To draw water from the farm pump instead of dipping it from lake or beck . . . To have the tents not in a wood, or on the fell, or even in an ordinary field, but in an orchard, with apple and damson trees in their neat rows . . . Why, Mrs. Tyson was quite right, and they might almost as well be in the Beckfoot garden. (103)

Only when they are allowed to move further up the track to their own camp can the children fully inhabit the landscape in the way they want to: ‘But for the first time those holidays they were camping in the wilderness in a camp of their own, free from even the friendliest of natives’ (186). In this way, the tamed wildness of the landscape the children inhabit (high hills within a small circumscribed region) is matched by the degree of freedom they require. At the same time, playspace is relatively unpredictable because it is reliant upon adult liberality: adults with a greater need of authority and

12

Arthur Ransome, Pigeon Post (London: Jonathan Cape, 1936), 24.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

172

Playspace: Spatialising Children’s Fiction

control can destabilise it by upsetting the balance and redetermining the parameters. Indeed, playspace is often negatively determined by an adult who represents a threat to it, or is deeply resistant to that way of existing. The effect is strongly spatial since it creates a need to physically avoid a certain location or avoid being seen by a particular individual in a particular place. In the first book this role is played by Uncle Jim who responds to John in an unfair and hostile way as an unsympathetic adult (wrongly accusing him of theft from his houseboat and sending the police over to the island). Once various misunderstandings have been resolved, he is allowed to re-integrate into the children’s imaginative space on the understanding that he acknowledges his previous failure to do so: ‘Take back your Black Spot, then,’ said Captain Flint. ‘You keep it,’ said Nancy, ‘to remind you never to turn native again.’13

In Swallowdale the Great Aunt performs a similar function to a lesser degree, but this idea is picked up again and played out far more fully in The Picts and the Martyrs. In this last book of the series, Dick and Dorothea come to stay with the Blacketts whilst Nancy’s mother is abroad recuperating from illness. Even before the intervention of the Great Aunt (who takes it upon herself to visit and look after the girls in their mother’s absence, thus necessitating Dick and Dorothea going into hiding) the children were under orders to behave and not to initiate full dramatised play: ‘You do understand, don’t you?’ said Nancy. ‘It’s no good thinking of anything tremendous. It’s got to be plain ordinary life, to show Mother it was perfectly safe to leave us by ourselves.’ (29)

In this case the arc-narrative (which was not supposed to be happening at all) is necessitated and determined by the Great Aunt’s unexpected arrival, compelling Dick and Dorothea to hide out in a hut in the woods as early Britons (Picts) and forcing Nancy and Peggy into the role of perfect nieces (Martyrs). Playspace has to reform itself around the resistant presence – creating an absolute split between real world and play world and the two zones in which these occur. This, in turn, results in a split personality for Nancy and Peggy, signalled most clearly in their changes of clothing and demeanour (as witnessed by the others): Then they both gasped at once. The boat had come into sight below the lawn, gliding down the river. They had last seen it with a crew of wet dishevelled savages.

13

Arthur Ransome, Swallows and Amazons (London: Jonathan Cape, 1930), 297.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

The Lakes As Playspace

173

Now it was being rowed by a girl in a white frock with a pink ribbon round her hair. Another girl, just like her, was sitting in the stern, idly trailing a hand in the water. (99)

The two worlds, rather than having points of easy overlap, with adults who are able to occupy and inhabit both (as in Swallows and Amazons for the most part), must be kept absolutely apart, or one will destroy the other. Adults such as Cook and the Doctor, who are caught in the knowledge of both worlds, are forced to be complicit against another adult in a way they are not happy with: ‘And you haven’t told her anything about us?’ ‘No I haven’t,’ said the Doctor almost angrily, ‘She didn’t ask any questions. But I’ll be in trouble if she finds out just the same.’ (87)

All the books play with the idea of withholding knowledge from the adults (as ‘natives’), of having secrets, but the question is how far to go – or to allow Nancy to make you go. Dorothea anxiously considers: ‘A horrible thought had struck her. What if the doctor and Timothy were right? What if the whole plan was a mistake?’ (86). The major threat to both worlds – the Picts being discovered by the Great Aunt – does in fact occur just as they have been sent out onto the lake to avoid such an eventuality. They accidentally find the Great Aunt on the houseboat when she is thought to be missing on the hills, and row her back to Beckfoot. But although the two worlds are literally brought together – with Dick and Dorothea rescuing the Great Aunt and bringing her safely to shore – the knowledge of their true situation eludes her and she leaves without having discovered the deception. The children display varying levels of conformity to the adult world that bounds them. Nancy and Peggy defy authority to a far greater degree than the other children and have greater freedom as a result. Dick and Dorothea are always on their own, with no parents present, but have strongly internalised values and are sensible and intelligent, so are trusted to look after themselves (although they do put themselves in danger). For the Walkers there is still an unwritten understanding that they must be within call, sight or easy reach of their mother and will not exceed the freedoms permitted them. This is internalised by Susan and, to a lesser extent, John, who become the surrogate parents responsible for policing the boundaries and ensuring the safety of the party. Part of the pleasure of existing in playspace lies in an implicit sense of eternality; no harm will come to you in this endless summer of childhood. However, Susan is never able to fully enter into this state because she is responsible for maintaining it

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

174

Playspace: Spatialising Children’s Fiction

for others. The interim state of the older, more mature child provides an important role. In fact, the children’s zone of protected escapism is, in every book, challenged and threatened by a real-life natural disaster with potentially life-threatening consequences. When this happens, the older children have the potential to ‘go native’ – in other words to respond to the situation from a realist adult perspective. Playspace is threatened when the imaginary dangers and fears inherent in being, say, an Arctic explorer, threaten to become all too real. The younger children and those less responsible are unaware of the danger they put themselves in, even when real danger occurs, because the boundaries between imaginative adventure and those of real life are so fluid and because playspace is protected for them by the responsibility of older children first, and then adults. So, in Winter Holiday for example, when the inexperienced Dick and Dorothea accidentally set off for the top end of the lake in a snowstorm with the potential for real tragedy, Nancy simply enjoys the extreme drama of the situation: ‘This is miles better than anything we planned . . . Sailing to the Pole in a gale of wind and a snowstorm.’ (352). Only when the adults finally arrive at the shelter of the ‘North Pole’ to find them safe does a hint of the real danger involved register with her, as her mother states: ‘We’d have been here before, but, of course, he [Captain Flint] had to race round to stop the search parties . . .’ ‘Search parties?’ said Nancy, ‘Not real ones?’14

When this is confirmed, however, Nancy’s response is not to experience any retrospective anxiety or fear but to express disappointment that drama has been halted: ‘“And you stopped them all” said Nancy regretfully’ (359). The threat of imaginary adventures becoming real is felt most strongly in Pigeon Post, when the younger children are in genuine danger twice over – first from a landslide and then from a forest fire. Ransome manipulates his narrative structure in relation to the climax of each book by splitting the narrative at such points so that the older children are separated off from the younger and the group as a whole is no longer able to work together and protect each other. In the most dramatic moment of Pigeon Post, as the older children rush back to where the younger children were left sleeping, even Nancy feels a sense of responsibility and fear of having failed in the substitute parent role:

14

Arthur Ransome, Winter Holiday (London: Jonathan Cape, 1933), 358.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

A Material and Visual Playspace

175

‘The wood itself may be on fire,’ said John, as he dashed forward with the ashes smoking about his feet. ‘They may be asleep in their tents,’ cried Susan. ‘Oh no . . . No . . .’ cried Nancy fiercely, ‘They aren’t utter galoots . . .’ (352)

Again, these adventures – in each case the dramatic climax of the book – are determined by the degree of freedom allowed the children, even as they also implicitly critique it. If the worst happened, would that level of freedom be considered appropriate, or irresponsible? At the end of Pigeon Post Nancy’s mother leaves her sentence unfinished: ‘“Oh my dears, think what your parents would be saying to me if . . .”’ (372). Ransome is bold in both articulating playspace and in pushing it to its limits.

A Material and Visual Playspace At a level of content, Ransome creates a perfect ground for children’s fiction and one that allows for that world to be open-ended and rediscovered over time. If this is true at a level of verbal representation it is also true visually in that the same space – represented in the iconic endpaper maps – is returned to over and over again by author and characters, but with variations that allow it to be renewed. Maps are vitally integrated with, and dramatically visualise, acts of mapping and re-mapping inside and outside the text. The material nature of the book means that endpaper maps physically frame and ‘protect’ the space of the fictional world held between them.15 In an article on the meaning held within endpaper designs in children’s books, Lawrence Sipe and Caroline McGuire convincingly argue that the endpapers in young children’s picture books form an important part of the overall experience, not just a design element.16 Front and back cover illustrations are ‘the first parts of the interior of the book to be seen when the book is opened, as well as the last to be seen after the story has been read and the book is about to be closed’ (292). The endpaper image has a doubled temporality, created by its material position which bears upon meaning. If the endpaper is also a map then this process of anticipation and

15

16

See Margaret R. Higonnet, ‘The Playground of the Peritext’, Children’s Literature Association Quarterly 15.2 (1990): 47–9; 47. Lawrence R. Sipe and Caroline E. McGuire, ‘Picturebook Endpapers: Resources for Literary and Aesthetic Interpretation’, Children’s Literature in Education 37 (2006): 291–304; 292.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

176

Playspace: Spatialising Children’s Fiction

review is intensified by being focused specifically on the space and place of the text. Ransome’s endpaper maps have a complicated history that can be briefly summarised here. For the first book of the series (Swallows and Amazons) Jonathan Cape hired and paid Stephen Spurrier to illustrate the entire book (including maps on the cover and endpapers), but Ransome disliked the illustrations so much that none were used in the first edition, and only the endpaper map, and map of Wild Cat island, by the illustrator survived. For the second edition of Swallows and Amazons and the first edition of Swallowdale, illustrations were made by the famous engraver, Clifford Webb, but after this Ransome illustrated all the books himself (re-illustrating Swallowdale). As a result, all of the endpaper maps are directly authorial except that for the first book. Stephen Spurrier’s stylised, two-colour map for Swallows and Amazons provides a sharp contrast with Ransome’s own style (see Fig. 5.1). In terms of perspective, Spurrier’s map mingles aerial and bird’s-eye views. The lake and islands are presented as if from above, but the sailing ship on the lake is not, nor are the images of houses on the shore, while the hills in the distance are more or less horizontal. The style is archaic and highly decorative and depicts entirely imaginative elements such as the whale,

Fig 5.1. Endpaper Map from Swallows and Amazons (For the colour version, please refer to the plate section.)

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

A Material and Visual Playspace

177

the octopus in the lagoon and the personified figure of the wind blowing up the lake. Bottom left there is an emblem of Swallows and Amazons with a shield, flags and a man and woman dressed in seventeenth-century piratical outfits. Place names on the map are clear and capitalised, though with the appearance of being handwritten. In various ways the map deliberately merges, or slips between, the ‘real’ landscape of the book and the child’s imaginative reappropriation of it, or seeks to represent both simultaneously. This is seen most clearly in the double naming of ‘Holly Howe’ and ‘Dixon Farm’ also as ‘native settlements’, but also in the subtle suggestion made by the outlining of land around the lake in blue and the presence of the sailing ship, implying that the lake is a sea in the imaginative world of the children. The map thus unites a self-conscious doubling of play and realist occupation of place within the text. We can compare this map directly with Ransome’s map for the second book in the series: Swallowdale (see Fig. 5.2). The first thing to note is the primitivism and realism of his map style in comparison with the ornateness of Spurrier’s. All of the more extravagant elements are stripped away. This supports the comment of Ransome’s biographer, Hugh Brogan, that: ‘He seems to have held that pictures in a story should be there merely to help readers’ visualisation’: they should meticulously depict details of the

Fig. 5.2. Endpaper Map from Swallowdale (For the colour version, please refer to the plate section.)

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

178

Playspace: Spatialising Children’s Fiction

province of the children in the book, not of the readers (who appear to be the primary audience for the endpaper map).17 The map now adopts a far more straightforward bird’s-eye view perspective, with the lake in the foreground being seen from a near-vertical position, but Kanchenjunga in the distance tending towards the horizontal, and with the hills and valleys much more realistically represented. Ransome moves the lake lower down the image to reflect the shift of emphasis on mapping the land in Swallowdale (as opposed to a focus on sailing on the lake in Swallows and Amazons). He keeps the same outline shape of the lake but inverts the map (presumably because the narrative is situated on the far shore). As a result, the compass orientation is also inverted. In spite of differences, however, Ransome did take four core stylistic elements from Spurrier that he then applied to all the maps: the decorative compass rose; the black-and-white edging lines; the colour shading around the edge of the lake; and the limited use of colour combinations. Limited use of colour becomes a highly distinctive feature in terms of presenting the same map across a series since it allows for an interplay between sameness and difference. So, if we compare the maps for Swallowdale (Fig. 5.2) and the next book, Winter Holiday (Fig. 5.3), the first map is in the colours of high summer – dusty yellow and a deep (almost stagnant) green. By contrast the same space in winter is left largely in black and white to indicate a snow-covered landscape with only blue edging around the lake and red ink for the words describing the children’s different routes to the North Pole. The colour schemes allow for dramatic contrast between maps but also closely relate to the holistic play ‘genre’ that the children themselves select and impose onto their environment (reminding us that the children’s own imaginative inhabitation is itself fully integrated with place and season). The way in which the children imaginatively possess the landscape is most clearly illustrated in the play of sameness and difference that the maps explicitly visualise. In Winter Holiday Ransome inverts the map again (so that it is now the same way up as Spurrier’s). The main scene of the action shifts from land to water but this is transformed into a navigable frozen space. Onto this all the characteristic elements are then replayed with a change of season and arc-narrative: the base is first ‘the igloo’, then shifts to the Houseboat, now renamed ‘The Fram’ (after the Norwegian explorer

17

Hugh Brogan, The Life of Arthur Ransome (London: Jonathan Cape, 1984), 341.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

A Material and Visual Playspace

179

Fig. 5.3. Endpaper Map from Winter Holiday (For the colour version, please refer to the plate section.)

Nansen’s ship which was frozen into the ice). The island becomes ‘Spitzbergen’; a trip inland is to ‘High Greenland’ where they rescue a ‘polar bear’ (sheep); the farms are ‘Eskimo settlements’. Thus the map visually and spatially represents each new play and totalises it on the same topographical region. One other highly distinctive characteristic, not present on Spurrier’s map but introduced by Ransome, is the integration of elements of narrative directly onto the map.18 This uncovers, first, the way in which the map works to represent simultaneously the zone, mode and arena of play, but also the core events of the sequential narrative. The map thus cleverly works in two dominant ways: to introduce and establish the terms of that play (the same place but at a different season) but also to work in an expository and integrated way alongside the reading of the text. So, Ransome adds dotted lines on the maps to indicate the movement of characters

18

Ransome’s map style, and indeed the children’s mode of imaginative mapping, owes much to Richard Jeffries’ Bevis (1882). The frontispiece map for Bevis (illustrated by E. H. Shepard) gives exotic names – ‘The Mississippi River’; ‘The Nile’ – to tributary rivers feeding into a lake in a way that anticipates the ‘River Amazon’, whilst also containing narrative information such as ‘swimming learnt here’. For a discussion of Bevis as an influence see Hunt, Approaching Arthur Ransome, 96–7.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

180

Playspace: Spatialising Children’s Fiction

Fig. 5.4. Detail from map for Swallowdale (For the colour version, please refer to the plate section.)

Fig. 5.5. Detail from map for Winter Holiday (For the colour version, please refer to the plate section.)

across the landscape (in very small neat lettering that does not distract from the larger visual). When we look closely at these (Figs. 5.4 and 5.5) we see that they correspond to a core feature of his narrative structure: the central spatial event that forms a dramatic climax in the books (usually a journey or expedition of some kind that goes wrong). For Swallows and Amazons this is when they stay out too late on the lake and have to sail home in the dark; for Swallowdale this is when they climb Kanchenjunga (Coniston Old Man) and Titty and Roger get lost in the fog on the way down; for Winter Holiday it is when Dick and Dorothea mistakenly set off for the top end of the lake and get caught in a snow blizzard; for Pigeon Post it is when the children are trapped behind the lines of a forest fire. The crisis is itself directly linked to place and season and usually to an act of disorientation (created in turn by: darkness; fog; snow; and fire). If we look at the wording written on the Swallowdale map then this reads: Overland track from the r. Amazon. Half Way Camp. Track of A.B. once lost in the fog. Here Roger twisted his ankle.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Referential or Non-referential?

181

Charcoal Burner’s Hut. Track of Stretcher Party.

The annotations act as a condensed summary of the core ‘adventure’; the dramatic climax of the narrative that is also a test of the children’s independence from adult authority and the adults’ sense of what level of freedom is required. Of course, the reader when he or she first comes to the map cannot make full sense of, and thus order, these annotations, since the spatial simultaneity of the map works against the sequential nature of the unfolding narrative. They are only activated during the reading of the story. However, at the point where the act of physical disorientation (caused by weather conditions) occurs for the children, the reader can turn to the map and locate where they are. In the map for Winter Holiday, core annotations are marked in red on the frozen lake and simply indicate the three routes to the pole: ‘D’s track to the North Pole’; ‘Nancy’s Track to the North Pole’; ‘Track of Rescue Party’ (see Fig. 5.5). The red lines make even more visually explicit than for Swallowdale that the mapped area of the text corresponds to the climactic core narrative.19 The map series makes clear the way in which each book is a reworking of the one before, a return to the same place, which can be endlessly revisited by the children.20 This also by extension connects out of the book to the reader, allowing for the possibility of them doing the same both across the series (mentally or physically connecting different seasonal maps) and out into the real world.

Referential or Non-referential? The apparently referential nature of his books has had a negative effect on critical appreciation of Ransome. It is notable that almost all of the booklength studies dedicated solely to him involve a playing-out of literary tourism through acts of mapping and re-enactment of literary events onto the real geography of the Lakes rather than full critical consideration.21 19

20

21

Other minor events are also noted on this map, however, but in black – so we have ‘Here they rescued the cragfast sheep’ top left, and a secondary narrative also on the map bottom centre ‘Here they Landed’; Here they Held up the Doctor’ etc. See also Victor Watson, Reading Series Fiction: From Arthur Ransome to Gene Kemp (New York and Oxford: Routledge, 2000). I am thinking of: Christina Hardyment, Arthur Ransome & Captain Flint’s Trunk (London: Jonathan Cape, 1984); Claire Kendall-Price, In the Footsteps of the Swallows and Amazons (Dorchester: Wild Cat Publishing, 1993); Roger Wardale, In Search of Swallows and Amazons (Wilmslow: Sigma Press, 1996). A rare early exception is Hunt’s Approaching Arthur Ransome which does offer a more literary-critical, though not theoretical, approach.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

182

Playspace: Spatialising Children’s Fiction

Famously, Ransome describes his imagined geography as a kind of mosaic in which specific elements (island; lakeside town; harbour; mountain) from Windermere and Coniston in the Lake District are amalgamated: Their lake is not altogether Windermere, though Rio is, of course, Bowness, because I had to take a good deal from Coniston. No island on Windermere has quite so good a harbour as that among the rocks at the south end of Peel Island on Coniston . . . And a good many people have spotted that Kanchenjunga must be the Old Man [the mountain of Coniston]. But Cormorant Island is Silverholme on Windermere . . .22

There is a degree of slippage here in how the author himself talks about the relationship between represented place in the books and the actual geographical places from which this space is generated. Ransome states ‘Rio . . . is Bowness’, ‘Cormorant Island is Silverholme’, in a way that suggests direct transposition. In a comment similar to that of Stevenson (whom Ransome references in his letter and by whom he is deeply influenced) we see again a reversal of representational priority as the place and space of the books starts to become ‘more real’ than that of the actual Lake District: Then, too, there had to be a little pulling about of rivers and roads, but every single place in all those books exists somewhere, and by now, I know the geography of the country in the books so well that when I walk about in actual fact, it sometimes seems to me that some giant or earthquake has been doing a little scene-shifting overnight. (5)

In terms of the authorial maps presented as endpapers, Ransome also introduces another kind of instability or slippage around the identity (real or fictional) of the map’s maker that is odd, but actually highly characteristic of literary mapping (comparable to Verne, Milne and others). Is the map to be understood as embedded within the act of representation – as a product of the children’s making – or is it outside it? Since the style of the map is the same across all of the series, and clearly authorial, the reader knows at one level that the map is made by the author, Arthur Ransome. However, in different books Ransome (as map-maker) either chooses to leave this implicit by not signing the map in any way (e.g. Swallowdale); or suggests shared ownership in a way that crosses and confuses representational boundaries (the author outside and the child within the text). A note to the List of Illustrations in Pigeon Post, for example, states: ‘As usual I have to thank Miss Nancy Blackett, whose drawing is improving hand over fist, for much 22

Arthur Ransome, ‘A Letter to the Editor’, The Junior Bookshelf 1 (1937): 3–5; 4–5.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Referential or Non-referential?

183

help with the illustrations. A.R.’ (PP 9). Elsewhere Ransome apparently hands ownership over to the children, led by Nancy, as in Winter Holiday where the map is signed ‘North Polar Exhibition: S’s, A’s, and D’s. Map Showing Routes to the Pole’. This stands in direct contrast to Swallows and Amazons where the endpaper map was definitely not that made by the children within the narrative and was signed by Spurrier (bottom left). The paratextual space of the map for Ransome thus contains the voices of both third-person author (‘Here they rescued the sheep’) and character – ‘Future explorers must not rely on this map in calculating distances. Capt. Nancy Blackett’ – in a way that slides across representational boundaries. Mapping for imaginative repossession is also explicitly replicated by the children within the books, creating a sense of generational rediscovery which is relevant for the doubled audience for children’s Fiction: the nostalgic adult revisiting place in memory and the child’s reappropriation of that space for themselves. In Swallows and Amazons the presence of two maps (the official adult one and the children’s own home-made one) makes this doubleness explicit. John asserts before they go to camp on the island that ‘“We ought to have a chart of some kind”’ (33). This leads them to find ‘a good map that showed the lake in a local guide-book. Titty said it wasn’t really a chart. John said it would do’ (33). Thus, there are two maps at work in the text (the ‘real’ map of the tourist guidebook and the home-made one), just as for Ransome there is the real Lakes of his childhood memories and his imaginative recreation of it. Within Swallows and Amazons, the participants tend to describe their map as a ‘chart’. This partly relates to the use of a naval register, which all the children adopt (after their father), and the focus on the lake itself as the centre of the map, but it also relates to more active use of a chart. As Arthur Robinson puts it in Elements of Cartography, ‘maps are to be looked at while charts are to be worked on; courses are plotted, positions determined, bearings are marked’ (9). The children’s own map is not a static object but an interactive one that can be enlarged as their knowledge of the landscape grows. The doubled nature of real-world map/created chart draws attention to a constant slippage between the realist landscape of the Lake District (also the world of the adults) and the more imaginative and romanticised literary landscape which the children create within it, which is essential to the whole way in which the books work. In Swallowdale the children explicitly stop each other using real place names for the hills: ‘That’s the hill above the valley the Amazon River comes from,’ said John. ‘I’ve seen it on the map. Its name is . . .’

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

184

Playspace: Spatialising Children’s Fiction ‘Let’s have it for Kanchenjunga,’ said Titty, ‘And then we can explore the sources of the Amazon and climb Kanchenjunga at the same time. Real exploring. . . .’ ‘Kanchenjunga’s a gorgeous name, anyhow,’ said John.23

When one of the children does supposedly use a topographical place name, the author edits it out for the reader: ‘Why have we never been up here before?’ said Peggy. ‘It’s not really very far. Look, there’s the . . .’ She said the name of the big hill with the peak. But the able-seaman, who had heard her, was quick in putting her right. ‘That one’s Kanchenjunga,’ she said . . . (174)24

As these examples make clear, much of the desire for renaming is led by Titty whose authority clearly comes from her status as ‘writer’ and the most literary amongst the children. She constantly seeks to translate the actual landscape into an imaginative zone of play, and the use of real names disrupts such a desire. So, the children have two sets of names and roles for themselves (e.g. Susan/First mate) relating to the arc-narrative that determines the play genre for the book. They also use names to signal which adults are allowed to enter into the imaginary world or not. Those that are so permitted are given two names accordingly. Uncle Jim is also ‘Captain Flint’, and the Swallows’ mother is an ‘honorary native’ allowed to know about the other world and partly enter into it. Other adults, or real-world places, are assimilated generically in relation to the arc-narrative. So, for example, Bowness is ‘Rio’ with native settlers in the high summer of Swallows and Amazons but becomes an Arctic settlement full of Eskimos and seals in Winter Holiday. The children themselves also occupy multiple levels of distance from reality signalled by their use of names and roles. The elder children only adopt one alternative identity but Titty and Roger, as the youngest children, when they are separated from their elders frequently slip into still further roles. Titty invokes both Treasure Island and Robinson Crusoe when she is alone, or alone with Roger, articulating a secondary level of imaginative recreation in which the older children never participate.

23 24

Arthur Ransome, Swallowdale (London: Jonathan Cape, 1931), 161. The only exception to the forbidden naming of real-world locations occurs when they are on the summit of Kanchenjunga when the children happily point out real places to each other: ‘“You can see right into Scotland,” said Nancy. “Those hills over there are the other side of the Solway Firth”’ (Swallowdale, 335). It seems as though, once on the summit, the children come out of their play world to enjoy the real view.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Referential or Non-referential?

185

In her introduction to The Literary Tourist, Nicola Watson comments that: To tour Ransome country is not merely a recapitulation of the original act of possession-by-reading performed by Ransome’s characters, but a repossession of that imaginative act, not because the map of the lakes corresponds in every detail with the world of Swallows and Amazons but precisely because it does not. (4)

The ‘inaccuracy’ of the correspondence between map and text, actual map and literary mapping is more significant than its accuracy. Moreover, the way that the children take hold of and creatively re-inhabit space also allows for the possibility of the book’s future readers doing the same, coming after them in an act of imaginative re-inhabitation. As Watson states: ‘The children set off to possess the land by re-naming it, an act of literary imagination’ (4). Acts of renaming and re-mapping assert that one can go on possessing and repossessing space and allow for the extension of the activities of the fictional children out into the real world for their readers. They also connect the doubled perspectives of adult and child within and beyond the text. The second book, Swallowdale, builds this concept directly into the narrative. When the children arrive at the summit of ‘Kanchenjunga’ (Coniston Old Man) they find a tin buried in the cairn in which their parents had left a record of their own climb as children up what they called ‘the Matterhorn’ (338). Titty draws explicit attention to the spatial and temporal connection that is being made: ‘“You found it,” she said. “You put it back, and then perhaps in another thirty years . . .”’ (339). When the children have made their own additions to the time-capsule, she comments that: ‘“And now perhaps it won’t be found for ages and ages till people wear quite different sorts of clothes”’ (340). The slightly disturbing backwardlooking history for the children – to an earlier, prior, act of naming by their parents (one of whom is now dead) – also implicitly works as a link forwards to current readers who, by implication, can re-inhabit and rename the same landscape for themselves. The resulting cultural value and power of the spatial in Children’s Fiction across generations is strongly felt in the comments of adult readers returning to Swallows and Amazons recorded by Fiona Maine and Alison Waller.25 The maps at the front of such works play an important role here. The researchers note that: ‘For some the physical object of the book and its 25

Fiona Maine and Alison Waller, ‘Swallows and Amazons Forever: How Adults and Children Engage in Reading a Classic Text’, Children’s Literature in Education 42 (2011): 354–67.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

186

Playspace: Spatialising Children’s Fiction

illustrations are the most memorable aspects’ (360) and record the following responses by three Adult Readers (ARs): AR Tom . . . admits that he struggled to remember the plot . . . but he had a clear recollection of the cover of his childhood copy of Swallows and Amazons, which features an image of a lake scene and a boat . . . More crucial for Tom as an adult reader is the map included in the preliminary pages which prompts him to recall his own playful map-making. (360) AR Ian and AR Sarah also mention the importance of the map, not so much to make sense of a fantastical story but as part of the creation of a whole and realistic world within the covers of the book. (361)

The researchers record comments by adult rereaders that strongly confirm the influence of the literary upon real play: ‘For some of the adult readers, the imaginative landscape of Swallows and Amazons represented by the map and described in the novel seems to have overlaid [sic] their own childhood worlds’ (361). In other words, for children, playspace easily extends beyond the bounds of fictional representation and starts to influence the experience of place and space in the real world that remains a powerful memory in later adulthood. In this way then it becomes multitemporal and potentially inter-generational.

Returning to Rousseau The analysis of Ransome’s maps has shown the power of fictional cartographic representation for children’s fiction and the ability of the map to bring to the surface the shared layers of real and represented worlds that inform playspace and underpin the genre. I want now to go back to Rousseau, in order to explore more fully the spatial dynamic between adult and child that underlies this state of being and fully establish this positive counter-model to the dominant Freudian one. Playspace emerges from a concept at the heart of Rousseau’s model, that of ‘la liberté bien réglée’ (livre II, 321) (‘well-regulated freedom’ (book II, 92)). In Émile or On Education, Rousseau argues that there are three ways in which the child is educated (by nature, man and things): Cette éducation nous vient de la nature, ou des hommes, ou des choses. Le développement interne de nos facultés et de nos organes est l’éducation de la nature; l’usage qu’on nous apprend à faire de ce développement est l’éducation des hommes; et l’acquis de nôtre propre expérience sur les objets qui nous affectent est l’éducation des choses . . . Or de ces trois éducations différentes, celle de la

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Returning to Rousseau

187

nature ne dépend point de nous; celle des choses n’en dépend qu’à certain égards; celle des hommes est la seule dont nous soyons vraiment les maitres . . . Puisque le concours des trois éducations est nécessaire à leur perfection, c’est sur celle à laquelle nous ne pouvons rien qu’il faut diriger les deux autres. (I, 247) (This education comes to us from nature or from men or from things. The internal development of our faculties and organs is the education of nature. The use that we are taught to make of this development is the education of men. And what we acquire from our own experience about the objects which affect us is the education of things . . . Now, of these three different educations, the one coming from nature is in no way in our control; that coming from things is in our control only in certain respects; that coming from men is the only one of which we are truly the masters . . . Since the conjunction of the three educations is necessary to their perfection, the two others must be directed toward the one over which we have no power . . . (I, 38–9))

In this account of the principles underlying the argument for a natural education Rousseau argues for the privileging of nature’s influence as the force most external to us which is only able to act upon us and shape us for the better. He asserts that the child is naturally good: Posons pour maxime incontestable que les premiers mouvemens de la nature sont toujours droits: il n’y a point de perversité originelle dans le coeur humain. Il ne s’y trouve pas un seul vice dont on ne puisse dire comment et par où il y est entré. (II, 322) (Let us lay it down as an incontestable maxim that the first movements of nature are always right. There is no original perversity in the human heart. There is not a single vice to be found in it of which it cannot be said how and whence it entered. (II, 92))

The second book of Émile is thus, unsurprisingly, centred upon the imaginary pupil’s education in the natural world in order for that education to be as uncorrupted as possible. Rousseau chooses to ‘élever Émile à la campagne loin de la canaille des valets’ (II, 326) (‘raise Emile in the country . . . far from the rabble of valets’ (II, 95)). He asserts that: Les prémiers mouvemens naturels de l’homme étant donc de se mesurer avec tout ce qui l’environne et d’éprouver dans chaque objet qu’il apperçoit toutes les qualités sensibles qui peuvent se rapporter à lui, sa première étude est une sorte de physique expérimentale rélative à sa propre conservation et dont on le détourne par des études spéculatives avant qu’il ait reconnu sa place ici-bas. (II, 369–70) (Since man’s first natural movements are, therefore, to measure himself against everything surrounding him and to experience in each object he perceives all the

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

188

Playspace: Spatialising Children’s Fiction qualities which can be sensed and relate to him, his first study is a sort of experimental physics relative to his own preservation, from which he is diverted by speculative studies before he has recognized his place here on earth. (II, 125))

A natural education is so twice over: it occurs in a natural (rural) environment removed from the perceived negative effects of society, thus permitting nature to act freely, but it also allows the child to learn the strengths and weaknesses of himself as an individual ‘naturally’ in ways that proceed directly from the body, through trial and error (though without endangering the child). At the heart of book II, Rousseau gives an account of the importance of the lesson of self-reliance implicit in such a model: Pour mon élêve, ou plutôt celui de la nature, exercé de bonne heure à se suffire à lui-même autant qu’il est possible, il ne s’accoutume point à recourir sans cesse aux autres, encore moins à leur étaler son grand savoir. En revanche il juge, il prévoit, il raisone en tout ce qui se rapporte immédiatement à lui . . . il prend ses leçons de la nature et non pas des hommes; il s’instruit d’autant mieux qu’il ne voit nulle part l’intention de l’instruire. Ainsi son corps et son esprit s’éxercent à la fois. (II, 361) (As for my pupil, or rather nature’s, trained early to be as self-sufficient as possible, he is not accustomed to turning constantly to others; still less is he accustomed to displaying his great learning for them. On the other hand, he judges, he foresees, he reasons in everything relating to him . . . He gets his lessons from nature and not from men. He instructs himself so much the better because he sees nowhere the intention to instruct him. Thus, his body and his mind are exercised together . . . (II, 119))

In a foundational (idealised) Romantic account of education, the child learns best when he does not realise that he is learning at all. This account centrally informs the spatial dynamic of children’s fiction as I wish to develop it here as a positive, healthy and necessary dynamic between adult and child. The child is strongly receptive to his environment, using all his senses to acquire experience based on an innate self-protection. At the same time, such self-reliance can only come about as a result of the sympathetic distance of a watchful adult, as Rousseau also makes clear: ‘Jeune instituteur, je vous prêche un art difficile; c’est de gouverner sans preceptes et de tout faire en ne faisant rien’ (II, 362) (‘Young teacher, I am preaching a difficult art to you, that of governing without precepts and doing everything by doing nothing’ (II, 119)). This secondary spatial aspect of the Romantic child (self-reliance within a framed environment) has tended to be overshadowed by the recent ‘innocence’ debate in literary criticism of children’s fiction. Nonetheless, it centrally informs the represented dynamic between adult and child in

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Returning to Rousseau

189

late nineteenth- and twentieth-century children’s fiction. At a level of environment and physical geography, it determines a dominant trope of the ‘rural idyll’ emerging strongly in works such as Richard Jeffries’ Bevis, the works of E. Nesbit and their many literary descendants (including Arthur Ransome). At another level, in terms of the child–adult dynamic it establishes a healthy norm (the right level of framing) which most works of children’s fiction either conform to or deviate from – with immediate and obvious consequences for the physical and mental health of the child. Whilst Rose’s critique of the idealised Romantic child may work theoretically at a level of language and text, it overlooks the fact that, even if the construct of the child does not function as a pure and innocent originary ‘other’ to the adult, it is still the case that real children are of a different order from adults. Children are far more in a state of flux and growth than an adult; more impulsive and less risk-averse; more playful; more able to recover rapidly; more likely to put themselves in danger. In reality adults cannot take themselves out of the equation altogether. Thus, some kind of frame is surely essential for the real child (and correspondingly in literature for representation of the child). The Romantic construction of childhood can be understood to be negatively determined (through the psychological frame of an adultimposed model that is unconsciously generated by the repressed child within, or the needs of larger society), but it can also bear upon children’s fiction spatially in ways that offer a more positive reading of the adult–child dynamic. Here, I align myself with Alan Richardson, whose article on ‘Romanticism and the End of Childhood’ seeks to offer ‘a post-Romantic approach to childhood which avoids transcendentalism but provides an alternative to exclusively self–regarding (“adultist”) or narrowly instrumental views of the child’.26 Richardson, in his controlled critique of Rose and others, makes the point that one aspect of the Romantic model is only able to be criticised because of another aspect of the same model: Those who are presently attempting to frame such a rhetoric have relied, selfconsciously or not, on the distinctively Romantic construction of childhood, which facilitated the democratization of an extended educative childhood in the first place. (179)

There is no question that, as Rose suggests, the Romantic account depicts the child as a being of a different order from the adult (indeed this 26

Alan Richardson, ‘Romanticism and the End of Childhood’, Nineteenth Century Contexts 21 (2008): 169–89; 180.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

190

Playspace: Spatialising Children’s Fiction

realisation is what underpins Romantic interest and investment in childhood).27 At its most extreme this difference is idealised and the child is registered as higher, purer and partaking of the divine (essentially ‘unfallen’). However, we can also immediately start to question the oversimplification of the Romantic model as an idealisation of innocence, or at least recognise that that ideal relies upon a particular kind of self-conscious framing by the adult (and/or older child) in ways that I want to suggest playspace necessarily requires. William Blake, for example, in Songs of Innocence and Experience, presents innocence as a temporary, vulnerable condition that sets the child apart because it is also a necessary ignorance of the world that means the child requires protection if such a condition is to be maintained. In a poem such as ‘Nurse’s Song’ protection comes from the adult, who knows she must bring the children safely indoors as darkness gathers but permits them to ‘go & play till the light fades away’ (line 13).28 In ‘The Chimney Sweep’ protection is partially offered by an older boy, more knowing than the new sweep, but still not disillusioned enough to fully realise the injustice of his life and so seeking to make the best of things: There’s little Tom Dacre, who cried when his head That curl’d like a lamb’s back, was shav’d, so I said, ‘Hush, Tom never mind it, for when your head’s bare, You know that the soot cannot spoil your white hair.’ (p. 26, lines 5–8)

Innocence is emphatically a temporary condition, always about to become something else, but that state is only possible at all because of those situated around it. The emphasis even here, then, is on the necessary presence of protection if any kind of childhood freedom is to be truly experienced. In other words, innocence is always framed by knowingness: in fact it has to be to exist at all. Karen Lesnik-Oberstein – a critic who adopts Rose’s position but applies it back onto the history of Children’s Literature – dwells at length on Rousseau in terms of the ‘education–amusement’ divide. In many ways her account anticipates my own here, when she notes that: ‘In Rousseau, 27

28

Rousseau famously states: ‘La nature veut que les enfans soient enfans avant que d’être hommes . . . L’enfance a des maniéres de voir, de penser, de sentir qui lui son propres’ (Émile, II, 319); (‘Nature wants children to be children before being men . . . Childhood has its ways of seeing, thinking, and feeling which are proper to it’ (Émile, II, 90)). William Blake, Blake’s Poetry and Designs, ed. Mary Lynn Johnson and John E. Grant (New York and London: W.W. Norton & Co. Ltd, 1979), p. 35.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Mapping Negative Playspace

191

even more strongly than in Locke we see how the “child” is the product of the paradox of teaching liberty’ (Children’s Literature, 94–5). Rousseau’s educational model is caught in a problematic position between a desire for the child to be brought up ‘naturally’ and for ‘natural education’ to then make him a better member of society and shaped to meet its needs. LesnikOberstein states that: The child’s liberty is a specialized liberty, particular to childhood, and is part of Rousseau’s argument for the specialized concept of ‘child’ as a whole. It is a liberty without power that Rousseau sees for the child. (97)

She concludes that: Children’s literature criticism and production almost wholly ignores this context . . . Because of this it does not identify any need to examine or explore the problems of freedom within restriction. (98–9)

Lesnik-Oberstein also recognises the uniqueness of this space as a ‘specialised liberty’ that is ‘particular to childhood’ but for her the structure of ‘freedom within restriction’ can only be read negatively as ‘liberty without power’: the adult is always manipulating the child to either personal or societal ends. Of course, this must be true to some extent – Rousseau’s entire educational rationale is about creating the best possible human being by shaping him or her early – but what would be the desired alternative? Total freedom and autonomy? Children left entirely to themselves are in danger, as even Rose acknowledges: ‘Education can only be justified first by the fact that a child left to its own devices would perish (this may seem obvious but it is important if you are arguing for a self-sufficient natural state)’ (The Case of Peter Pan, 44). In any case, there is no option for not shaping the child; a child will be influenced one way or another as a result of being an immature form of life. What is really important, then, is that the structure (the necessary framing of the child’s experience by adult or older child) is one that determines healthy growth and development and allows children to thrive, to journey, to explore, rather than restricting or damaging them.

Mapping Negative Playspace As Ransome’s maps so clearly demonstrate, positive playspace is eminently mappable. Indeed, the act of mapping is a pleasurable element in the production and establishment of this protected zone, asserting ownership

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

192

Playspace: Spatialising Children’s Fiction

by the children on their own terms. In the most shared of spaces adult and child both purport to partake in the making of the map (as we saw with Ransome’s maps). So, for example, E. H. Shepard’s map for Winnie-thePooh is signed at the bottom: ‘Drawn by me and Mr Shepard helpd’, with the adult illustrator adopting the persona of Christopher Robin as map-maker. However, if, as I am suggesting, playspace is a core spatial structure capable of determining the genre, then both positive and negative models ought to apply. That is, an absence of playspace should be as powerful as its presence. I want to consider this briefly here, and also consider how such negation bears upon the map. Here, that same (multiple) work that was so central to the debate around the existence of children’s fiction itself proves significant once again: J. M. Barrie’s Peter Pan. As the setting for Peter’s world changes and darkens across different versions of the text, place moves from being necessarily mappable to resisting and denying any attempt to map it. The two earliest versions of the narrative – The Little White Bird and Peter Pan in Kensington Gardens – present a real-world setting for the story in London, reinforced by an accurate map entitled ‘Peter Pan’s Map of Kensington Gardens’ (see Fig. 5.6). This map represents: actual places

Fig. 5.6. ‘Peter Pan’s Map of Kensington Gardens’ from Peter Pan in Kensington Gardens

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Mapping Negative Playspace

193

(‘the Round Pond’; ‘the Broad Walk’); fictional sites that feature in the story (‘The Faeries Winter Palace’, Peter Pan in his boat on the Serpentine); and fictionalised versions of real sites (‘the Baby’s Palace’ for Kensington Palace; ‘the Gold King’ for the Albert Memorial). So the map merges real and imagined place. This mapped space visualises the park in terms of strong forms of physical and human protection for the young child, with fencing all round it and figures of women with young children and prams depicted on the pathways. The text reasserts this: The Gardens are bounded on one side by a never-ending line of omnibuses, over which your nurse has such authority that if she holds up her finger to any one of them it stops immediately. She then crosses with you in safety to the other side. (Peter Pan in Kensington Gardens, 3)

The book itself then opens with a ‘Tour’ that verbally explores the Gardens in a way that directly corresponds to the map: ‘We are now in the Broad Walk’ (4); ‘Next we come to the Hump’ (5). The calm, adult voice of the narrator, guiding and leading the second-person ‘you’ round key sites creates a further level of safety and control within the text. Thus although in this first version baby-Peter is able to fly out of his own nursery ‘by the window, which had no bars’ (13) and escape to the park ‘after Lock-Out Time’ (15), he is both out and in – not where he should be but still in a safe, mappable, protected space. The changes that occur in the representation of Peter himself across the play and book versions of his story bear directly upon playspace, which morphs from positive to negative in ways that begin to signal a separation of the shared dynamic between adult and child. In Peter Pan and Wendy, the same imaginative space that holds a strong attraction and charm for the children is threatening to the adults who seek to protect them.29 On the night that the children’s parents go out, allowing Peter to fly in the window and take the children off, Mrs Darling feels ‘a nameless fear’ (86). Michael asks: ‘“Can anything harm us, mother, after the night lights are lit?”’ to which she replies: ‘“Nothing, precious . . . they are the eyes a mother leaves behind her to guard her children”’ (86). From the perspective of the adult, wanting the child to play in safety, Peter is a dangerous threat, precisely because of his desire to remove them entirely from all forms of adult protection.30

29

30

J. M. Barrie, Peter Pan in Kensington Gardens and Peter and Wendy, ed. Peter Hollindale (Oxford: Oxford University Press, 1991), 77. See also Rose, ‘There are in fact two conflicts in Peter Pan, that between the nursery and the Never Land . . . and that waged by Peter against both’ (The Case of Peter Pan, 35).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

194

Playspace: Spatialising Children’s Fiction

When they set out on their adventure, the children start off unaware of the dangers of a world of play without any boundaries. To them Neverland is the ultimate playspace, containing all of their desired elements: Of all delectable islands the Neverland is the snuggest and most compact; not large and sprawly, you know, with tedious distances between one adventure and another, but nicely crammed. (74)

Quite rapidly, however, even on the way to it, as they are in danger of being left behind by Peter, the children begin to realise their own vulnerability in a world with no adult controls of any kind: In the old days at home the Neverland had always begun to look a little dark and threatening by bedtime . . . Of course the Neverland had been make-believe in those days; but it was real now, and there were no night-lights, and it was getting darker every moment, and where was Nana? (106)

Wendy, as the oldest child (and substitute parent), recognises the threat soonest as Peter takes delight in Michael falling asleep whilst flying: The awful thing was that Peter thought this funny. ‘There he goes again!’ he would cry gleefully, as Michael dropped like a stone. ‘Save him, save him!’ cried Wendy, looking with horror at the cruel sea far below. (103)

Part of the horror lies in Peter’s total lack of any sense of responsibility. This also applies to the uneasy relationship between ‘make-believe’ as Peter’s normative state and the ‘real’ which the other children are all aware of. For ordinary children play is only a temporary condition. A return to reality is asserted by the interrupting adult calling the child in for meals, or to get ready for bed. When this presence is removed, with no counterframe to the world of ‘make-believe’ (which Peter entirely inhabits), things threaten to become disordered: The difference between him and the other boys at such a time was that they knew it was make-believe, while to him make-believe and true were exactly the same thing. This sometimes troubled them, as when they had to make-believe that they had had their dinners. (128)

Peter’s condition – and the nature of the split spatial world that comes into being around him – is a direct consequence of the core act of adult rejection that holds firm across all four versions of his story.31 In Peter 31

J. M. Barrie, The Little White Bird (Hodder & Stoughton, 1902) consists of episodic narratives for which chs. 13–18 are about Peter Pan; performance of Peter Pan, or The Boy Who Would Not Grow Up at the Duke of York’s Theatre, London, 27 December 1904 (145 performances);

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Mapping Negative Playspace

195

Pan in Kensington Gardens, Peter toys with returning home (making it safely back to his grieving mother but, after observing her unseen, deciding not to stay). On a later return, after some months have passed, he finally decides to go back: [S]o eager was he to be nestling in her arms that this time he flew straight to the window which was always to be open for him. But the window was closed, and there were iron bars on it, and peering inside he saw his mother sleeping peacefully with her arm around another little boy. Peter called, ‘Mother! mother!’ but she heard him not; in vain he beat his little limbs against the iron bars. (‘Lock-out Time’, Peter Pan in Kensington Gardens, 40)

In the most disturbing of all the accounts, the unspoken agreement that the window ‘was always to be open for him’ is broken and the absolute trust of childhood destroyed. As a result, the eternality that should apply to the mother’s love for her child, is displaced onto Peter himself.32 His state is not a simple inversion of playspace but an impossible aberration – an entirely safe non-adult space that needs no boundaries and has no limits. The same story of abandonment, rejection and substitution is reiterated in the play version, but with far greater distance. Peter retells his tale factually and it ends with a stage direction in brackets that is rendered comic by the level of understatement: . Wendy, you are wrong about mothers. I thought like you about the window, so I stayed away for moons and moons, and then I flew back, but the window was barred, for my mother had forgotten all about me and there was another little boy sleeping in my bed. (This is a general damper).33

Finally, the tragic story is retold again in Peter Pan and Wendy, but immediately undercut by the narrator: ‘Long ago,’ he said, ‘I thought like you that my mother would always keep the window open for me; so I stayed away for moons and moons and moons, and then

32

33

Peter Pan in Kensington Gardens (Hodder & Stoughton, 1906) contains the chapters from The Little White Bird published as a separate short work; Peter and Wendy (Hodder & Stoughton, 1911); Peter Pan, or the Boy Who Would Not Grow Up (Hodder & Stoughton, 1928), the final publication of the playscript. The rejection of the mother is only subjectively experienced by Peter. She may still be grieving for him and wanting him to return and the bars on the window may only be there to ensure that the same loss does not occur again with a second child. At the same time, however, the barring of a window that should remain forever open is clearly a negative symbol. J. M. Barrie, Peter Pan: A Fantasy in Five Acts [1928] (London and New York: Samuel French, 1956), 59.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

196

Playspace: Spatialising Children’s Fiction flew back; but the window was barred, for mother had forgotten all about me, and there was another little boy sleeping in my bed.’ I am not sure that this was true, but Peter thought it was true; and it scared them. (167)

The primary danger of wandering too far from the adult is that the adult might lose interest in you altogether (or replace you with a newer model). But the truly dark side of Neverland, as of Peter himself, is that if you stay away too long, you cease to even care. As Peter himself changes, so does the world to which he escapes, and its ability to be subjected to any form of cartographic control. Where the park provided a home-from-home in Peter Pan in Kensington Gardens, in the next (play) version Peter has been relocated to ‘the Never Land’. What was before explicitly mapped by the adult narrator (‘it will be difficult to follow Peter Pan’s adventures unless you are familiar with the Kensington Gardens’ (4)) now starts to resist and defy representation. It cannot be fully visualised or glimpsed: ‘What you see is the Never Land. You have often half seen it before, or even three-quarters’ but ‘if you were to see the island bang . . . the wonders of it might hurt your eyes’ (Act II, p. 24). The parental space of daytime nannies and nurses that is freed at night from adult authority is replaced by something more child-centred but also more dangerous because it is now anchored only to Peter himself. In its final form in Peter and Wendy, the route to and from the Never Land is unlocatable without Peter – ‘How could we ever find our way back without him?’ (103) – and also far more sinister: ‘They all recognised it at once, and until fear fell upon them they hailed it . . . as a familiar friend’ (105). Now the place is fluid and changeable – ‘nothing will stand still’ (Peter and Wendy, 74) – and cannot be fixed and held in any stable form. In the case of Peter Pan, then, playspace increasingly resists and denies the attempt to take control through mapping – just as Peter himself denies any form of adult authority once he has been rejected. The danger of this place is complex and subtle. As John Griffith puts it: ‘The worlds of Treasure Island and Narnia do not threaten or lure the characters in quite the same way. The Neverland is more disturbing . . .’34 Other, darker, forms of children’s fiction actively negate playspace and make use of the map to do so, such as Alan Garner’s The Weirdstone of Brisingamen (1960).35 In this book the map of ‘The Journey from Highmost 34

35

John Griffith, ‘Making Wishes Innocent: Peter Pan’, The Lion and the Unicorn 3 (1979): 28–37; 29. Alan Garner, The Weirdstone of Brisingamen (London: Collins, 1960).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

Mapping Negative Playspace

197

Redmanhey’ shares many features with Ransome’s maps: a dotted line for the route of the children; the use of text to summarise key events at key sites within the narrative: ‘Here they sheltered from the storm’. However, it also signals layers of history and the juxtaposition of past and present, from the modernity of Macclesfield to the representation of country houses such as ‘The Riddings’ and finally to the ancient sites of ‘Rednor Mere’ or ‘Clulow Cross’. The map is not dark or sinister in itself, yet the marked route through woods and off roads suggests the necessity of cover and flight. Labels on the map that at first sight seem entirely neutral – ‘Here they met Harry Wardle’ – identify key moments of understanding in the narrative about the distorted world that the children have entered. In the meeting signalled here, the protective father-figure of Gowther is glad to encounter Wardle who is ‘all reet, I’ve known him since we were lads’ (197). However, the innocent encounter has a hidden darker meaning. When they go back to this spot they find him gone – strongly suggesting that he is working against them and spying for the enemy. Despite its surface banality, the label on the map is there because the encounter is significant. It points to a complete inversion of positive playspace – a world in which adults are not kindly protectors, nor clear threats but – far more disturbing – posing as what they are not (pretending to be the former, but actually being the latter). It reveals to both adults and children that no one can be taken for granted, not even the childhood friend. Similarly, in Garner’s Elidor (1965) although a map is not given at the front of the book the connection between modern and ancient or other worlds is initiated through the form of an interactive street map – ‘ a tall machine of squares and wheels’ – that excites the boy Roland: ‘It’s smashing,’ he said. ‘Come and look. See this roller? It’s the street index . . . you can find any street in Manchester. It’s easy, watch.’36

Roland’s apparently random spin of the index wheel is what leads him to the site of the abandoned church, where worlds collide. Again, the apparent control of space that the map allows (‘“you brought us straight here, anyway, Roland” said Nicholas’ (17)) is underlain with something deeper and darker, capable of asserting itself beneath, and even through, apparent order and modernity. The different versions of Peter Pan trace the transition of playspace from positive to negative, but in a world that is still vitally bound to and 36

Alan Garner, Elidor (London: Collins, 1965), 12.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

198

Playspace: Spatialising Children’s Fiction

emanating from the imagination of the child. Garner’s negative model is more fearful and threatening. Now the danger is not somewhere remote, apart from the real world; it threatens to enter, destabilise and destroy. Even more disturbingly, the core dynamic between adult and child is also inverted. The adult protectors who should be establishing the safe bounds of play are those that threaten, and the children, instead of ceding responsibility and being released into play, are responsible for the fate of all. The absence, or inversion, of playspace is as powerful within the genre as its presence.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.006

6

Mapping Worlds: Tolkien’s Cartographic Imagination

[In] such a story one cannot make a map for the narrative, but must first make a map and make the narrative agree. J. R. R. Tolkien, Letters The map is hell! J. R. R. Tolkien, Letters

What to Do First 1. Find the MAP. It will be there. No Tour of Fantasyland is complete without one. It will be found in the front part of your brochure . . . 2. Examine the Map. It will show most of a continent . . . If you are lucky the Map will carry an arrow or compass heading somewhere . . . and this will show you which way up to hold it . . . but there is no scale of miles and no way of telling how long you might take on the way to see these places. In short, the Map is useless, but you are advised to keep consulting it, because it is the only one you will get.1 In her tongue-in-cheek A–Z of useful information, Diana Wynne Jones begins (‘obviously’ as she informs us on the dust jacket) with the map. Her advice to the visiting tourist of ‘Fantasyland’ tells us a lot about common assumptions concerning the relationship between map and text for the genre of Fantasy: that the map is endemic to the point of redundancy for this genre (as it is not for any other); that the map does not function like a real-world map, or contain information that is of any real help (since it does not correspond to any known coordinates); but that, despite this, a need for the map remains. It reassures the reader, giving an illusion of partial control: this may be an entirely alien environment, but at least you can map it. It serves to ‘authenticate’ something which is entirely

1

Diana Wynne Jones, The Tough Guide to Fantasyland (London: Gollancz, 1996), 1–3.

199

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

200

Tolkien’s Cartographic Imagination

inauthentic; it fools us into responding to this world as if it can be mapped. But of course, the map is useless twice over; not only does it often fail to provide the kind of cartographic information needed – most notably a scale – but it does so because it depicts an entirely alternative universe to which the rules of human cartography probably do not apply. Why does the fictional map emerge so strongly, almost as a necessity, for this particular genre? Since maps are of our world, and Cartesian cartography is a way of asserting control over space, we might logically expect maps to be found frequently in realist works of fiction – functioning as an act of verisimilitude intended to make the realism even more convincing – rather than in Fantasy. Instead, somewhat paradoxically (as we shall go on to explore in Chapter 7), the reverse is true. Since fantasy as a genre explicitly creates a world that signals its difference from our own in time and space, it must follow that it is precisely because fantasy is so selfconsciously unfamiliar that the need is felt for a map. Bjorn Sundmark comments on this contradiction: The inherent world-making potential is the most important aspect of the map iconotext. It also explains why maps are particularly common in fantasy fiction. World-making or secondary world creation as Tolkien calls it, strives towards fantasy and verisimilitude at the same time.2

Similarly, in one of the earliest articles on the subject of literary maps (1987), Peter Hunt describes fantasy maps as: both reductive and suggestive; they stabilize the fantasy, while releasing greater imaginative potential . . . they also symbolise the tension that exists for the writer between the real landscape and the fantasy which inhabits it.3

Hunt articulates a near-contradiction: in relation to the text, and the world it represents, such maps fix and potentially limit understanding of the representation of place; in relation to the reader they release imaginative engagement in what Anthony Pavlik has described as ‘an experiential relationship’.4 Elements that are not fully mapped, or the way in which

2

3

4

Bjorn Sundmark, ‘Mapping Middle Earth: A Tolkienian Legacy’, in Maps and Mapping in Children’s Literature: Landscapes, Seascapes and Cityscapes, ed. Nina Goga and Bettina Kümmerling-Meibauer (Amsterdam: John Benjamins Publishing Co., 2017), 221–37; 223. Peter Hunt, ‘Landscapes and Journeys, Metaphors and Maps: The Distinctive Feature of English Fantasy’, Children’s Literature Association Quarterly 12.1 (1987): 11–14; 11. Anthony Pavlik, ‘A Special Kind of Reading Game: Maps in Children’s Literature’, International Research in Children’s Literature 3 (2010): 28–43; 38.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

What to Do First

201

map exceeds text, work to ‘open up the potential of spatiality for the reader’ (39). The ubiquity of the map in works of fantasy is created by the presence of maps in the initiating work of a new genre (as has been true for other examples considered in this study), which brings us rapidly to J. R. R. Tolkien’s The Lord of the Rings. Even in the 1980s, R. E. Walker noted of Tolkien’s maps that ‘Not only have they set a high standard, they seem to have created an interest in fantasy maps . . . so that a map has become almost de rigueur.’5 Although fantastic works of fiction pre-date Tolkien, these are set in, or extend out of, the real world (what Tolkien calls the ‘Primary World’).6 Tolkien’s creation of an entirely self-sufficient ‘Secondary World’ in effect means that he creates a new genre (‘High Fantasy’) or, at least, as Stefan Ekman puts it: ‘the publication of Tolkien’s novel marked the beginning of seeing fantasy as a genre’.7 The presence of a map as a problematic (because assumed) element of fantasy can also be charged against Tolkien. In Rhetorics of Fantasy, Farah Mendlesohn analyses the way in which the genre is constructed by defining four broad categories from which she explores ‘what ideological consequences emerge from the rhetorical structures’ (xviii).8 A brief engagement with (and dismissal of ) the fantasy map occurs in her discussion of history within the ‘portal-quest’ form. She states: ‘since the late 1970s, genre fantasy has frequently been signaled by these two devices: the map . . . and the fixed and narrated past’ (14). The focus of her anti-map stance lies in its reinforcement of history as a dead past, no longer open to debate and argument. Thus: The structure becomes ideological as portal-quest fantasies reconstruct history in the mode of the Scholastics, and recruit cartography to provide a fixed narrative, in a palpable failure to understand the fictive and imaginative nature of the discipline of history. (14)

5

6

7

8

R. E. Walker, ‘The Cartography of Fantasy’, in Mythlore: A Journal of J. R. R. Tolkien, C. S. Lewis, Charles Williams and the Genres of Myth and Fantasy Studies 7 (1981): 33–8; 37. A Primary World, in which the fantastic extends out of a real-world setting, corresponds to ‘Low Fantasy’ (e.g. Gulliver’s Travels). As Stefan Ekman notes, however, there is a degree of slippage here between ‘Primary’ as the actual real world of the living author and a represented fictional ‘real-world’ setting (Here Be Dragons, 45). Ekman, Here Be Dragons, 9. Bjorn Sundmark also reminds us of the early generic indeterminacy around The Lord of the Rings: ‘Today we anachronistically think of it as fantasy, simply, but on publication it was a ground-breaking hard-to-define children’s book’ (223). The four forms are: the portal-quest, the immersive, the intrusive and the liminal. These are determined by ‘the means by which the fantastic enters the narrated world’ (Farah Mendlesohn, Rhetorics of Fantasy (Middletown CT: Wesleyan University Press, 2008), xiv).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

202

Tolkien’s Cartographic Imagination

The blame is laid at Tolkien’s door: ‘Tolkien set the trend for maps and prehistory, establishing a pattern for the quest narrative in . . . the insistence that there is past and place behind’ (14). From such a perspective, maps of fantasy worlds are not dynamic and meaningful, integrating with core spatial meanings in the work, but instead function to ‘fix the interpretation of a landscape. Maps are no more geography than chronology and legend are history, but in portal-quest fantasies, they complete the denial of discourse’ (14). Stefan Ekman’s recent study Here Be Dragons (2014) suggests that such depreciation of the fictional fantasy map, and of Tolkien’s role in relation to it, may be about to change.9 The primary focus of his book is on the concept of division as central to fantasy but this leads him to focus on different kinds of division, the smallest of which is that represented by the map. In contrast to Mendlesohn, then, Ekman devotes an entire chapter to ‘Maps’. Although his starting point is once again map ubiquity (‘it is often taken to be a hallmark of fantasy and one of the genre’s most distinctive characteristics’ (11)) Ekman sets out to explore and challenge assumptions. His survey of two hundred works of fantasy fiction leads him to conclude that: ‘no more than 40 percent and possibly as few as 27 percent of all fantasy novels actually contain any maps’ although, as we might expect, ‘maps are much more common in fantasy set in a secondary world’ (22).10 Because Ekman is only looking at one genre containing fictional maps, he is disappointed by this finding, and concludes that ‘maps are not the compulsory ingredients they are widely held to be’ (22). However, if they are not indeed ‘compulsory’ they are still almost certainly occurring at a far higher rate for this genre than for any other. In the context of this study and this chapter, Tolkien is approached as a pre-eminent figure for fictional maps in general, not just in relation to the emergence of (high) fantasy as a distinct genre in the mid-twentieth century. No other author uses maps on the scale and level, in private and public, that he does.11 This chapter therefore unashamedly focuses solely on Tolkien’s integrated model of writing and mapping. It does so in order

9 10

11

See Ekman for a useful overview of previous critical readings’ 15–19. Oddly Ekman’s survey does not record a distinction between primary and secondary fantasy so that he later suggests: ‘A valid question that remains for a future study would thus be to what extent high-fantasy novels come with maps’ (65). The only other authors whose creative cartography might compare to Tolkien’s are Jules Verne (for whose published maps see the extremely useful website http://verne.garmtdevries.nl/en/ maps/originals.html; and, much more recently, Reif Larsen in The Selected Works of T. S. Spivet (London: Vintage, 2010).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

‘Each Is Both Prior to the Other and Later Than It’

203

to explore the fictional map across the entirety of process and in order to consider the use of multiple maps in relation to a literary totality (not just a place but an entire world-system).12 The chapter therefore moves across writerly (‘working’) maps, readerly published maps and into post-authorial re-mappings. The first section traces the way in which, for Tolkien, the literary work develops holistically in a manner that vitally connects the visual and verbal, mapping and writing; before exploring the relationship between working authorial maps (J. R. R. Tolkien) and multiple published maps (Christopher Tolkien). The final part of the chapter looks at the production of post-authorial atlases where Tolkien’s accurate and selfconsistent literary cartography stimulates and enables creative re-mapping.

‘Each Is Both Prior to the Other and Later Than It’13 Like other writers considered in this book, such as Stevenson or Ransome, Tolkien’s literary creativity cannot take place without external visualisation. What this means in relation to the mapping of literary place and space is a fundamental distinction between authorial (writerly) mapping as part of the act of literary production (what we might call ‘the working map’) and readerly mapping – the use of given maps by the reader in relation to the published work. For the writer, the map itself functions like a realworld map, albeit one that corresponds to a previously non-existent place that its creation helps to bring into being. Topography and geography for the cartographic writer – taken to its extreme in Tolkien – is never ‘background’; rather it initiates the whole, of which it later becomes a part. As Karen Wynn Fonstad puts it in her account of making The Atlas of Middle-Earth: ‘I had learned in the end what Tolkien had known at the outset: that the overall picture was necessary before the detailed stories could be fitted into place.’14 Of course, the map once made does act to fix places relationally (as those who adopt a critique of Cartesian cartography would argue) and in Tolkien’s case, it does this with an unusually high degree of internal consistency. However, the key point to grasp is that, particularly from a compositional 12

13 14

A totalising tendency is also felt in the way that Tolkien insisted that his three books were not a ‘trilogy’ and originally wanted them to be called simply The Lord of the Rings Vols. I, II, III. His publishers insisted upon him providing separate book titles. (See Letters of J. R. R. Tolkien, ed. Carpenter, 167, 169–70.) J. Hillis Miller, Topographies (Stanford, CA: Stanford University Press), 21. Karen Wynn Fonstad, ‘Writing “TO” the Map’, Tolkien Studies 3 (2006): 133–6; 134.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

204

Tolkien’s Cartographic Imagination

perspective, the map does not delimit the understanding of place and space but enables and initiates it. It is worth comparing Tolkien with Robert Louis Stevenson at this point. In both cases there is a doubled act of private map-making for creative production which has to be translated into a ‘public’ version of the map for use by the reader (also in both cases this involves a father-and-son/stepson relationship). In a letter to the proofreader for Lord of the Rings Tolkien states: I wisely started with a map, and made the story fit . . . The other way about lands one in confusions and impossibilities, and in any case it is weary work to compose a map from a story . . . (Letters, 177)

This comment is remarkably similar to Stevenson’s account of his way of working. In ‘My First Book’, as discussed in Chapter 3, Stevenson describes the drawn outline of the island and it is worth looking at this again since it provides an exemplary account of the cartographic imagination at work: Somewhat in this way, as I paused upon my map of ‘Treasure Island’, the future character of the book began to appear . . . and their brown faces and bright weapons peeped out upon me from unexpected quarters, as they passed to and fro, fighting and hunting treasure, on these few square inches of a flat projection. (118)

Visualisation of place comes first, but, once achieved, the other elements of literature – plot (the search for treasure); character; movement and event (fights) – are able to be imaginatively generated out of the ‘few square inches of a flat projection’ as the writer moves immediately into verbal articulation: ‘The next thing I knew I had some papers before me and was writing out a list of chapters’ (118). Tolkien’s difficulty with producing maps for final publication stems directly from a division between writerly and readerly needs in relation to map-making. A much-quoted letter to his publishers after submission of part I of The Lord of the Rings gives a good sense of Tolkien’s relationship to the maps: Maps are worrying me. One at least (which would then have to be rather large) is absolutely essential. I think three are needed: 1. Of the Shire; 2. Of Gondor; and 3. A general small-scale map of the whole field of action. They exist, of course; though not in any form fit for reproduction – for of course in such a story one cannot make a map for the narrative, but must first make a map and make the narrative agree. 3. is needed throughout. 1 is needed in the first volume and the last. 2. is essential in vols. II and III. (Letters, 168)

There is a clear distinction between the use of working maps by the writer (‘not in any form fit for reproduction’) and the need for maps to guide the Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

‘Each Is Both Prior to the Other and Later Than It’

205

reader in negotiating literary place and space. Tolkien is also attentive to map duration – linking active reading of different maps to different volumes and envisaging overlapping usage (e.g. of map 3 continuously and alongside maps 1 and 2). A second letter, six months later, states: The Maps. I am stumped. Indeed in a panic. They are essential; and urgent; but I just cannot get them done. I have spent an enormous amount of time on them without profitable result . . . the shape and proportions of ‘The Shire’ as described in the tale cannot (by me) be made to fit into shape of a page; nor at that size be contrived to be informative . . . Even at a little cost there should be picturesque maps, providing more than a mere index to what is said in the text. (Letters, 171)

As this letter makes clear, Tolkien had problems, as his own cartographer, in undertaking the necessary processes of selection and simplification. He is in fact the worst kind of map-maker; one who desires to include everything. A later comment on Christopher’s final maps – that they are ‘beautifully clear, as far as reduction in reproduction allows; but they do not contain everything, alas!’ (177) – makes this tendency explicit (and echoes Stevenson’s similar sentiments). As Sabine Timpf points out, in her recent discussion of map-making in The Lord of the Rings: This onerous task was not only left to Christopher Tolkien because of time pressure but mainly because the process of zooming out and imposing coherence on a map was a task almost impossible for J. R. R. Tolkien. He was much too immersed in his world . . .15

Evidence from the Letters above would certainly seem to support such an account. In this sense Tolkien’s imagination is not merely cartographic, rather the cartographic impulse is one aspect of his need to visualise. To some extent he seems to embody the tension between map-making as an artistic process and map-making as a scientific endeavour that informs cartography itself as a discipline. On the one hand, he insists upon accuracy, scale and consistency in the mapping of his world. On the other, his own maps for The Hobbit remain highly pictorial and ornate and his desire to fully represent an entirely imaginative place visually as well as verbally spills over into other visual forms of representation for key sites and scenes.16 15

16

Sabine Timpf, ‘Insights into Mapping the Imagined World of J. R. R. Tolkien’, in Binding Them All: Interdisciplinary Perspectives on J. R. R. Tolkien and His Works, ed. Monika Kirner-Ludwig, Stephan Köser and Sebastian Streitberger, (Zurich: Walking Tree Publishers, 2018), 231–52; 237. My thanks to Kate Ingle for pointing this inner tension in Tolkien out to me.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

206

Tolkien’s Cartographic Imagination

In the analysis that follows, then, I want to explore the nature of Tolkien’s ‘cartographic imagination’ across the draft process in three ways: his use of the road and river as the means of establishing and enlarging the narrative of Lord of the Rings out of The Hobbit through enlargement of the map; his integration of visual and verbal creative expression for the creation of specific sites as understood through representation of the city of Minas Tirith; and the multiple layers of private and public mapping involved for Lord of the Rings.17

Writing and Mapping: Rivers and Roads In an ‘Introductory Note’ to the essays in Tree and Leaf, Tolkien relates the writing of these early pieces at a time when: ‘The Lord of the Rings was just beginning to unroll itself and to unfold prospects of labour and exploration in yet unknown country as daunting to me as to the hobbits.’18 Tolkien describes imaginative creation itself as a kind of ‘unrolling’ of a map in a way that proves highly appropriate for his method. He continues: ‘At about that time we had reached Bree, and I had then no more notion than they had of what had become of Gandalf or who Strider was’ (9). In this account, writer and characters are set on an equal plane, both entering unknown territory together. The making of the map is also the writing of the text. Scholarly resources for exploring Tolkien’s creative-cartographic process are now comprehensive. Christopher Tolkien’s remarkable series of volumes for The History of Middle-Earth provides complex and detailed accounts of the development of both maps and texts as well as transcriptions of handwritten notes on visual images. Alongside these, two recent volumes edited by Wayne G. Hammond and Christina Scull – The Art of The Hobbit (2011) and The Art of The Lord of the Rings (2015) – present a comprehensive representation of all the major visual manuscripts.19 This 17

18

19

Christopher Tolkien notes that ‘My father was greatly concerned to harmonise Bilbo’s journey with the geography of The Lord of the Rings, especially in respect of the distance and time taken’ (J. R. R. Tolkien, The Return of the Shadow Vol. VI. History of Middle-Earth, ed. Christopher Tolkien (Boston, MA: Houghton Mifflin Co., 1988)), 204. J. R. R. Tolkien, Tree and Leaf; Smith of Wooton Major; The Homecoming of Beorhthoth Beorhtheln’s Son (London: George Allen & Unwin, 1975), 9–10. In The Art of The Lord of the Rings (London: HarperCollins, 2015) editors Wayne G. Hammond and Christina Scull tell us that: ‘In The Art of The Hobbit we aimed to publish every illustration, map, piece of formal calligraphy and design Tolkien produced . . . For The Art of The Lord of the Rings we were asked to do the same’ (13). They divide Tolkien’s art into five

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Writing and Mapping: Rivers and Roads

207

is, therefore, a relatively new (but potentially rich) area for Tolkien scholarship. Recent articles explore the ‘immersive’ nature of the maps in relation to the reader (Timpf) and the visual/verbal interface (Macleod and Smol) in ways that partly connect with my approach here.20 In their work, Macleod and Smol explore the integration of word and image on the draft pages for ‘Shelob’s lair’ which ‘illustrate[s] the importance of the visual in drafting the verbal text’ (116). They also undertake a close analysis of the drawing of the Tower of Cirith Ungol on a page of manuscript writing that shows the full integration of visual and verbal creativity: Our conclusion, therefore, is that the original pencilled text made way for the drawing, and then the inked text overwrote the pencilled one. In other words, the verbal description paused while a visual description in the form of the sketch was worked out before the verbal description was taken up again. (117)

Macleod and Smol make the point that Tolkien needed to fully externalise his imaginative vision in order to create: ‘It seems that Tolkien not only had to see this landscape in his mind’s eye, he had to see it with his physical eye’ (117). Timpf takes this one stage further, arguing for a fully immersive imagination: Tolkien started from scenes that he wrote and views that he sketched . . . the perspective (of a scene and a view) is a sign of being immersed in the narrative as well as in the geography. This is very different from having an overview in which the perspective [is] . . . able to see everything at once. (237)

Whilst I agree with Timpf that Tolkien’s creativity is holistic, involving full visualisation beyond the merely cartographic, my focus here is on the earliest acts of creative cartography relating to the transition from The Hobbit to The Lord of the Rings and the development of narrative alongside and through the drawing of the map. I want to explore relationships between the Wilderland map as published at the back of The Hobbit – which provides the starting point for creating the topography and the major maps of The Lord of the Rings – and the draft maps which come after it.21 The focus of my

20

21

categories: ‘the cartographic; drawings; illustrations; art for display; facsimiles’ (12–13). Manuscripts are held at Marquette University, Milwaukee, WI, and the Bodleian Library, Oxford, with the majority of the texts at Marquette and the majority of the art-based manuscripts at the Bodleian. Timpf, ‘Insights into Mapping the World of J. R. R. Tolkien’; Jeffrey J. Macleod and Anna Smol, ‘Visualizing the Word: Tolkien As Artist and Writer’, Tolkien Studies 14 (2017): 115–31. Alice Campbell’s entry for ‘Maps’ in The J. R. R. Encyclopaedia: Scholarship and Critical Assessment, ed. Michael Drout (London: Routledge, 2006) gives a useful summary of the dates and number of maps produced.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

208

Tolkien’s Cartographic Imagination

Fig. 6.1. Map for ‘A Part of The Shire’ from The Fellowship of the Ring

analysis is therefore on the extension of cartography out of the Map of Wilderland for The Hobbit to the East (to create a map for The Shire) and to the South (into Middle-Earth proper). Across all of the Tolkiens’ maps, even the best-known General Map (the black and red map of the whole region by Christopher Tolkien), natural features strongly predominate. Roads are marked only by faint dotted lines that make them seem minor by comparison to the mountains, forests and rivers; living creatures are dwarfed by the landscape around them. The only map for which this is not true is the published map for ‘A Part of The Shire’ (see Fig. 6.1). Any critic interpreting this map in isolation might well conclude, as Ekman does, that: ‘what stands out most is actually the network of roads’ (46) and that ‘In a story that comes with a map that includes so many roads and so many places to travel off the map, journeys are inevitable’ (47). This is true because the colouring of the river in black (the same colour as all other features) and the roads in red privileges the roads – which are of central importance to the work as a whole. However, when we compare the published map with early draft maps of The Shire, we immediately see that the reduction of the image to two colours greatly diminishes the visual power of the river. On the earliest working map of The Shire (see Fig. 6.2) there are two dominant features originally: the river in thick blue ink running down the page, with

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Writing and Mapping: Rivers and Roads

209

Fig. 6.2. ‘First sketch map of The Shire’ (For the colour version, please refer to the plate section.)

transverse tributaries feeding into it, and the red lines of the roads.22 The centre of the image is the cross created where the major roads and rivers meet at the Brandywine bridge. Rivers and roads are also vitally integrated in early maps and sketches which contain only these two elements (see also Figs. 6.2 and 6.3). There is a sense in which the landscape from The Shire across to Mirkwood is generated immediately around and along rivers and roads and this is also felt on the General Map (Fig. 6.5) where large areas of white space remain around the central ‘East Road’ that links The Shire to Wilderland. Both road and river also feature strongly within the text. Frodo’s song of the road in The Fellowship of the Ring is followed by his recollections from Bilbo: He used often to say there was only one Road; that it was like a great river: its springs were at every doorstep, and every path was its tributary . . . ‘You step into 22

Christopher Tolkien describes ‘four extant maps of the Shire made by my father, and two which I made’ (Return of the Shadow, 106); Hammond and Scull tell us that ‘Tolkien drew at least eight maps of The Shire’ (Art of LOTR, 28). They reproduce all of these (26–39).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

210

Tolkien’s Cartographic Imagination

Fig. 6.3. ‘Sketch map of roads and rivers in The Shire’ (For the colour version, please refer to the plate section.)

Fig. 6.4. Back endpaper ‘Map of Wilderland’ from The Hobbit

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Writing and Mapping: Rivers and Roads

211

Fig. 6.5. Christopher Tolkien’s ‘General Map for Middle Earth’ from The Fellowship of the Ring (For the colour version, please refer to the plate section.)

the Road, and if you don’t keep your feet, there is no knowing where you might be swept off to.’ (FR, 83)23

The close association of road and river here is directly relevant to Tolkien’s model of map-enlargement – developing along the road from Hobbiton and down the Brandywine for The Shire and along the Old Forest Road and down The Great River for the central geography of The Lord of the Rings (as we shall see). When Bilbo warns Frodo ‘“Do you realize that this 23

J. R. R. Tolkien, The Fellowship of the Ring: Being the First Part of The Lord of the Rings (London: George Allen & Unwin, 1954), 83. Henceforth FR.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

212

Tolkien’s Cartographic Imagination

is the very path that goes through Mirkwood, and that if you let it, it might take you to the Lonely Mountain or even further and to worse places?”’ (FR, 83) he directly anticipates not only Frodo’s quest but also the author’s creative-cartographic method. The local river (the Brandywine/Branduin/ Anduin) functions as a natural boundary for The Shire and all that it represents. The crossing of it represents a first step into the unknown and this is registered subjectively within the text: ‘Sam was the only member of the party who had not been over the river before. He had a strange feeling as the slow gurgling stream slipped by: his old life lay behind in the mists, dark adventure lay in front’ (FR, 97).24 In his account of Tolkien’s writing of the early chapters in The Return of the Shadow, Christopher Tolkien notes at this point that ‘This, then, is where the narrative stopped and stayed stopped for some six months or more’ (109). Thus, as he himself tells us in Tree and Leaf, Tolkien comes to a halt where his characters halt (at Bree), after a significant milestone is reached: the crossing of the natural boundary to their homeland. Once writing has recommenced, for book II of The Fellowship of the Ring, with the chapter ‘The Ring Turns South’, Tolkien (and his characters) enter entirely new territory for the first time as they move south off the Wilderland Map. One phase of the journey is over and the next phase is to begin. This is as true for the writer as it is for the Fellowship. Where before the Road has led the way, it now becomes dangerous and as Elrond puts it: ‘“Therefore it must be shunned. It will be watched. Too often the Elves have fled that way. Now at this last we must take a hard road, a road unforeseen”’ (FR, 280). In the written manuscript at this point, Tolkien makes notes about the potential route that the chosen few must take: ‘South along mountains. Over the Red Pass down the Redway to the great River’ (Return of the Shadow, 397). In the end, the characters more-or-less follow this route. It that keeps them on the lefthand side of the mountains that run along the main valley of the Great River: At the Ford of Bruinen they left the Road and turning southwards went on by narrow paths among the folded lands. Their purpose was to hold this course west of the Mountains for many miles and days. (FR, 294)

24

An earlier version more explicitly describes the river as a border: ‘But Odo had never been so far East before. He had a queer feeling as they crossed the slow silent river, as if he had now at last started, as if he was crossing a boundary and leaving his old life on the other shore’ (Return of the Shadow, 100).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Writing and Mapping: Rivers and Roads

213

As the characters head down towards Mirrormere (the point at which they will cross the mountain range), they gaze ahead of them and the vista reveals their differing stances towards what they see: ‘But the mountains are ahead of us,’ said Pippin. ‘We must have turned eastwards in the night.’ ‘No,’ said Gandalf. ‘But you see further ahead in the clear light. Beyond those peaks the range bends round south-west. There are many maps in Elrond’s house, but I suppose you never thought to look at them?’ ‘Yes I did, sometimes,’ said Pippin, ‘but I don't remember them. Frodo has a better head for that sort of thing.’ ‘I need no map,’ said Gimli . . . . ‘There is the land where our fathers worked of old . . . There the Misty Mountains divide, and between their arms lies the deepshadowed valley which we cannot forget: Azanulbizar, the Dimrill Dale, which the Elves call Nanduhirion.’ ‘It is for the Dimrill Dale that we are making,’ said Gandalf. (FR, 296)

In this passage, Pippin misreads the lay of the land because of his inability to read maps or convert them into an understanding of topography (the same mountain range which they have been following south has now curved around in front of them). Gimli responds to the landscape historically and culturally, respecting his own and the culture of others – as signalled through his use of multiple names for the same place. Finally, Gandalf remains focused on the key objective, the goal of their journey: the pass over, or through, the mountain range at Mirrormere. It is worth comparing the experience of the characters at this point with the ‘unrolling’ landscape of creative composition. In The Return of the Shadow, Christopher Tolkien presents an ‘extremely rapid, rough and now tattered map’ which ‘was my father’s first representation of Middle-earth south of the Map of Wilderland in The Hobbit’ (438) (see Fig. 6.6). The materiality of the manuscript page makes it clear that Tolkien is using exam papers to draft on. When we recall that Tolkien’s first words for The Hobbit were also written on the blank page of an exam script we might consider whether this is just the using up of scrap paper or has become something of a writerly superstition.25 The place name written and circled at this point, 25

Tolkien states: ‘All I remember about the start of The Hobbit is sitting correcting School Certificate papers in the everlasting weariness of that annual task forced on impecunious academics with children. On a blank leaf I scrawled: “In a hole in the ground there lived a hobbit.” I did not and do not know why. I did nothing about it, for a long time, and for some years I got no further than the production of Thror's Map’ (Letters, 215). The ‘Codeword, or Number’ top centre on this map clearly compares with the ‘Map of the Brandywine and Withywindle’. This has the same ‘Codeword or Number’ top centre but also ‘No. of Question’ top left confirming that it is an exam paper (Hammond and Scull, Art of LOTR, 39).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

214

Tolkien’s Cartographic Imagination

Fig. 6.6. Working Sketch Map: ‘Sketch Map of the Misty Mountains and the Lands South and East’

‘Carrock’ – denoting an eyot in the middle of the river – proves useful because it provides a reference point by which to compare different maps.26 (We can assume that this is also of use to the writer-cartographer for whom it allows for a degree of overlap with the prior map.) Drawing on this

26

Hammond and Scull note that ‘The Carrock, at top, is a landmark shared between the two [maps of Middle-Earth and of Wilderland]’ (Art of LOTR, 59).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Writing and Mapping: Rivers and Roads

(a)

(b)

(c)

215

(d)

Fig. 6.7. Comparing ‘Carrock’ on the maps. From left to right: ‘Map of Wilderland’ (published at the back of The Hobbit); Working Sketch Map (unpublished); The First Map (unpublished); ‘General Map for Middle-Earth’ (published in The Lord of the Rings) (For the colour version, please refer to the plate section.)

convenient fact, I have created a ‘strip map’ of the centre of the map across four versions (see Fig. 6.7), with Carrock marked on each by a red dot for direct comparison. This allows us to look across the developing landscape (from left to right) to compare: a. Wilderland Map; published at the back of The Hobbit; JRRT b. Working Sketch Map; unpublished; JRRT c. The so-called ‘First Map’ that pastes different parts together into a totality; unpublished; JRRT d. General Map of Middle-Earth; published but without a title; CT We can also explore the interactive relationship between mapping and writing that Tolkien himself describes when he states: ‘the thing seems to write itself once I get going, as if the truth comes out then, only imperfectly glimpsed in the preliminary sketch’ (Letters, 104). For the Working Sketch Map, then, which is the primary focus of analysis (see Fig. 6.6), we can surmise that Carrock is Tolkien’s starting point and, thus, that the river is the feature which first enables him to extend creatively downward because, in terms of shaping the landscape, it is the primary natural feature. So Tolkien outlines the Great River and extends it on below a split at ‘Gladden’ (also drawn on the Wilderland Map), giving

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

216

Tolkien’s Cartographic Imagination

Fig. 6.8. ‘Sketch of part of the Misty Mountains showing Dimrill Dale’

it (and ‘Gladden Fields’) a name for the first time.27 On the Working Sketch Map page, there is a continuous line to just above this point, then the fork is drawn slightly more heavily. In an unconfident hand Tolkien does not draw a fluid downward line for the river but appears to work upward from the bottom left corner in overlapping short lines to create the ‘Great Bend’ and its corresponding valley. These are sketched in to the left from the top of the page with a series of arrow markings down to the Gladden, at which point they bow slightly to the distinctive feature in the middle of the page. Once the main river is in place, Tolkien most likely sketches in the next natural feature: the mountains which drift leftwards, allowing for the Great Bend of the river. The first mountain range tails out here (as it does on the final General Map at Lorien) and Tolkien struggles with names for the three major peaks in this region. Clearly marked on the page, though, is a dark circle labelled ‘Mirrormere’ with a further mention of ‘Moria’.28 Here we see Tolkien’s spatial and cartographic imagination rooting itself in key spots of importance, around which visual images and verbal narrative will develop. Hammond and Scull also reproduce two further detailed sketches for Mirrormere, one viewing it from above, the other envisaging the enclosed valley ‘Dimrill Dale’ with the peak of Caradras behind it at ‘17500 ft’ (60–1), (see Fig 6.8).

27

28

I am drawing on Christopher Tolkien’s transcriptions of words on the map in The Return of the Shadow (438–41). By the time we get to the published General Map, ‘Mirrormere’ is no longer labelled, although ‘Moria Gate’ is.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Writing and Mapping: Rivers and Roads

217

Details on the map are thus worked out fully through other forms of visualisation for key sites that then help to determine an ‘itinerary’ for the narrative that is structured by the travellers on their quest. It is worth remembering, however, that the focal area on the map as it develops cartographically – the valley of the Great River – does not correspond directly to the route the characters take in the corresponding narrative. They avoid this more obvious route because of Sauron’s spies. Thus, although the narrative moves down the map, it does so in a way that is essentially off centre (on the other side of the mountain range) and deliberately so. Where for the earlier map of The Shire, visualisation of the landscape did not extend far beyond the immediate environs of the narrative journey, this is no longer the case. If we return to the full-page Working Sketch Map (Fig. 6.6), we see that the new mountain range extends south at the river Redway (probably drawn earlier as a tributary to the Great River also coming up from the bottom left corner of the page).29 The area of Fangorn forest is sketched very loosely with a dotted circular line bottom right and later moved upward. Then, at the bottom of the page, Tolkien begins to outline zones or other regions with particular identities that will bear upon both people and topography. So, bottom right we have Rohan or ‘Horsekingsland’. The messy extension of mountains at the bottom of the page already anticipates the next version of the map as registered in the scrawled words bottom left: ‘Place this pass into Rohan further south’. On the strip map, if we compare the Working Sketch Map with its subsequent iteration (the First Map to the right of it), the latter is at a smaller scale (clearly seen by comparing the river fork at Gladden) and the mountain ridge in black now begins to dominate over the paler blue of the river. As the map takes shape, the mountain range is smoothed into one continuous run that stretches down and meets the White Mountains at the bottom. The point at which it divided in the draft version now becomes the key point of passage over or through the mountain. So, the shaping of the landscape again matches the development of the narrative. At this point, Lorien is not yet drawn, but in the next version, once the characters survive Moria and come out on the other side of the mountains, rivers flow down and through to Lorien (the next major destination on their route). Describing the Working Sketch Map, Christopher Tolkien tells us that it can, ‘with complete certainty, I think, be ascribed to the time of the original 29

This river was meant to be that running out of Mirrormere but is replaced on the later map with the river Silverlode.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

218

Tolkien’s Cartographic Imagination

writing of this chapter’ (Return of the Shadow, 438).30 He does not pin down relations between draft text and the loose sheet of paper of the map any more tightly than this, but what we see in the corresponding manuscript (a continuous long piece of writing) is that ‘My father wrote nearly all of it in ink, but he wrote extremely fast’ (415) and that this section contains a ‘remarkable outline of future events found on the back of a rejected page of the text of the Council of Elrond’ (410). This manuscript page includes comments such as ‘Gollum must reappear at or after Moria’; ‘Fangorn Forest . . . Frodo must get separated from the rest’; and ‘Fangorn is an evergreen (oak holly?) forest’ (410). It is surely at least possible to deduce that the sudden surge of active creativity and anticipation of key events yet to come is stimulated by the sketching out of key places on this map. Equally, at the end of the chapter ‘The Ring Goes South’ comes a ‘two-page Sketch of the Mines of Moria’ with comments such as ‘How far to go’, ‘How long it will take’ (442). Christopher Tolkien’s use of the word ‘sketch’ is highly appropriate here since Tolkien is literally sketching out the land, and key features, visually, and then verbally sketching the actions and events that are to occur at such points. I want to conclude this analysis of combined creativity (mapping-towrite) by looking briefly at a page of Tolkien’s creative cartography for the defensive city of Minas Tirith. The development of detailed description of place out of, and in relation to, mapped representation of natural features can clearly be seen here on a single manuscript page (see Fig. 6.9). So, from a manuscript in the Marquette J. R. R. Tolkien Collection, Tolkien writes a description of the city and its protection by the mountain (Mindolluin) behind (transcription from mid-sentence gap four lines up from first image): A strong citadel indeed it was and not to be taken by a host of men if there were any within that could hold weapons, unless some enemy could come behind and scale Mindolluin and so come behind upon the shoulder that joined . . .31

The copying-out of the narrative pauses here as he stops to visualise the description, drawing two versions of the position of the city in relation to the mountain. The first is annotated with ‘The summit of Mindolluin’ and ‘M.T’ top left but crossed through and immediately redrawn (making it

30

31

Return of the Shadow, 438. The image given here is the colour version reproduced by Hammond and Scull (Art of LOTR, 58). Marquette J. R. R. Tolkien Collection, MSS 3/1/24/3b. Hammond and Scull provide a transcription in Art of LOTR, 45.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Writing and Mapping: Rivers and Roads

219

Fig. 6.9. Redrafting of topography and description for Minas Tirith

at first appear as part of the same land mass). The mountain is then re-marked ‘Mindolluin Peak’ and ‘Minas Tirith’, with a higher protective summit and a less separated city at its base. The written narrative then continues straight on, beneath the image, picking up the text at midsentence: . . . the Hill of Guard to the mountain mass. But that shoulder which was at the height of the fifth wall was walled right up [to] the precipice that overhung it, and there stood the great domed tombs of bygone kings and lords, both at once memorials and fortresses . . . (Marquette J. R. R. Tolkien Collection, MSS 3/1/24/3b)

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

220

Tolkien’s Cartographic Imagination

At this point, a second smaller sketch literalises the written description. Tolkien draws the ‘shoulder’ extending out of the mountain at the fifth level (and then those above and beneath it) around the words ‘of bygone kings’, breaking up the writing before the final words to the right of this image: ‘if need should come’.32 Circles sketched on the defensive walled layers indicate the location of gates at each layer. On just a single page, then, we can clearly see Tolkien actively writing and mapping together, reshaping the larger land mass in relation to the highly defended city in an act of integrated visual/verbal creativity. Across the extended period of composition for The Lord of the Rings (1938–54), there are numerous similar examples of Tolkien moving between text and image directly on the manuscript page across the archive.33 This occurs at a macro-level, as we have seen with the extension of the larger map, then for detailed areas on that map, and finally for specific places or sites or structures in relation to immediate topography, as we see for Minas Tirith. In other words, Tolkien maps-to-write at multiple levels and in multiple ways privately throughout his creative process to an unprecedented degree. He is the ultimate creative literary cartographer.

Multiple Mapping in The Lord of the Rings So far in this chapter I have looked closely at Tolkien’s creative practices and map-making as an active element in his creative process. Now I want to turn to the maps that appear in the published works. There are three major levels of private and public mapping for The Lord of the Rings. Firstly, there are Tolkien's own rough sketch maps and drawing-out of details used by him throughout composition (although, as we have seen, there may be multiple stages within this). Secondly, there is a working master-plan of the whole (‘The First Map’ – see diagram below) constructed by Christopher for his father in order to provide an overview, but still only used as part of the private creative process. Thirdly, there are the maps made by Christopher from his father’s originals and published with the first and subsequent editions.

32

33

It is possible that Tolkien chooses to write the lines immediately around an already drawn image on the page but if so he writes very close to it (e.g. ‘of’). Either way, the verbal and the visual are tightly integrated. See also Macleod and Smol, ‘Visualizing the Word’, referenced above.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Multiple Mapping in The Lord of the Rings

221

In volume VII of The History of Middle Earth: The Treason of Isengard, Christopher Tolkien helpfully creates a diagram of ‘The First Map’ and describes it as ‘an evolution, rather than a fixed state of the geography’.34 Tolkien’s master-map consists of ‘pages glued together and on to backing sheets’ (Treason of Isengard, 295). The idea of ‘pasting over’ is quite common in nineteenth-century manuscripts, where an earlier copied passage may be heavily reworked and need a clean copy stuck on top, but here the pastingover is of the cartographic ‘manuscript’ not the text (see Fig. 6.10). The first map section is ‘A’ – so that the map begins by overlapping with the area covered in The Hobbit and extending that map downwards.35 The map is then extended to the north and west by section ‘B’. Section ‘C’ extends further south but also overlaps with ‘A’; and section ‘D’ is described by Christopher Tolkien as being ‘replaced over and over again, and is by far the most complex part of the map, as the region covered is also crucial in the story’ (296). This overlapping image clearly illustrates the active nature of cartographic composition for the work at a macro-level, as the map is repeatedly redrawn and refined alongside the development of the narrative. It is time to turn from the use of maps as part of the creative process by the writer to the use of maps in relation to the published text for the readers. When we consider the interrelations between the maps that are presented across the published texts we might surmise that the way different maps work together in process as part of ‘The First Map’ directly informs the intended usage of multiple maps for the readers during the experience of the text. In other words, writerly mapping informs intended readerly mapping. For the first edition of The Lord of the Rings the initial map for The Shire (see Fig. 6.1) at a larger scale was given at the start of the first volume; a general fold-out smaller-scale map of the West of MiddleEarth (‘The General Map’, see Fig. 6.5) was reproduced at the back of the first two volumes (The Fellowship of The Ring and The Two Towers); and a further fold-out larger-scale map of the imagined realms of Rohan, Gondor and Mordor was given at the end of the third volume (The Return of the King), along with various appendices (see Fig. 6.11) . How do these maps connect to each other, and to the narrative, and how are the readers expected to interpret or move between them? 34

35

J. R. R. Tolkien, The Treason of Isengard: The History of the Lord of the Rings Part Two, ed. Christopher Tolkien (Boston, MA: Houghton Mifflin, 1989), 296. The image given here is that supplied by Denis Wood and John Krygier in the second edition (not present in the third edition) of Making Maps: A Visual Guide to Map Design for GIS (New York and London: Guilford Press, 2011), 64. Wood also discusses Tolkien’s ‘First Map’ briefly in The Power of Maps (New York and London: Guilford Press, 1992), 30–2.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

222

Tolkien’s Cartographic Imagination

Fig. 6.10. ‘Christopher Tolkien’s Reconstruction of Tolkien’s Working Practices; An Annotated Version’. Reproduced by John Krygier and Denis Wood

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Multiple Mapping in The Lord of the Rings

223

Fig. 6.11. Christopher Tolkien’s ‘Map of Gondor, Rohan and Mordor’ from The Return of the King

When we compare them directly, the three published maps have contrasting styles suited to their particular purpose. The first map for ‘A Part of The Shire’ performs an important function within and beyond the epic for the reader. Its presence links this narrative spatially and geographically to the prior narrative of The Hobbit, taking the same exact location as its starting point, with Hobbiton and The Hill just off the centre of the map. So, the journey across and out of The Shire, which forms the first section of the major fantasy epic, both initiates and plays out in a more homely way the repeated pattern of the entire structure. Although Frodo emphasises that this quest is very different from that of The Hobbit – ‘This is no treasure-hunt, no there-and-back journey. I am flying from deadly peril into deadly peril’ (FR, 102) – the initial situating of the epic narrative here does implicitly create a parallel trajectory which is ultimately replicated, though in darker ways (ending with the ‘scouring’ of The Shire) in The Return of the King. Frodo’s comment also makes clear the highly spatialised nature of his journey: he is compelled to leave home, must undertake a quest and, by means of the Ring, is irresistibly drawn towards an end. The maps, especially the General Map, emphasise how far from home he has been/will be made to go. Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

224

Tolkien’s Cartographic Imagination

Whilst the larger-scale map of ‘A Part of The Shire’ is still in red and black, the font style is far plainer and simpler and both colours are unthreatening in comparison with the later maps. In some ways, this map conforms to standard plan mapping. Lettering is entered across different regions (South Farthing; North Farthing etc.) in a large, spaced way and inverted, as on a real-world map, and simple symbols are used for buildings, woods etc.36 However, in the middle of the map around Greenhill country (and at the top) map perspective shifts from an aerial view to horizontal so that we see the low hills and woods from the side. Overall then, the topography is unthreatening and the map feels homely and familiar: ‘the Shire map creates the “small, safe and understood world” that a portal-quest fantasy . . . requires for its starting point’.37 The way in which the map is presented in volume I at the end of the ‘Prologue: Concerning Hobbits, and Other Matters’ (FR, 10–25) also makes it feel part of a fictional ‘doceme’ within the secondary world. This explains that the entire narrative comes from the now lost ‘Red Book of Westmarch’ (originally four volumes containing Bilbo’s ‘private diary’ with Frodo’s story added in ‘three large volumes’) of which ‘many copies were made’ (FR, 23). Christopher Tolkien’s large fold-out maps are made visually attractive and easy to read because of the way they are presented as landform maps. Such a map has a plan base but uses a symbol for land form that adopts an oblique perspective so that the land-surface is given a vertical dimension. On the General Map this perspective is most visible in relation to wooded areas and mountain ranges where, instead of an area symbol or the use of contour lines for such features, we are given oblique views (see Fig. 6.5). The mountain ranges, in particular, stand out as threatening natural barriers, with stylised steep peaks breaking through the land beneath like teeth. Place names are entered in red in a stylised font that is hand-drawn to look like print, and the red and black contrast is also subliminally menacing. The mountains around Mordor are awkwardly (unnaturally) at right angles, creating a box around the area and visually emphasising the realisation within the text that, ‘As he gazed at it suddenly Sam understood, almost with a shock, that this stronghold had been built not to keep enemies out of Mordor, but to keep them in.’38

36

37 38

Ekman makes the point that ‘Having the Farthing names facing different ways suggests, faintly, that this map did not originally belong in a book’ (Here Be Dragons, 50). Ekman, Here Be Dragons, 50. J. R. R. Tolkien, The Return of the King: Being the Third Part of The Lord of the Rings (London: George Allen & Unwin, 1955), 176. Henceforth RK.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Multiple Mapping in The Lord of the Rings

225

By not using a key for map symbols, nor contour lines, and by presenting mixed perspectives and handwritten print, the General Map signals that this is not a modern map but something else. The ‘hand-made’ nature of the mapped representation subtly reinforces a key aspect of geography in Fantasy discussed at the start of this chapter (that it is a mappable world, but a different kind of one from that in which we live). The Shire is still represented on the General Map, to the west, but is entirely overwritten by the large lettering of the region ‘Eriador’ running across it.39 The centre of the map shows Mirkwood and the Lonely Mountain and in so doing incorporates and encompasses the earlier Hobbit map of Wilderland (as already discussed). This links the two texts cartographically whilst emphasising the change of scale for the major epic. For volume III, the larger-scale map of Gondor, Rohan and Mordor (according to Tolkien ‘5 times enlarged exactly from that of the general map’ (Letters, 210)) is much more topographical in feel and, as a result, is even more distancing and menacing than the General Map (see Fig. 6.11). Unlike the other maps, it uses only a plan perspective and represents hills and ridges in contour lines. It functions largely to locate the final part of the quest and the defensive capabilities of Mordor, emphasising the apparent impossibility of Sam and Frodo’s quest at a geographic level (ironically, of course, they themselves possess no such map). The land rises steeply in inaccessible ridges whilst the thick black ink lines of the river hint at darkness and death, as does the name ‘Mordor’ both semantically and visually. The image of the map as reproduced here – with its clear fold lines – also reminds us of the materiality of it within the book in the first edition. Tolkien emphasises the status of the map as an object in its own right by having it pasted into the back cover of the volumes. On the one hand, this encourages it to be used by the reader alongside the narrative as a map might be used in a guidebook, reinforcing the sense of it as a ‘real’ landscape to be negotiated. On the other, it allows the possibility of the map being detached literally (many maps have been cut out from such volumes in libraries) and conceptually – anticipating the reproduction of it in other forms entirely apart from the text (as discussed in ‘Post-Authorial Mapping’ below). What we see across the volumes of Lord of the Rings, then, is that as the narrative gets richer, darker and more complex, so do the maps.

39

The route of The Hobbit also appears to be marked on this map as a small dotted line leading across from The Shire through Mirkwood.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

226

Tolkien’s Cartographic Imagination

Mapping and Not Mapping The relationship between hobbits and mapping is significant. In The Hobbit, Bilbo is described as having a great liking for cartography: ‘He loved maps, and in his hall there hung a large one of the County Round with all his favourite walks marked on it in red ink’ (31).40 However, his map does not extend beyond the area that he knows from direct experience so that, in spite of his love of maps and desire to travel, he appears to have little sense of the larger world around him. When he and the dwarves set off on their adventure, Gandalf is very clearly the guide. If he has to leave the others on their own he gives them simplistic directional instructions (e.g. pass safely through Mirkwood by staying on the path) that they then fail to obey. The narrative, then, is not primarily driven by following the map, but, if anything, by the lack of a map. This sense of not quite knowing where – or who – one is, characterises Bilbo himself as well as his movement. He is reluctantly employed by the dwarves as ‘Burglar’ at Gandalf’s insistence and neither he, nor they, really know what his role will be. The dwarves’ scepticism about the usefulness of Thror’s map is echoed in scepticism about the use of Bilbo (‘“Will he do, do you think? . . . He looks more like a grocer than a burglar”’ (28)) but both prove true in the end. Bilbo rescues the dwarves on a number of occasions and, whilst he is lucky, he also instinctively seems to find the right way. When he frees them from imprisonment by floating them down the river in barrels, the reader is told: Those lands had changed much since the days when dwarves dwelt in the Mountain, days which most people now remembered only as a very shadowy tradition. They had changed even in recent years, and since the last news that Gandalf had had of them . . . The elf-road through the wood which the dwarves had followed on the advice of Beorn now came to a doubtful and little used end at the eastern edge of the forest; only the river offered any longer a safe way from the skirts of Mirkwood . . . So you see Bilbo had come in the end by the only road that was any good. (197)

Even the most reliable guides are working with out-of-date information. Bilbo also frequently takes the narrative of The Hobbit in unexpected directions. Most important of all, for this and the subsequent epic, is the way in which chance actions that occur when Bilbo is lost prove highly significant. When he is abandoned underground he reasons with himself: 40

The Hobbit or There and Back Again (London: George Allen & Unwin, 1937), 31.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Mapping and Not Mapping

227

‘“Go back?” he thought. “No good at all! Go sideways? Impossible! Go forward? Only thing to do! On we go”’ (81) and this decision leads him to Gollum and to stumbling across the ring.41 The finding of the ring occurs when he is utterly disoriented and happens unintentionally: His head was swimming, and he was far from certain of the direction they had been going in when he had his fall. He guessed as well as he could, and crawled along for a good way, till suddenly his hand met what felt like a tiny ring of cold metal lying on the floor of the tunnel. It was a turning point in his career, but he did not know it. He put the ring in his pocket almost without thinking . . . (80)

Important things can be found when one is lost. Unlike in The Hobbit, where ‘Thror’s Map’ figures as an important motivating element of the story (even if it is not always used along the way), the characters in the three volumes of The Lord of the Rings are constantly journeying but not constantly referring to a map. Gandalf is, once again, the most knowledgeable guide, with an apparently total perspective. He is able to work out his location even when disoriented: in the underground darkness of Moria he states ‘Unless I am quite astray, I guess that we are above and to the north of the Great Gates’ (FR, 332). However, the resulting over-reliance of other characters upon him puts the whole quest in danger. When he is apparently lost to them, the others are left directionless, ‘“We have not decided our course,” said Aragorn. “Beyond Lothlórien I do not know what Gandalf intended to do”’ (FR, 382). Gandalf leads them not only morally and metaphorically but also literally. The limits of a hobbit’s map-reading skills and knowledge are laid bare in The Lord of the Rings: ‘Frodo began to feel restless . . . He looked at maps, and wondered what lay beyond their edges: maps made in the Shire showed mostly white spaces beyond its borders’ (FR, 52). In fact, a hobbit’s identity is partly defined by this lack of a need to know of other places and the hobbits are the least able to map-read. Not only do they have little larger geographical or geopolitical knowledge or sense of orientation, they also have no way of measuring how much of their journey remains. The narrator comments that ‘Maps conveyed nothing to Sam’s mind, and all distances in these strange lands seemed so vast that he was quite out of his reckoning’ (FR, 299). Such a lack of wider knowledge, and of the wherewithal to use smaller-scale maps of unknown regions, considerably 41

In The Hobbit the finding of the ring appears to be a chance occurrence, but in The Lord of the Rings it is interpreted otherwise ‘“It was the strangest event in the whole history of the Ring so far: Bilbo’s arrival just at that time, and putting his hand on it, blindly, in the dark . . . Bilbo was meant to find the Ring”’ (FR, 65).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

228

Tolkien’s Cartographic Imagination

disadvantages the hobbits the moment they move beyond their own known lands. It also has the secondary effect of increasing the expression of geographical awareness by other characters, who must help them to locate themselves through description of the features around them that are visible to the naked eye. So, for example, Aragorn describes their situation in great detail to them (and to the reader by proxy): ‘But here we are not above sixty leagues, I guess, south of the Southfarthing away in your Shire, hundreds of long miles yonder. You are looking now south-west across the north plains of the Riddermark, Rohan, the land of the Horse-lords. Ere long we shall come to the mouth of the Limlight that runs down from Fangorn to join the Great River. That is the north boundary of Rohan . . .’ (FR, 397).

The hobbits must thus rely on guides who have specialist knowledge of different regions. They are led in turn by: Aragorn; Gandalf; Aragorn (again); Faramir; and Gollum. Whilst Aragorn and Gandalf have a wide range of geographical knowledge in a full cultural and historical context, Gollum, like the hobbits, knows only routes on the ground and particular areas in which he has had personal experiences. Nonetheless, whilst the hobbits’ limited map-reading skills at first appear to be a weakness, in the end it turns out to be a strength. In part it protects, even liberates, them since they cannot fully anticipate, or even imagine, the extent of the undertaking to which they have committed themselves. It also releases them into a subtly different experience of place and space, one that is both less specific and more symbolic than that of the other characters. An indistinct sense of future direction enables them to continue when someone else with greater knowledge and understanding might assume that all is hopeless and give up. So, The Lord of the Rings ultimately plays one kind of cartographic knowledge and experience off against another to surprising ends. It is those without a map, a sense of direction or an enlarged perspective who fulfil the quest, where others with far more sophisticated cartographic knowledge have failed. As they draw near to their final objective, Frodo asks Sam, ‘“How far is there to go?” “I don’t know,” said Sam, “because I don’t know where we’re going”’ (RK, 218).

Post-authorial Re-mapping One final point of interest about maps in The Lord of the Rings is the extent to which the mapped, fictional landscape threatens to displace the literary

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Post-authorial Re-mapping

229

space entirely, becoming part of a larger cultural capital. For these works, both authorial and post-authorial maps and atlases have been reproduced entirely apart from the literary text. Such a tendency was encouraged by authorial approval. Tolkien permitted Pauline Baynes (the illustrator of the Narnia books) to create a poster map for Middle-Earth in 1970 and also to produce maps that appeared alongside his own drawings as part of the 1973 Tolkien Calendar.42 At first glance, post-authorial atlases of Middle-Earth appear to exist as a kind of homage to Tolkien, in line with his cult following, but in actuality they are doing something different which is enabled by the systematic structuring of his own cartographic practices. To understand what is going on here we need to go back to Tolkien’s own influential account of Primary and Secondary worlds for Fantasy. In his essay ‘On Fairy Stories’ Tolkien put forward a definition that redetermined Fantasy as a genre in its own right – and that provided a generic classification for his own work which had previously confounded critics. He distinguished the genre from earlier fantastic forms such as Gulliver’s Travels and ‘any story that uses the machinery of Dream’ (‘Tree and Leaf’, 20) because: the vehicle of the satire, brilliant invention though it may be, belongs to the class of travellers’ tales. Such tales report many marvels, but they are marvels to be seen in this mortal world in some region of our own time and space . . . (19)

This leads Tolkien into a critique of the Coleridgean concept of ‘the willing suspension of disbelief’ for imaginative literature.43 In his original account in Biographia Literaria, Coleridge had described the need for the writer to draw upon real feeling and experience deeply enough ‘so as to transfer from our inward nature a human interest and a semblance of truth sufficient to procure for these shadows of imagination that willing suspension of disbelief for the moment, which constitutes poetic faith’ (6). Tolkien’s famous definition emerges in direct opposition to this:

42

43

Pauline Baynes worked closely with Tolkien to ensure the accuracy of the map at the centre of the poster. Their map annotations in preparation for it can be found in the Bodleian’s Tolkien archive (MS. Tolkien Drawings: 132). Tolkien’s account clearly defines itself against and through Coleridge’s account of the Imagination in Biographia Literaria – including the concepts of ‘the willing suspension of disbelief’; the distinction between Primary and Secondary workings of the Imagination; and that between Imagination and Fancy. See The Collected Works of Samuel Taylor Coleridge Vol VII. Biographia Literaria, or Biographical Sketches of My Life and Literary Opinions, ed. James Engell and W. Jackson Bate (Princeton, NJ: Princeton University Press, 1983).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

230

Tolkien’s Cartographic Imagination That state of mind has been called ‘willing suspension of disbelief’. But this does not seem to me a good description of what happens. What really happens is that the story-maker proves a successful ‘sub-creator’. He makes a Secondary World which your mind can enter. Inside it, what he relates is ‘true’: it accords with the laws of that world. You therefore believe it, while you are, as it were, inside. (40–1)

In Coleridge’s account, the writer makes the world seem ‘true’ for the reader only by an agreed (willing) compact into which both writer and reader enter. For Tolkien, this is problematic because it involves a deliberate act by the readers and any such knowingness is itself a barrier. That is, it involves a conscious splitting of the self (a ‘suspension’) – whereas the readers must feel themselves to be entirely in the imaginary realm: The moment disbelief arises, the spell is broken; the magic, or rather art, has failed. You are then out in the Primary World again, looking at the little abortive Secondary World from outside. (41)

‘True’ belief in the imagined world must be unconscious and full, because ‘if they really liked it, for itself, they would not have to suspend disbelief: they would believe’ (41; my italics). The secondary world must be as ‘real’ on its own terms as the Primary World out of which it is created. The consequence of this for Tolkien, and for his map-making (as well as in relation to the massive prehistory of The Silmarillion and the creation of Elvish and other languages) is a need for the Secondary World to be absolutely consistent and accurate within itself. It is not surprising, then, that the level of detail and nature of the literary-cartographic processes underpinning the maps for Lord of the Rings is of such a high standard that it provides what we might think of as ‘neutral’ cartographic data.44 So, for example, Karen Wynn Fonstad’s The Atlas of Middle Earth offers a ‘Pathway Table’ which records each day’s distance covered and hours travelled for the entire journey. As she puts it: ‘Only Tolkien’s tremendous attention to detail made it possible to trace the pathways so closely.’45 At this point, Middle-Earth becomes mappable in the same way as a real-world country or region is mappable: able to be mapped in different ways by a range of cartographers, reflecting changes in design over time.46 44

45

46

This is highly unusual in the context of literary mapping. As Ricardo Padrón notes, ‘Rarely do literary texts provide enough of the right kind of detail to allow us to map their worlds in conclusive, indisputable ways’ (‘Mapping Imaginary Worlds’, 265). Karen Wynn Fonstad, The Atlas of Middle-Earth (Boston, MA, and New York: Houghton Mifflin Co., 1991), 156. It should be noted that this is entirely unique in literary mapping terms.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Post-authorial Re-mapping

231

Two of the most accomplished post-authorial re-mappings of The Lord of the Rings are those of Barbara Strachey, in The Journeys of Frodo, and Wynn Fonstad’s Atlas of Middle-Earth.47 Journeys of Frodo contains fifty-one maps covering the entirety of the three volumes of The Lord of the Rings. The mapmaker is not a professional cartographer or artist. Her rationale for the collection is that it ‘serves as a companion to the three books of The Lord of the Rings, to be used whilst reading (and re-reading) the story’ (frontispage). Strachey’s maps (unlike Tolkien’s) use varying scales of 1 inch : 100 miles and 1 inch : 0.25 miles, with Hobbiton Hill as the ‘0, or Greenwich point’ (frontispage). Strachey maps time as well as place, giving ‘each day’s journey its date by the Hobbit calendar’ (frontispage) and also adding symbols for the phases of the moon. In style and design, Strachey keeps to the two-colour red and black used by the Tolkiens, but makes the maps feel more realistic and consistent by using contour lines and map symbols so that there is a continuous aerial perspective. The font style is plain and closest to the style of The Shire map, loosely corresponding to that of an early twentieth-century Ordnance Survey map. Strachey takes Tolkien’s impulse for visualisation and translates it into literary cartography of an almost literal kind. We can compare this directly to Tolkien’s original ambitions for The Hobbit. He originally intended there to be five maps in total that would ‘trace Bilbo’s journey: in “logical order, following the course of the story and they would have made a neat cartographic parallel to the text”’.48 This would have created a very different structure and experience of reading and mapping across the work. For Journeys of Frodo, then, the reader’s urges replicate those of the author and of his working practices: a desire to have the entire narrative journey represented visually through a detailed cartographic equivalent. Strachey also draws upon Tolkien’s own illustrations as well as the text. So, for example, Map 2 (‘Hobbiton to Bywater’) is based on Tolkien’s picture of the village, but converted into a mapped form (see Fig. 6.12).49 At times Strachey’s maps produce an almost abstract effect – as with Map 9 for the Barrow-downs in which the hobbits get lost in the fog. The thicker red line of the route entwines and entangles itself with the red contour lines of the Barrow. The cartographic experience of reading the text alongside the map 47

48 49

Barbara Strachey, Journeys of Frodo: An Atlas of J. R. R. Tolkien’s The Lord of the Rings (London: Unwin Paperbacks, 1981). The Art of The Hobbit, ed. Hammond and Scull, 91. Strachey also does this for other detailed paintings by Tolkien. See also Map 16. Rivendell (‘based on the three drawings made of it by Tolkien’); Map 31 for Helm’s Deep; and Map 38 for Cirith Ungol.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

232

Tolkien’s Cartographic Imagination

Fig. 6.12. ‘Map 2: Journey from Hobbiton to Bywater’ from Journeys of Frodo

differs most from that of The Lord of the Rings towards the climax of the action as Sam and Frodo journey inside Mordor. Strachey’s level of detail provides a much clearer sense, both of their slow progress day by day and of the move into deep and sinister territory – as in Map 47 which shows Sam and Frodo’s precarious and slow trudge around the interior of the Mordor ridge towards Mount Doom. Where Strachey’s maps are primarily centred on tracing the narrativeas-journey whilst incorporating a sense of time, Wynn Fonstad’s Atlas ‘incorporates the entire world-system as well as systematically mapping onto the texts of The Hobbit and The Lord of the Rings’.50 Wynn Fonstad’s account of her initial interest in Tolkien as a cartography student determines the nature of her response: ‘Immediately I developed an explorer’s need to map and classify this (to me) newfound world . . . I wished for one gigantic indexed map showing every place-name and all the pathways’ (ix). Thus, she includes: maps of the first, second and third ages from The Silmarillion; regional and thematic maps; cultural maps; maps of battle flow and troop movements; pathways; and site maps. She approaches the 50

Wynn Fonstad’s ‘Pathways’ section (156–76) towards the back of her Atlas more or less encompasses the entirety of Strachey’s Journeys (though at a smaller scale).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Post-authorial Re-mapping

233

cartographic project by addressing some of the challenges presented by literary as opposed to real-world mapping. So, for example, she notes that there is an ambiguity over whether the world is round or flat in this universe: ‘Tolkien was envisioning his world much as our medieval cartographers viewed our own. They showed the earth as a disc, with oceans around the circumference’ (ix). After the Change recorded in The Silmarillion, Tolkien’s world does indeed become round but it continues to be mapped as if it were not. Wynn Fonstad concludes that: ‘The only reasonable solution is to map his maps – treating his round world as if it were flat’ (x). In a similar way, The Hobbit proves problematic (for her and for Tolkien) because of the lack of any scale on these maps (134). Wynn Fonstad’s most characteristic and distinctive maps occur for The Lord of the Rings. These are pages that present a major map of the area from a bird’s-eye-view perspective with insets adopting other viewpoints, usually including cross-section, aerial and detail. The power of her Atlas lies in the ability to re-visualise simultaneously from multiple cartographic perspectives. This works particularly well when there are both above ground and under-ground elements (e.g. for Isengard with depictions of the tower above and below the surface). The example given here (Fig. 6.13), for the mines of Moria, gives cross-sections through the hillside, maps of the entrance and aerial views.51 The Moria maps enable users to chart fully the route of characters within the book and to make sense of the confusing underground layout. This provides far superior understanding, both to that experienced by the characters in the dark, or even that of readers of the original text. Wynn Fonstad follows the Tolkiens in employing a limited colour palette, but the use of orange and grey (rather than red and black) softens the images and makes them feel more earthy. Her handwritten fonts also feel far more modern than the maps of either Tolkien or Strachey and thus draw attention to their difference and distance from the original. These maps are far more purely cartographic than those of the Tolkiens in that their primary purpose is to enable users to fully situate themselves and understand the topography of the landscape and the structures within it. In other words, there is not really much that is ‘literary’ about such maps (as there still was in Strachey’s Journeys). By this stage, the applied real-world accuracy – that for Tolkien was such an important part of creating verisimilitude within the ‘Secondary World’ – enables entirely non-authorial re-mapping of the imaginary realm. 51

The Minas Tirith map was one of Wynn Fonstad’s earliest created site maps, sent as a sample to the publisher. See Fonstad, ‘Writing “TO” the Map’, 133.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

234

Tolkien’s Cartographic Imagination

Fig. 6.13. ‘Map of the Mines of Moria’ from Karen Wynn Fonstad, The Atlas of Middle-Earth (For the colour version, please refer to the plate section.)

It is a small step from such re-mappings to Peter Jackson’s 2001–2003 films of The Lord of the Rings, in which the map also plays a prominent part.52 52

For critical discussions of the film, see Brian Rosebury, Tolkien: A Cultural Phenomenon (London: Palgrave Macmillan, 2003); Janice M. Bogstad and Philip Kaveney eds., Picturing Tolkien: Essays on Peter Jackson’s The Lord of the Rings Film Trilogy (Jefferson, NC: McFarland and Co., 2011).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

Post-authorial Re-mapping

235

Here, visualisation of the map forms part of the title sequence and an interactive map is provided under ‘Middle-Earth Atlas’ in the Special Extended DVD Edition. Because of its non-referential nature, MiddleEarth is easily able to be transferred from a vague authorial location in the West Midlands (corresponding to an idealised pre-war rural England) to Peter Jackson’s New Zealand birthplace.53 In fact, this is exactly what the director articulates in an interview on the DVD: If you live in New Zealand and you read that book you . . . somehow feel that you are surrounded by Middle-Earth. It’s not a conscious thing but you’re just aware of the landscape and you read the book and the book is all about landscape and somehow the two . . . are related in some form.54

The ‘world’ of The Lord of the Rings is able to be ‘authenticated’ and grounded not through reference to the setting of the text or the birthplace of the author, but through the set for the making of the film. So New Zealand touristic sites can make claims such as: ‘Hobbiton Movie Set Tour: Discover the real Middle-Earth’ (in Auckland) or The Telegraph can have the headline: ‘Ten Epic Middle-Earth locations that really exist in New Zealand’.55 At this point the relationship between Secondary and Primary world essentially inverts. The imaginary landscape is not merely imitated and replicated in the real world but starts to act upon and shape the larger culture as ‘literary meanings continue to be appropriated into new contexts and in the process, transformed’.56 This occurs directly in the form of literary tourism (its primary articulation) which demands a real-world correspondence for fictional place and space, but also indirectly in the playing out of nostalgia, idealism and escapism onto a particular landscape. In the global culture of the twenty-first century, literary place and space can transfer from one location to another, and even from one side of

53

54

55

56

Tolkien states of The Shire that ‘It is in fact more or less a Warwickshire village of about the period of the Diamond Jubilee’ (Letters, 230). The Shire is commonly compared to Sarehole near Birmingham The ‘Diamond Jubilee’ is that of Queen Victoria (1897). The Lord of the Rings: The Return of the King (Special Extended Disc Edition). Disc 3: The Appendices Part Five: ‘Designing and Building Middle-Earth’, 0:31-38 seconds. www.newzealand.com/uk/feature/the-lord-of-the-rings-trilogy-filming-locations/, accessed 13 September 2018; www.telegraph.co.uk/travel/destinations/oceania/new-zealand/articles/reallife-new-zealand-filming-locations-that-starred-in-the-hobbit-middle-earth/, accessed 13 September 2018. Shelagh Squire, ‘Landscapes, Places and Geographic Spaces: Texts of Beatrix Potter as Cultural Communication’, GeoJournal 38.1 (1996): 75–86; 78. This article provides an excellent account of the ways that texts, contexts and cultural communication intersect. See also Shelagh Squire, ‘The Cultural Values of Literary Tourism’, Annals of Tourism Research 21(1994): 103–20.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

236

Tolkien’s Cartographic Imagination

the world to another, as Middle-Earth becomes a shared culturally worldwide phenomenon. Tolkien’s cartographic imagination is so powerful, and so internally consistent and accurate, that the fictional geography of the literary text can be reattributed to any corresponding, real-world landscape.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.007

7

Fearing the Map: Representational Priorities and Referential Assumptions

It is to be regretted that no mental method of daguerreotype or photography has yet been discovered by which the characters of men can be reduced to writing and put into grammatical language with an unerring precision of truthful description. Trollope, Barchester Towers Sooner or later, . . . the effort of mapping is interrupted by an encounter with the unmappable. J. Hillis Miller, Topographies

In his satiric essay ‘On the Impossibility of Drawing a Map of the Empire on a Scale of 1:1’ (1982), Umberto Eco considers the challenges of such a task with Swiftian attention to detail, including the practicalities of folding and of cartographic self-representation: Now suppose that each subject grasps a bit of the edge of the map and begins folding it . . . The subjects must [therefore], as the folding gradually proceeds, leap instead outside the map and onto the territory itself, where they will continue folding from outside . . . When the map is installed over all the territory (whether suspended or not) the territory of the empire has the characteristic of being a territory entirely covered by a map. The map does not take account of this characteristic, which would have to be presented on another map that depicted the territory plus the lower map. But such a process would be infinite (the ‘third man’ argument).1

By taking all too seriously the possibility of its existence, Eco’s essay mocks both the mindset or culture that might desire to produce a 1:1 map, and the inherent representational contradiction that such an object must be.

1

Umberto Eco, ‘On the Impossibility of Drawing a Map of the Empire on a Scale of 1:1’, in How to Travel with a Salmon and other Essays, trans. William Weaver (London: Secker & Warburg, 1994), 103, 105.

237

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

238

Fearing the Map

The well-known distinction between map and territory that Eco draws upon here was originally made by Alfred Korzybski (who also provided the concept of necessary reduplication): A map is not the territory it represents, but, if correct, it has a similar structure to the territory, which accounts for its usefulness. If the map could be ideally correct, it would include, in a reduced scale, the map of the map; the map of the map of the map; and so on, endlessly . . .2

Evidently, the map’s difference from the world it represents is an essential part of its nature and value: it must be analogous – ‘of a similar structure’ – to the land mass but not identical. In contrast, the paradoxical map that is fully identical to the land it represents pushes a fallacious desire for absolute accuracy (in the form of direct correspondence) to such an extreme that the map loses its core function and identity. The metaphor of the 1 : 1 map has, famously, been given high theoretical status by way of its reproduction (in the form of one of Borges’s short stories) in the opening chapter of Simulacra and Simulations by Jean Baudrillard (1981).3 Here he employs it as ‘the most beautiful allegory of simulation’ (1) in ways that prove highly relevant to this chapter.4 The earlier distinction between ‘map’ and ‘territory’ served to define the act of representation in terms of correspondence between object and world in a way that ‘stems from the principle of the equivalence of the sign and the real’ (6). However, simulation involves ‘models of a real without origin or reality’ (1) concerned only with ‘signs that dissimulate that there is nothing’ (6). Baudrillard’s radical disruption of the relationship between object and referent, sign and signifier, has the end result that ‘The territory no longer precedes the map, nor does it survive it . . . the map . . . engenders the territory’ (1). Unsurprisingly, this is exactly what is predicted in the example given by Eco: ‘the empire fulfills its own most secret dream, that of making itself imperceptible to enemy empires; but . . . it would become imperceptible to itself as well’ (106). The end result is to turn the 2

3

4

Alfred Korzybski, Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics, 1933, 5th ed. (New York: Institute of General Semantics, 2000), 58. Korzybski also states: ‘If we reflect upon our languages, we find that at best they must be considered only as maps. A word is not the object it represents’ (58). The concept of the 1 : 1 map used by Borges in ‘On Exactitude in Science’, Suárez Miranda, Travels of Prudent Men Book IV, Ch. XLV, Lérida, 1658, in Collected Fictions, trans. Andrew Hurley (London: Penguin, 2004), is preceded by Lewis Carroll in Sylvia and Bruno Concluded (The Complete Illustrated Works of Lewis Carroll (London: Chancellor Press, 1982)). Jean Baudrillard, Simulacra and Simulation (1981), trans. Sheila Farer Glaser (Ann Arbor: University of Michigan Press, 1994), 1.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Reasons for the Absence of Maps

239

referential relationship between object and world on its head: ‘at the moment the map is realized, the empire becomes unreproducible’ (106). The first two chapters of this study established a critical/theoretical and then a literary-historical frame for the fictional map that, in turn, necessitated a focused exploration of maps defined by their roles within specific genres for Chapters 3 to 6. In these final two chapters, the study shifts focus once more. The final chapter will focus entirely on the experiential role of the reader in order to explore the nature of the inner readerly map and the new kinds of literary mapping offered by digital space. This chapter begins to move towards such a position by considering writerly concerns about visualisation of realist space and place in the nineteenth century and the effects of this on readerly imaginative engagement as well as on the presence or absence of maps alongside the text. Why do maps occur so frequently in popular genres but extremely infrequently in canonical texts, especially the realist novel?5 What underlying assumptions lead to this position? Where such maps do occur, how do they function? These are the questions that need to be addressed. Literary mapping is unusual in comparison with maps in other disciplines in that the question of why a map is absent, or is withheld, can be of as much interest as its presence.6 This issue will be considered here through two linked debates around the value of illustration and the nature of realism in nineteenth-century French and British realist authors. The second half of the chapter will then analyse two realist authors (Trollope and Hardy) who are highly unusual in that they do use a map as part of the creation of an imaginative region (Barsetshire and Wessex).

Reasons for the Absence of Maps Before entering into more theoretical territory, it is necessary to clarify practical issues of production: how easy was it to reproduce a map alongside a text in the nineteenth century? In the Victorian period, advances in printing and publishing led to the explosion of new forms for fiction (such as the serial magazines) and the development of lavishly illustrated printed

5

6

It should be borne in mind that the realist novel was a ‘popular’ novel in its own time – but this is all the more reason why more maps should occur far more frequently in this form than they do. Cf. Harley’s comment that: ‘that which is absent from maps is as much a proper field for enquiry as that which is present’ (‘Silences and Secrecy’, 86). He is talking about what is on the map though, not the map itself.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

240

Fearing the Map

books. These developments also bore upon the printing of maps. At the start of the nineteenth century, paper-making was still a costly hand-made industry ‘able to produce white paper only from white rags because there was no means of bleaching pulp’.7 However, the development of a machine by Donkin and Fourdrinier (patented 1801) that was capable of making 600 lb of paper in twenty-four hours, revolutionised the industry. David Smith tells us that ‘by 1830 half Britain’s paper was machine-made and by 1860 machines produced 96 per cent’ (30). At the same time, laws around Stamp Duty (originally designed to control seditious libel) reduced across the period and were abolished in 1855; tax on paper halved in 1836 and was abolished in 1860; while a network of cheap distribution was considerably enlarged by the creation of the railways. All of these factors ‘made possible the mass-production which brought cheap maps within the financial reach of the vast majority of a population hungry for them’ (30). The development of lithography was also significant, being much cheaper than engraving and allowing a map to be drawn directly (not in reverse) onto limestone by a transfer technique using specially prepared paper. This process – ‘fairly well established by 1825’ – enabled images to be copied from earlier engravings and reproduced either fully or in part and combined with other images and letterpress.8 Although formal Ordnance Survey maps were deliberately undecorative and utilitarian in design, by the 1830s and 40s a popular form was to combine a plain, functional map with a far more decorative vignette to the extent that: ‘on occasion, the map actually became subsidiary to the topographical or architectural print itself, being used merely to locate the subject of the print within its neighbourhood or area’ (52).9 It is clear then, that, from the 1830s onwards, there was nothing, in terms of production processes, to stop a map image being produced alongside or within another illustration or text.10 Having laid aside technical factors, we need to consider forms of publication and their bearing upon possible visualisation of map images 7 8

9

10

David Smith, Victorian Maps of the British Isles (London: B. T. Batsford Ltd, 1985), 30. Walter W. Ristow, ‘Lithography and Maps: 1789–1850’, in Five Centuries of Map Printing, ed. David Woodward (Chicago and London: University of Chicago Press, 1975), 77–112; 93. See, for example, Reuben Ramble’s Travels through the County of England (London: Darton & Clark, 1845–50). ‘Reuben Ramble’ was the pseudonym of Samuel Clark who presented a series of maps of British counties for children, in which the map is inset within an illustrated vignette (lithograph). Another possible restricting factor on the use of maps alongside texts might be thought to be copyright law around the reproduction of Ordnance Survey maps. However, the Ordnance Survey had quite a relaxed attitude towards plagiarism and use of their data during the nineteenth century (Smith, 33–4).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Reasons for the Absence of Maps

241

alongside the text. In Victorian Novelists and Their Illustrators, John Harvey states: The traditional novel had little place for pictures. In the 1830s the novel proper would normally come out in three or four small volumes. Occasionally there might be an elegant frontispiece, but usually there were no illustrations at all. And for two decades more the three-volume novel continued to resist illustration.11

Even where a novel had previously been published with illustrations in serial form (as with Trollope below) these were often reduced in the reprinted book version, while a writer who had happily published with illustrations for one work in serial form would still not use them in a subsequent three-volume novel.12 The exact reasons for this are unclear, but we can assume that they are implicitly to do with the requirements of the anticipated audience and assumptions around the need for full visual representation. Those able to afford books were likely to be the more educated middle classes who did not need the attraction and stimulus of illustrations and may even have been insulted by them. This also seems to hold for library users. Paul Goldman describes the purchase only of ‘the unillustrated “Three-Decker”’ by Mudie for his circulating libraries and states, ‘there is no sign of illustrated books being available’ (60). Traditionally the ‘three-decker’ declared its seriousness, and the educated nature of its readership, in part by being unillustrated. In a general sense, this begins to explain the absence of maps for realist novels: if all forms of visualisation are denied to the three-volume novel then so is the map. But what of the absence of maps in serial publication of novels? Serials were designed to make novels affordable to a much wider readership by being published in episodes over time. Here there were certainly illustrations, because: the proprietors of illustrated periodicals destined for readers further down the social and economic scale knew how crucial illustrations were in maintaining support among the less confident newly literate. (62)

Harvey confirms that ‘It made little difference to the first edition of a three-volume novel whether there were illustrations or not. But with

11

12

John Harvey, Victorian Novelists and Their Illustrators (London: Sidgwick and Jackson, 1970), 8. For example, for The Small House at Allington although the full-page illustrations by Millais were kept, vignettes at the opening of chapters were removed (Paul Goldman, Victorian Illustrated Books 1850–1870: The Heyday of Wood Engraving (London: British Museum Press, 1994), 29). Equally Harvey tells us, ‘Thackeray, who illustrated his serial novels himself, had no pictures in Esmond’ (8).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

242

Fearing the Map

the monthly-part novel the illustrations were not merely desired, they were needed’ (8). The breakthrough literary work which made the serial an acceptable and popular first outlet for serious novelists was Dickens’s Pickwick Papers. Mary Hamer reminds us that this was originally designed ‘by the publishers, Chapman and Hall, not as a work of verbal narrative but as a collection of prints . . . accompanied by letter press: serial pictures, not serial fiction’.13 In other words the visual was at least as important as the verbal. However, Dickens reversed the priority of image over text (increasing the amount of text and decreasing the number of images). Even so, as Harvey points out, the groundbreaking nature of this work in relation to future illustration of the serialised novel occurred almost by accident because it was not a conventional literary form: Illustration was introduced by Pickwick, and no resistance was offered because Pickwick was not regarded as a novel. Dickens himself, while writing it, referred to his next work, intended for three volumes, as his ‘first Novel’. (12)

In an illuminating discussion of the relative power of image and text in Pickwick Papers, J. Hillis Miller describes ‘the relation between Phiz's illustrations and Dickens’s novel as not so much that of text and “elucidating” commentary . . . as that of two parallel, and to some degree incompatible, expressions with somewhat different traditions as controls on meaning’.14 Miller also relates Henry James’s recorded response to earlier illustrations of Dickens by Cruikshank: ‘In A Small Boy and Others he observes that for him when he was a child Dickens’s Oliver Twist meant more the powerful etchings by Cruikshank than Dickens’s text’ (70). The visual images have the potential to usurp their verbal equivalent, to which they ought to be subsidiary (because without them they would not exist at all). In the light of the importance and value of illustrations to serial publication of the novel, we must still ask why, when it was perfectly possible to present a map (a sketch map as illustration or a more decorative or formal map) alongside the text in this format, this was not done.15 In Chapter 4 we

13

14 15

Mary Hamer, Writing by Numbers: Trollope’s Serial Fiction (Cambridge: Cambridge University Press, 1987), 8. J. Hillis Miller, Illustration (London: Reaktion Books, 1992), 101–2. One of the first uses of the serial form as a means of publishing an expensive work over time was actually cartographic: ‘Before 1725 only about twenty titles had been brought out in this manner. Among them was Herman Moll’s Atlas Geographicus: it came out over the nine years from 1708–17 and made up five volumes when complete. It was the first book that had been

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Reasons for the Absence of Maps

243

saw how the use of house plans in newspaper reports of crime directly fed into the use of such plans in literary crime fiction, yet the same visual crossfertilisation does not occur for the serial or the novel – where the boundaries of the form of publication prove far stronger. If we take Dickens as an example, we find that despite the detailed visual and verbal representation of urban spaces in works such as ‘Sketches by Boz’ or Oliver Twist, despite his knowledge of London streets gained from walking them with Inspector Field as recorded in Household Words and his own insomniac night-walking, Dickens never in all the illustrations of his novels provides a map directly alongside the text.16 The only example I was able to locate of a map within an illustration occurs for chapter 21 of Martin Chuzzlewit. The illustration is of Martin’s visit to the agent’s office for the Eden Settlement, where he is deceived into purchasing land (see Fig. 7.1). The map representation – ‘a great plan which occupied one whole side of the office’ – reduces and falsifies reality by idealisation.17 It displays such a flourishing ‘architectural city . . . faithfully depicted in the view before them’ (354) that Martin (with architectural ambitions) is at first concerned ‘“that there’s nothing left for me to do”’ (354). The image is entitled ‘The thriving City of Eden, as it appeared on paper’ (a subsequent image being ‘The thriving City of Eden, as it appeared in fact’). The map is clearly used to gull investors into buying property in this ‘earthly paradise’ (362) by creating an illusion of a well-planned and perfectly ordered space, that in reality remains a diseased swamp. The map functions within the illustration as part of its meaning and, as such, the limits of map visualisation as opposed to illustration are also made clear. The clean, geometric lines and shapes of the map mislead the innocent map-reader depicted within the image. Equally, the map form cannot compete with the detail and dynamism of the illustrated human scene, in which the one who reads the map is clearly being taken in by the disreputable figures who encourage him to do so. The sole use of a map image by Dickens thus serves to subsume cartography within illustration in order to discredit it.

16

17

compiled expressly for publication in monthly parts, and it combined the dignity of a specialist study with the advantages of hire-purchase’ (Hamer, Writing by Numbers, 2). A useful comprehensive source for Dickens’s illustrators and illustrations is found at: https:// charlesdickenspage.com/illustrations-sketches.html. Accessed 19 December 2018. Charles Dickens, Martin Chuzzlewit (Oxford: Clarendon Press, 1982), 353.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

244

Fearing the Map

Fig. 7.1. ‘The thriving City of Eden as it appeared on paper’ from The Life and Adventures of Martin Chuzzlewit

Fearing Illustration Having dismissed practical reasons for the absence of fictional maps alongside realist novels in the nineteenth century, we need now to take into consideration the broader, underlying debate around visualisation at this time that re-emerged along the lines of ut pictora poesis (the classical

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Fearing Illustration

245

comparison of painting and poetry as ‘sister arts’ of mimesis).18 This debate surfaces once more in the nineteenth century, re-ignited by the invention of photography and the use of photographs in place of graphic (etched) illustrations (which superficially appears to make the visual representation more direct, more ‘true’, than the verbal). Two key concerns emerge as relevant to fictional maps: first, a concern about the inability of visual representation to adequately correspond to the richness and depth of the verbal (and thus the danger of reduction); second, anxiety about the effect of visual representation alongside the verbal on the internal mental actions of the reader. In Flaubert and the Pictorial Arts, Adrianne Tooke provides a fuller context for this nineteenth-century anxiety. Tooke makes clear that ‘the point of writing, for Flaubert, was “faire voir”’ (71) – ‘to show’ or ‘make seen’.19 Tooke quotes from Flaubert’s Souvenirs: ‘Quand on écrit . . . on se compose des tableaux qu’on voit’ (71) (‘When one writes . . . one composes the scenes which one views’). A doubled process of visualisation is envisaged: writing is the medium that allows translation of the writer’s internal vision into words by means of which the reader can subsequently internally visualise for him or herself.20 Flaubert is interested in the imaginative capacity that words hold (for both writer and reader): ‘the capacity to “faire rêver”’ (27) (to ‘make [one] dream/imagine’). Tooke tells us that: Flaubert is drawn to enigmatic art, of the kind which evokes a feeling he describes as ‘comme un souvenir de choses que je n’avais pas vues.’ (27)21

In other words, it is part of the essence of communicating through language that it is not absolute and precise because this allows space for the reader to actively internalise and personalise the world of the text.

18

19

20

21

Ut pictura poesis (‘as is painting so is poetry’ or ‘as it is in painting so let it be in poetry’, after Horace) assumes a likeness between visual and verbal forms of representation centred upon a third object in the world. By the eighteenth century a naturalistic emphasis on imitative reproduction (mimesis) was increasingly replaced by an anti-pictorial model which shifted the valuing of the visual inwards into a more psychological response to an idealised or imagined version of reality. Adrianne Tooke, Flaubert and the Pictorial Arts: From Image to Text (Oxford: Oxford University Press, 2000); 71. My translation. Trollope’s narrator discusses something similar in Barchester Towers (though with the opposite desire to achieve direct correspondence, although possibly with a degree of irony – see epigraph to this chapter). “[L]ike the memory of things that I have never seen.” My translation.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

246

Fearing the Map

A similar position is articulated by Henry James in his ‘Preface’ to the New York Edition of The Golden Bowl (1909). The author should ideally reduce the reader to: such a state of hallucination by the images one has evoked as doesn’t permit him to rest till he has noted or recorded them, set up some semblance of them in his own other medium, by his own other art.22

However, while it is perfectly acceptable for readers to want to externally reproduce evoked images for themselves (active), it is not acceptable to passively rely on images externalised from another’s (the illustrator’s) imagination. It is perhaps in Flaubert’s response that we find the best clue to the absence of maps as a consequence of deeply held principles around the nature of literary representation. He strongly resists seeing photography as an art form in its own right. Tooke describes him as ‘extremely reluctant to have his image taken . . . no doubt through some instinctive fear of the image’ (57). This leads her to conclude that: ‘Flaubert's well-known dislike of the art of illustration stems . . . from his resistance, instinctive as well as intellectual, to this idea of fixity, of a one-to-one correspondence between text and image’ (75). Tooke’s account describes Flaubert’s position very much in terms of accuracy in ways that draw to mind the map. For him, ‘“Accuracy” [here] masquerades as truth’ (56). It is also worth comparing Flaubert briefly with Stendhal, whose position in relation to readerly imaginative engagement also values indirectness for the reader. In Vie de Henry Brulard (1835) he ‘asserts that it is crucial not to describe key moments in detail – for then too little is left to the reader’s capacity to dream and imagine’.23 Like Flaubert, Stendhal engages directly with the problem of ‘referential illusion’ though in a much more active, creative way, involving maps. His posthumously published autobiographical novel (Vie de Henry Brulard) is filled with hand-drawn sketches of Grenoble (the place of his birth and childhood) but also of various key sites. These maps are most commonly plan views, but also involve allegorical maps, itineraries, panoramas, affective maps.24 It is important to

22

23

24

Henry James, Literary Criticism: French Writers; Other European Writers; The Prefaces to the New York Edition (New York: Library of America, 1984), 1326. Alison Finch, ‘Reality and Its Representation’, in Cambridge Companion to the French Novel from 1800 to the Present, ed. Timothy Unwin (Cambridge: Cambridge University Press, 1997), 36–53; 47. See Patrick Bray, The Novel Map: Space and Subjectivity in Nineteenth Century French Fiction (Evanston, IL: Northwestern University Press, 2013), 51.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Fearing Illustration

247

remember that this is an unfinished, unpublished, work. Thus, such maps can only really be considered as ‘working maps’ since we do not know if Stendhal intended to publish them (although they seem deeply integral to the meaning of the work). Vie de Henry Brulard is a work which from the start plays with concepts of ‘truth’ and ‘accuracy’ in ironic ways in relation to self-representation and memory. In the first chapter, ‘Henry’ asserts the need to recall his own past self honestly and without delusion and declares: Quel encouragement à étre vrai, et simplement vrai, il n’y a que cela qui tienne. (What an encouragement to be truthful and nothing but truthful, that’s all that counts.)25

It is not entirely clear whether the hand-drawn plans found alongside the text are the spontaneous product of an authorial need for externalised visualisation in relation to his own past self – which Stendhal repeatedly articulates and questions – or form part of an ironic attempt to capture the truth through rational, controlled means (though with the certain knowledge that it cannot be captured). So he repeatedly makes comments such as: Je me figure l’evénement, mais probablement ce n’est pas un souvenir direct, ce ne’st que le souvenir de l’image que je me formai. (44) (I can picture the event, but probably this isn’t a direct memory it is only the mental image of the memory I formed. (56))

As Anders Engberg-Pedersen (whose interest in Stendhal is primarily focused on later maps relating to the Napoleonic Wars) drily notes: ‘Stendhal’s mental maps are by no means self-explanatory’ (Empire of Chance, 213). On the one hand, these visual images create a kind of order (purport to be objective, in control), but, on the other, their content and the verbal discussion of them often pulls against this. As Patrick Bray puts it: ‘They seem to correspond to the textual impulse for autobiographical truth and objectivity, while often revealing its limits’ (42). Thus, the constant movement between visual and verbal projections of the protagonist’s own past memories inVie de Henry Brulard problematises the act of selfrepresentation. Bray tells us: ‘Representations (writing, drawing, maps, engravings) enter into the mind and replace memory, threatening the self 25

Stendhal, Oeuvres intimes de Stendhal: La Vie de Henry Brulard, ed. Henri Martineau (Paris: Librairie Gallimard, 1955), 9; Stendhal, The Life of Henry Brulard, trans. John Sturrock (New York: New York Review Books, 1995), 10.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

248

Fearing the Map

for the construction of the subject’ (58). He concludes his examination of Stendhal by raising a question concerning the relationship between map and text that is highly relevant for this study: Do the maps in Vie de Henry Brulard refer to the ‘real’ Grenoble or to the Grenoble in the narrative, or even just to the map itself . . . The image stands independent of the text, but participates in the network of meaning. (35)

In this fictional-autobiographical work, the presence of multiple (often repeated and redundant) maps alongside the text works to destabilise self-representation. All of this takes us back to the earlier question of what exactly is feared in the representation of a map alongside the text. From the debate above we can identify the following: a fear of the visual impacting upon or undermining the verbal meaning to which it purports to correspond; a fear of being called to account for the ‘accuracy’ or truth of the representational form because the apparent naturalism of the map referent brings this into play; a fear of visualisation acting negatively upon readerly imaginative internalisation. We might go further and suggest that underlying these surface fears, only half-sensed, perhaps, in the nineteenth century, is the possibility of inversion between the representational and the real; of a loss of representational priority.

Realist Principles and the Absent Map ‘The word evokes. The illustration presents’ (Illustration, 67). With these words, Hillis Miller concisely sums up the arguments being made above. Language is a form of representation concerned with evocation; it is indirect rather than direct; it must be concerned with indeterminacy, rather than over-determination. From such a position, we can easily see why the idea of representing the space and place of a novel by means of a map given alongside it would be anathema to such writers. It appears to fix the space and to attempt exactly the kind of 1 : 1 correspondence of fictional place and space to its representation in the text (or to the ‘real world’ supposedly behind the text) that Flaubert resisted. The issue of a direct referential relationship to the world is in play for realism as it is not for other literary genres. Pam Morris tells us: There is a popular and somewhat paradoxical assumption that realist fiction is to be judged according to how faithfully it corresponds to things and events in the realworld. The more exact the correspondence, the more a one-to-one concordance can

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Realist Principles and the Absent Map

249

be recognised between words and world, the more the realist writer is to be praised for having achieved his or her aim.26

Such an assumption can be traced back to core statements by major commentators in France: The realists’ encompassing motto is succinctly summarised in Balzac’s claim at the beginning of Le Père Goriot that ‘All is true’ which echoes the definition given in the Mercure de France of 1826: ‘la litterature du vrai’ (‘the literature of the true’).27

In Britain, George Lewes, writing in the Westminster Review, declared that: Art always aims at the representation of Reality, i.e. of Truth; and no departure from truth is permissible, except such as inevitably lies in the nature of the medium itself. Realism is thus the basis of all Art and its antithesis is not idealism, but Falsism.28

The obvious danger of assuming too transparent a position in terms of the correspondence between a representation and its ‘original’ is that it leaves the representational mode open to criticism on the grounds that it ‘practises a form of dishonesty, veiling its status as art to suggest that it is simply a copy or reflection of life’.29 However, this is to do realist writers a great disservice since it is clear that they were well aware that the relationship between medium and message was deeply problematic.30 Famously, George Eliot in Adam Bede evokes the metaphor of the mirror only to undermine it: [M]y strongest effort is to avoid any such arbitrary picture, and to give a faithful account of men and things as they have mirrored themselves in my mind. The mirror is doubtless defective; the outlines will sometimes be disturbed, the reflection faint or confused.31

By using the image of a mirror that is not external but ‘in my mind’, Eliot immediately qualifies the mimetic model she draws upon, rendering it subjective (and thus not required to be a direct or ‘accurate’ reflection of reality). 26 27 28

29 30

31

Pam Morris, Realism (London: Routledge, 2003), 5. Lilian R. Furst, ed. Realism (New York: Longman, 1992), 2. George Lewes, ‘Realism in Art: Recent German Fiction’, Westminster Review 70 (1858): 488–518; 492–3. Morris, Realism, 97. For example, note the qualification even in Lewes’s statement above: ‘except such as inevitably lies in the nature of the medium itself’. George Eliot, Adam Bede, ed. Carol A. Martin (Oxford: Clarendon Press, 2001);, bk II, ch. XVII, 164–5.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

250

Fearing the Map

In France, Guy de Maupassant offers the most intelligent critique of realism in his ‘Preface’ to Pierre et Jean (entitled ‘Le Roman’ (The Novel)) when he draws attention to the necessary craft involved in any attempt to directly represent reality: Raconter tout serait impossible, car il faudrait alors un volume au moins par journée, pour énumérer les multitudes d’incidents insignifiants qui emplissent notre existence.32 (It would be impossible to narrate everything because it would need at least a volume per day to enumerate the multitude of insignificant incidents which fill our existence.)

As Maupassant points out, the realist novel (like a map) must select and simplify in order to work. By its very nature it must reduce: ‘Un choix s’impose donc – ce qui est une première atteinte à la théorie de toute la vérité’ (xiv) (‘A choice is therefore imposed – which is the first attack upon the theory of the whole truth’). From this he concludes that ‘Faire vrai consiste donc à donner l’illusion complète du vrai’ (xiv) (‘To show the truth consists therefore of giving an illusion of the complete truth’) and he suggests that realists should really be called ‘illusionists’: ‘J’en conclus que les Réalistes de talent devraient s’appeler plutôt des Illusionnistes’ (xv) (‘I conclude that talented Realists should call themselves illusionists’).

Why a Map Is Not an Illustration [A] critic once remarked to me that nothing could give such reality to a tale as a map of this sort: & I myself have often felt the same thing.33

Is fear of the map valid? Does the presence of a fictional map undermine the illusion of the text to which it corresponds? Why couldn’t the opposite be true, as Hardy suggests above? Surely the presence of a convincinglooking map could work equally well to increase verisimilitude. In his ‘Preface’ to The Golden Bowl (defending the presentation of frontispiece photographs by Alvin Langdon Coburn for the New York edition), Henry James famously expressed concern over an unequal competition between 32

33

Oeuvres complètes de Guy de Maupassant, Vol. 19: Pierre et Jean (Paris: Louis Conard, 1929), xiv. My translation. Thomas Hardy, letter to Smith, Elder & Co. October 1878, Collected Letters Volume I: 1840–1892, ed. R. L. Purdy and Michael Millgate (Oxford: Clarendon Press, 1979), I, 61. (Henceforth CL.)

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Why a Map Is Not an Illustration

251

verbal and visual forms of representation, defending the ‘intrinsic virtue’ of verbal pictures against illustration: Anything that relieves responsible prose of the duty of being, while placed before us, good enough, interesting enough and, if the question be of picture, pictorial enough, above all in itself, does it the worst of services.34

For James, the presence of an explicit visual representation alongside the verbal implies a lack since the novel must intrinsically visualise through words rather than relying on some other medium. Perhaps surprisingly (in an argument against illustration), James goes on to make the point that, whilst illustration is unacceptable, the use of photographic images alongside the verbal in ‘as different a “medium” as possible’ does not pose a problem. This is because photographs do not ‘keep or . . . pretend to keep, anything like dramatic step with their suggestive matter’ (1327). Each medium respects the boundaries of the other. Logically, following from such a position, a map should also be exempt from the charge of overly explicit visualisation since (as noted at the start of this chapter) it registers both similarity to and difference from the ‘suggestive matter’ of its verbal equivalent. If James is satisfied that the presence of a photographic image alongside the text presents no threat or difficulty this is surely even more true of a map. A fictional map is not a direct translation of verbal into visual form but one that demands a fresh act of interpretation – it needs to be read – and requires the reader not just to look at it but to make an active correspondence between the two forms of representation (map and text). In his discussion of ‘Mapping Imaginary Worlds’, Ricardo Padrón makes just this point: ‘Mapping involves interpretation, and interpretation always contains an irreducibly subjective component.’35 Anthony Pavlik goes further to argue that ‘literary maps actually reveal their full potential not just in direct association with a text but through the reader’s imagination’.36 Where the map exceeds the limits of the narrative this is particularly true: the eye runs up against these vanishing points that articulate the known with the unknown. They create openings for our imaginations to travel beyond the geographies that the map provides.37 34 35

36 37

James, Literary Criticism, 1326. Ricardo Padrón, ‘Mapping Imaginary Worlds’, in Maps: Finding Our Place in the World, ed. James Akerman and Robert Karrow (Chicago, IL, and London: University of Chicago Press, 2007), 256–87; 265. Pavlik, ‘Special Kind of Reading Game’, 28. Padrón, ‘Mapping Imaginary Worlds’, 279.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

252

Fearing the Map

The map opens up spatial meanings beyond those actually depicted in the text. As Anthony Pavlik points out, ‘in theory at least, each novel has a potential for multiple paths (and therefore multiple possible narratives); there are always other paths not followed by the protagonist(s)’.38 The map allows the reader to develop an independent, personalised experience of literary spatiality precisely because it is not a direct representational equivalent but is only ‘analogous’ and ‘of a similar structure’.39 Pavlik continues: ‘the literary map creates its own territory, and its reference is as much to its own identity as it is to the accompanying text’ (38). What does this leave us with? Not literature as an illusion of reality but as an ‘illusion out of illusion’: A work of literature is rather a link in a chain of transformations and substitutions . . . the incessant displacement of figurative by literal and of literal by figurative which takes place within a text in its transactions with the social world and with its readers . . . This chain of substitutions and transformations creates illusion out of illusion and the appearance of reality out of illusion, in a play of language without beginning, end, or extra- linguistic foundation.40

What is true for the realist novel is also true for the fictional map. The paradoxical 1 : 1 map with which this chapter started pointed to the ways in which the map form can never be truly mimetic. Baudrillard states: Today abstraction is no longer that of the map, the double, the mirror or the concept . . . It is the generation by models of a real without origin or reality: a hyperreal. The territory no longer precedes the map, nor does it survive it. It is nevertheless the map that precedes the territory – precession of simulacra – that engenders the territory.41

The fictional map works in just this way.42 Because it has no real existence, it endlessly self-replicates (as we have seen in many examples across this study) and draws attention to its own fictionality. This also means that, for the reader, the fictional map functions in an ‘evocative’ way that acts upon readerly imagination and engagement in a way far more similar to verbal

38 39 40

41 42

Pavlik, ‘Special Kind of Reading Game’, 37. Korzybski, Science and Sanity, 58. J. Hillis Miller with David Borowitz, ‘The Fiction of Realism’, in Charles Dickens and George Cruikshank (Los Angeles: William Andrews Clark Memorial Library Publication, 1971), 1–69; 39. Simulacra and Simulation, 1. See also Pavlik, who notes that ‘the creative reader of maps in fact re-maps the territory’ (‘Special Kind of Reading Game’, 38).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Mapping Realism: Trollope

253

evocation than visual illustration, leaving room for the reader to engage actively. The fictional map works in just this way but with one vital difference: it constantly seeks to defy its own fictionality and create its own origin point in the world. We see this in the desire to create external material facsimiles of maps described within the text (Rider Haggard; Tolkien); in the transposition of fictional maps to real-world sites (The Hundred-Acre Wood of Winnie-the-Pooh); in the separation of map from text so that it can be published independently in its own right (The Map of Middle-Earth); in the actions of literary tourists re-inhabiting the fictional space out in the world and desiring ‘accurate’ correspondence between real-world (e.g. Wessex) and fictional place and space. The fictional map knows itself to be at two steps removed – the representation of a representation – yet it repeatedly seeks to make itself real. Thus, it possesses a heightened selfconsciousness in relation to the act of representation. So, it may be that one major reason for the absence of a map in the realist novel is that it doubles the central problem of realism (as a false representation of a false representation). Its presence draws unwelcome attention to that problem and to the way in which (whether Flaubert likes it or not) realist fiction is, like a map, necessarily reductive. The apparent verisimilitude of a map given alongside the text reminds us that what we are reading is no more real than the cartographic visualisation of it.

Mapping Realism: Trollope Up to this point, the chapter has explored possible reasons for the absence of maps in realist fiction. The final sections of the chapter turn now to two British realist writers – Anthony Trollope and Thomas Hardy – who do use maps in relation to the writing of a series of novels set in the same region. What makes these writers buck the trend and explicitly map realist space? The first thing that is immediately clear is that, for both writers, a complex mapping out (of region, character and class/power dynamics generated from relations between the two) is enabled by an enlarged model of literary exploration over time. A repeated return to the same region creates a greater need for a map (both for the writer planning out the enlarging fictional space and trying to keep it consistent and for the reader moving across it). A totalising model for realist fiction was first created by Honoré de Balzac who also introduced the device of recurring characters across novels

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

254

Fearing the Map

(as Trollope was the first to do in England).43 In his ‘Avant-Propos’ to La Comédie humaine (1842) Balzac undertakes ‘d’en raconter l’origine, d’en expliquer brièvement le plan’ (‘to relate its origin, and briefly explain the plan’).44 He argues above all for ‘l’unité de composition’ (xxv) (‘unity of composition’), following a scientific model for organising characters into social types in order to explore across the representation of literary place the effects of environment upon man in ‘a form of fictional interconnection that was as much spatial as sequential’ (Wall, ‘Trollope, Balzac’,123). Nevertheless, although Balzac maps out his entire social network on a massive scale, (involving an estimated 2,472 imaginary characters (123)), he does not actually make a map of it: his hold on the social, temporal and spatial connections between characters remains (impressively) internal.45 The absence of an actual map by the author himself creates a space that is then able to be filled 155 years later by Franco Moretti. It is no surprise that Franco Moretti should declare in Atlas of the European Novel: ‘What can I do, I like Balzac better than Dickens’ (124). Of course he does. Balzac provides Moretti with the perfect raw materials for the kind of narrative morphology he seeks to practise. Moretti can use the realist framework of nineteenth-century Paris (and its environs) that provides the setting for Balzac’s social experiments on environment and literally map out the structures in relation to types that Balzac creates – in terms of narrative, movement, trope and motivational forces (both internal and external). Young men are sucked into the capital from the provinces and then orient themselves around a desire for money and power that sees them move from the poor Left Bank towards the rich north of Paris: ‘the magnetism of desire “orients” the city along the axis . . . from the rive gauche, towards the ‘beehive’ in the north-west’ (95). Moretti deduces from this ‘the basic matrix of Balzac’s narrative: two poles, and a current . . . from one to the other’ (95). This, in turn, is a remapping of Propp’s

43

44

45

Stephen Wall, in ‘Trollope, Balzac, and the Reappearing Character’ (Essays in Criticism 25 (January 1975): 123–44), notes that ‘Thackeray’s novels interconnect but they do so in ways more peripheral than substantial’ (129). See also Walter M. Kendrick, ‘Balzac and British Realism: Mid-Victorian Theories of the Novel’, Victorian Studies 20 (1976): 5–24. Oeuvres complètes de Honoré de Balzac. La Comédie humaine. Vol. I., ed. Marcel Bouteron and Henri Longnon (Paris: Éditions Louis Conard, 1953), xxv. My translation. Balzac is interested in mapping, though. Anders Engberg-Pedersen argues that interest in the ‘representability of war’ leads into an enquiry into ‘the limits and possibilities of literary representation’ for Balzac that is embodied in his attempts to write La Bataille. He wants to create a ‘vision of a total literary cartography’ that is unachievable but out of which a double perspective (both aperspectival and situated) enables a new ‘simulated experience’ (Empire of Chance, 194).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Mapping Realism: Trollope

255

Morphology of the Folktale that ‘rests on the existence of two antithetical spaces from whose opposition arise all the fundamental events of the plot’ (107).46 Balzac’s masterplan – his verbal mapping-out of his strategies and aims for the whole – frees Moretti to pull out the underlying human geometry inherent in it at a macro level.47 Trollope is no Balzac, yet his slowly emerging model of interconnected narratives and characters is loosely comparable, as is his narratorial stance as ‘chronicler’ and his retrospective reflection upon the fictional totality, once it is gathered together into a whole. In his introduction to the collected Chronicles of Barsetshire (1878–9) Trollope discusses the relationships between these books, asserting that, apart from the first two, they were not ‘intended to be in any sequence, but . . . in all respects separate’.48 This decision was (echoing Balzac’s account) based on his observation of failure in creating a successful series by prior writers such as Scott and Thackeray: ‘I had taught myself to believe that few novels written in continuation, one of another, had been successful’ (425). So, Trollope creates instead a model of interconnected works that occupy the same region and share the same macro-tensions between region and capital, but across which the focus shifts to different properties and families. This allows minor characters in one novel, or one part of the county, to come to the centre in another. The critic Ellen Moody argues that mapping is in fact central to the entire nature of realism in Trollope. She goes so far as to suggest that Trollope ‘visualized the intersection of hierarchy with social behaviour and individual psychology . . . through movements of characters from site to site on a coherent map’ and that ‘In Framley Parsonage we literally move from one group of characters and their place to another as we read each chapter.’49

46

47

48

49

Moretti’s ultimate point is that Balzac triangulates the binary model through ‘The Third’ which is ‘the figure of social overdetermination’ (108). For a digital attempt at Mapping Balzac, see: http://blogs.memphis.edu/mappingbalzac/. Accessed 28 December 2018. Anthony Trollope, Barchester Towers (Oxford: Oxford University Press, 2014), appendix 1, at 425. ‘Mapping Trollope or Geographies of Power’, 4; www.victorianweb.org/authors/trollope/ moody3.html. Moody’s argument is compromised by the fact that Framley Parsonage is written in blocks of three chapters for serial publication and the fact that her statement is only true for certain sections of the book. See also Lucy Sheehan, ‘Anthony Trollope’s “English Tale, on English Life with Clerical Flavour”’, Fortnightly Review (18 September 2011): http:// fortnightlyreview.co.uk/2011/09/trollope-sheehan.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

256

Fearing the Map

It may be helpful to consider briefly a localised example of Trollope’s mode of spatial enlargement across the novels, by examining it at work in Barchester Towers. In this book, the shift from volume I to volume II is also a spatial move from the Bishop’s Palace (with the corresponding characters of the Proudies and Mr Slope) to nearby Plumstead Episcopi (the Grantlys and Mr Arabin). The spatial focus of the novel moves away from the centre (Barchester close) of the centre (the city of Barchester) to the Archdeacon’s house and then expands further eastwards through the device of the Archdeacon helping prepare Mr Arabin’s new house at Ullathorne. This introduces another property (Ullathorne Court) and established family (the Thornes) within Barchester Towers, which anticipates the introduction of Dr Thorne in the next book of the series (as a ‘second cousin’ of these Thornes living further east again at Greshamsbury). In this way Trollope builds outward from the centre and from the known to the unknown for both writer and reader. Gradual enlargement also means that in the early books a larger sense of the region can only be gestured towards because it remains extremely vague even for the author. So, for example, we are told in Barchester Towers that Lady de Courcy ‘lived in the county and within a sort of extensive visiting distance of Barchester’ (vol. III, ch. 3, 354) and that (when she uses the poor roads as an excuse for her delayed arrival and early departure) ‘no lamp could withstand the jolting of the roads of East Barsetshire. The De Courcy property lay in the western division of the county’ (vol. III, ch. 8, 397). Trollope cannot be more specific because he has not yet mapped out the region for himself (although these early gestures towards a West/East division anticipate the need for it). Trollope’s Barsetshire, then, is essentially an accretive space which I have attempted to show here by means of a choropleth map that visualises the zones of enlargement in terms of shading (see Fig. 7.2).50 The mapping-out of place across books takes on the form of a structure of radiating circles as Trollope enlarges from The Warden (1855) – focused on a small story centred around one individual and his daughter on one primary site (Hiram’s Hospital by the Cathedral Close at Barchester) – to Barchester Towers which takes the same characters and places but enlarges the powerplay and petty politics involved in the cathedral city (the close at Barchester being ‘the closest of all closes’) – and finally, in Framley Parsonage, to the entire imaginary county.51 The model can be easily delineated by mapping 50

51

See James R. Kincaird, The Novels of Anthony Trollope (Oxford: Clarendon Press, 1972). See particularly ch. 4, ‘The Barsetshire Chronicle’ (92–142). Anthony Trollope, Framley Parsonage (Harmondsworth: Penguin, 1984), 228.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Mapping Realism: Trollope

257

Fig. 7.2. Choropleth map of four Trollope novels mapped onto Michael Sadleir’s Map of Barsetshire

the zones covered by each novel onto Sadleir’s simplified transcription of Trollope’s own map of Barsetshire (Fig. 7.2).52 Trollope finds his way as he writes spatially, and there is a sense in which the model allows him to build confidence as he uses extension out of known regions, houses and people into new areas: ‘by placing Framley Parsonage near Barchester, I was able to fall back upon my old friends Mrs Proudie and the archdeacon’ (Autobiography, 92). Where for Hardy (as we shall see) the nature and tone of different kinds of topography strongly impinges upon the actions and events of characters in his novels, in Trollope, people are primarily invested in place through human ownership and power (or loss of it) (reflecting an underlying class distinction). For the most part, Trollope centres description of topography upon property and the corresponding family. He always introduces the most significant building within each region with care: ‘Gatherum Castle is a very noble pile and, standing as it does on an eminence, has a very fine effect when seen from many a distant knoll and verdant wooded hill’ (195).53 This is done primarily to signal the local hierarchy and control 52

53

Circles in order of size enlarge from: The Warden; Barchester Towers; Dr. Thorne; Framley Parsonage. The last two books position themselves against this space: just outside it in the adjacent region (South-East?) for The Small House at Allington; and in the unfashionable, marginalised North for The Last Chronicle of Barsetshire. Anthony Trollope, Dr Thorne (Harmondsworth: Penguin, 1991), 195.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

258

Fearing the Map

of place and space from that site. The difference between Trollope and Hardy, then, is inherent in Trollope’s own comment in his Autobiography that: ‘I knew all the great lords and their castles the squires and their parks, the rectors and their churches’ (98).54 Movement and the spatial dimensions of existence are directly tied into sites of regional power, and lives are reshaped if one is cast out.55 Works enlarge on setting not in terms of a distinct sense of place itself, but place as representative of a certain kind of attitude, of values and of societal dynamics. For Barchester Towers, the act of mapping Church and State politics, and thus power, onto an imaginary regional site was ready-made. As Robin Gilmour explains in his ‘Note on the Church’: Barsetshire is both an imaginary county in the south of England and a diocese, or administrative district, of the Church of England. The head of the diocese is the bishop whose see . . . Barchester is . . . bishops are nominated by the Crown, which in effect means the government of the day. . . . The Bishop in turn has power of patronage or preferment. (xxxii)

The full alignment of layers of power-play with regional space gives Trollope the unified structure he requires for the first full novel of Barsetshire (and allows him to link Whig/Tory politics to the changing nature of the Church (e.g. Dr Proudie is appointed as a liberal over the Archdeacon as a traditional Tory). However, in subsequent novels, as he moves away from the clergy, a different kind of secular power structure needs to emerge. It is at this point that he starts to enlarge his imaginary geographic realm and introduces the idea of regional division. The third book, Dr Thorne, thus opens with a full description of the county, explicitly introducing the reader to ‘some particulars as to the locality’ (5) and describing a ‘county in the west of England very dear to those who know it well’ (5). The move from the cathedral town and its ‘clerical aristocracy’ (5) to the gentry is also explicitly made: ‘In other respects, the greatness of Barsetshire depends wholly on the landed powers’ (5). Larger party politics now act directly upon the shaping of the region in terms of the integrated model of power/family/house: Barsetshire, however, is not now so essentially one whole as it was before the Reform Bill divided it. There is in these days an East Barsetshire and there is a West

54

55

Anthony Trollope: An Autobiography and Other Writings, ed. Nicholas Shrimpton (Oxford: Oxford University Press, 2014). 98. Vertical class structures act upon horizontal spatial structures often in ways unfair to the female gender. So, Mary Thorne and Lucy Robarts are exiled from the big house when the heir to the estate in each case wishes to marry them even though they have appropriately refused them.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Mapping Realism: Trollope

259

Barsetshire; and people conversant with Barsetshire doings declare that they can already decipher some difference of feeling, some division of interests. The eastern moiety of the county is more purely conservative than the western . . . the residence of two such great Whig magnates as the Duke of Omnium and the Earl de Courcy in that locality in some degree overshadows and renders less influential the gentlemen who live near them. (Dr Thorne, 6)

Across the series, then, Trollope gradually enlarges the scope and extent of Barsetshire in a way that reaches full expression in his fourth novel, Framley Parsonage, where the centres of regional control in East and West Barchester are pitted against each other and aligned to larger political positions (West Barchester is Liberal/Whig and East Barsetshire solidly Tory). This is the point at which Trollope himself moves from implicit and purely interior mapping to fully externalised visualisation. The point at which his map is made is significant in relation to the representation of place across a realist totality. The making of a map for his own use is described by Trollope in his Autobiography: Of Framley Parsonage I need only further say, that as I wrote it I became more closely than ever acquainted with the new shire which I had added to the English counties. I had it all in my mind, – its roads and railroads, its towns and parishes, its members of Parliament, and the different hunts which rode over it. I knew all the great lords and their castles, the squires and their parks, the rectors and their churches. This was the fourth novel of which I had placed the scene in Barsetshire, and as I wrote it I made a map of the dear county. Throughout these stories there has been no name given to a fictitious site which does not represent to me a spot of which I know all the accessories, as though I had lived and wandered there. (98)

Because Trollope only makes the map for himself and does not publish it (even though he describes it here), the true originality of his decision to do so for the British realist novel is lost, and he does not get the credit he deserves as a precursor to Hardy. Nonetheless, it is easy to see why Trollope makes the map at this point, for this novel. In Framley Parsonage he tells us: There is a parish called Hogglestock lying away quite in the northern extremity of the eastern division of the county – lying also on the borders of the western division. I almost fear that it will become necessary, before this history be completed, to provide a map of Barsetshire for the due explanation of all these localities. Framley is also in the northern position of the county, but just to the south of the grand trunk line of railway from which the branch to Barchester strikes off at a point some thirty miles nearer to London. The station for Framley Court is Silverbridge, which is, however, in the western division of the county. Hogglestock

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

260

Fearing the Map

Fig. 7.3. ‘Anthony Trollope’s Sketch Map of Barsetshire’ from Michael Sadleir, Trollope: A Commentary

is to the north of the railway, the line of which, however, runs through a portion of the parish. (185)

The verbal description of the novel can be compared directly with Trollope’s working sketch map (Fig. 7.3) to reveal more about his conception of place. For example, on his own map (a detail not accurately reproduced by Sadleir) we can see that there is redrawing around the Western boundary that allows for Hogglestock to be ‘on the border of the Western division’ – potentially significant when we consider moral sympathies for the region and in relation to its inhabitants. The map also makes clear the way in which the literary region is only explicitly situated in relation to one realworld place – London – emphasising its importance. This is signalled through the train line (the point of most direct and rapid contact with a larger world) which points to the right, ‘To London’.56 While the main line is set in a known real-world space, a branch line leads off it to the principal town of the imagined region. The new train line runs along the north of the 56

Ronald Knox’s later version of the map adds ‘To Exeter’ pointing left, but this is not authorial. See R. A. Knox, ‘A Ramble in Barsetshire’, London Mercury 5 (1922): 378–85.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Mapping Realism: Trollope

261

imaginative region, in contrast to ‘The Old Coach Road’ which passes right through the middle of it with almost all the main houses and places featured in the novels set along it or close to it, as we might expect. The layout of the routes across the county thus reflects changing modes of transport, but we might also note that Trollope’s own interest shifts across the series from south to north, from wealth to poverty. The only place situated remotely is Gatherum Castle – perhaps reflecting the fact that the Duke has his own gravitational force allowing him to ‘gather’ others in.57 Barchester is the only site where the two major forms of transport (old and new) intersect in the region as a whole. Faint larger lines by Trollope and a dotted line down the middle indicates the East/West split – which is also written onto the map in capitals (W.B; E.B). Why does the narrator ‘almost fear’ the need for a map and Trollope not provide a published version of it for the reader? Apart from the underlying principles of realism considered above, we might consider such a fear to be a reflection of theories of novel-writing at the time. If ‘the highest form of novelistic art effaces all signs of craft’, then the working map of an author might be understood to be part of such hidden work, associated with the process of creation, even if published in a cleaned-up version alongside the text.58 Whatever the reason, despite describing his making of the map in his posthumous Autobiography, Trollope never published it, even with the Chronicles of Barsetshire of 1878–9 (when we might have expected him to do so, alongside his account of the writing of the series in the Preface). The map was only brought to public attention for the first time in 1927 with the publication of Michael Sadleir’s The Trollope Companion which reproduced and transcribed it as well as giving two later reconstructions by others.59 As is characteristic of early critical attempts at literary mapping, the dominant approach of critics to the map was literary-touristic, centred

57

58

59

Again, Trollope’s own sketch differs from Sadleir’s in that Courcy was also originally set at a distance but later moved closer to the road. Mary Poovey, ‘Trollope’s Barsetshire Series’, Cambridge Companion to Anthony Trollope, ed. Carolyn Dever and Lisa Niles (Cambridge: Cambridge University Press, 2010): 31–43; 33. Michael Sadleir, Trollope: A Commentary (London: Constable & Co. Ltd, 1927). Sadleir provides maps by Spencer Van Bokkelan Nichols and Ronald Knox as well as a facsimile of Trollope’s own working map (reproduced above) but fails to reference this accurately: ‘Opposite to them was Trollope’s own map, which was drawn as he declared and was found three or four years ago among some papers’ (153). See also more recent Trollope maps (of Allington, Barchester Close, Bragton, Dillsborough, Framley, Greshamsbury, Guestwick and Ullathorne as well as Barsetshire) made by Florence W. Ewing and published in A Guide to Trollope, ed. Winifred Gregory Gerould and James Thayer Gerould (Princeton, NJ: Princeton University Press, 1968).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

262

Fearing the Map

Fig. 7.4. Embryo map from George Eliot ‘Quarry for Middlemarch’ notebook

upon its ‘accuracy’ in relation to the world and with a repeatedly expressed desire to ‘improve on’ the authorial sketch.60 Before leaving Trollope for Hardy, it is worth noting in passing that George Eliot (a good friend and admirer of Trollope’s work), drafted an embryo map of Middlemarch in a notebook entitled ‘Quarry for Middlemarch’ (MS Lowell 13, f 57 [seq 117] Houghton Library, Harvard) (see Fig. 7.4). Oh, what might have been. It is perhaps no coincidence that Eliot also acknowledged Trollope’s importance to her. She is reported as saying that she was: ‘not at all sure that, but for Anthony Trollope, I should ever have planned my studies on so extensive a scale for Middlemarch’.61 That said, as Ralph Pite points out, ‘geography matters less in her work than we might expect’.62 In contrast to the specificity of place in Hardy, or the sense of a region in Trollope: ‘Characteristically, a setting is given near the beginning of her novels and then its geographical position is allowed to slip into the background. The local becomes the world in miniature’ (70).

Mapping Realism: Hardy Like Trollope, Hardy worked for some years with a private personal map of his fictional region of Wessex but, unlike Trollope, he did go on to publish 60 61

62

See Knox, and Lance O. Tingay, ‘Mapmaking in Barsetshire’, Trollopian 3 (1948): 19–32. Comment to Eliza Lynn Linton, in T. H. S. Escott, Anthony Trollope: His Work, Associates and Literary Originals (London: John Lane 1913), 185. Ralph Pite, Hardy’s Geography: Wessex and the Regional Novel (London: Palgrave Macmillan, 2002), 70.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Mapping Realism: Hardy

263

it and, in so doing, had a significant impact on the real-world place to which that region purported to correspond. His example enables us finally to test directly those earlier fears: does the public presentation of a map of realist space and place interfere with, or draw attention to, problems of representation? Does it act negatively upon readerly imagination? Does it encourage readers to undertake a misplaced search for ‘accuracy’? The case of Hardy shows us that all of these fears are true, to a large extent – yet what we find is that they in no way diminish the literary work but rather enhance the author’s reputation and ultimately enrich and complicate the representational relationship between world, text and map. The first thing we need to establish is Hardy’s own sense of the relationship between the fictional world and the actual one. The best attempt to address this is that undertaken by Ralph Pite in Hardy’s Geography. Pite puts the problem simply and clearly when he states that: [T]reating Hardy as someone who records particular places and times, using invented names as only a thin disguise, leads people to read the novels too literally . . . When the match is absent or faulty, such critics condemn Hardy for inaccuracy. (2)

Instead he argues for a more sophisticated understanding of literary place and space than that commonly allowed: Hardy's locations still tend to be understood in these ways within either a realist framework or an anti-realist one. Both are mistaken, I believe. The difficulty lies in describing exactly how Hardy and his locations are linked together. (3)

For Pite, then, Hardy’s sense of place is not only complex but ultimately held in a kind of suspension: ‘neither a separate world . . . nor a world vaguely located somewhere within the real one’ (172) but ‘somewhere separate, equal and co-existent’ [to the real] (177). Is Hardy’s form of realism unique or are matters simpler than Pite allows?63 A letter to Sir Frederick Macmillan of 1911, relating to the Wessex edition of 1912, provides a useful summary of the maps from Hardy’s point of view: The Map of Wessex in the novels was first published in 1895, & is an extension of an idea I first used in The Return of the Native . . . to which was prefixed a map of 63

My position here is closer to that of W. J. Keith when he describes Hardy’s end position: ‘Resistance he seems to imply is hopeless; the man who is now devoting his creative energies wholly to poetry appears content to allow his readers to use his novels as travel-guides’ (‘Thomas Hardy and the Literary Pilgrims’, Nineteenth Century Fiction 24 (1969): 80–92; 90). See also Sara Haslam, ‘Wessex Literary Pilgrims, and Thomas Hardy’, in Literary Tourism and Nineteenth Century Culture, ed. Nicola Watson (London: Palgrave Macmillan, 2010): 164–74.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

264

Fearing the Map

Egdon Heath, the scene of the story. As to the map being an authorized picture of where the localities really are, it is not precisely so, for though I certainly drew it, it was rather unwillingly done, owing to the constant inquiries of readers for actual places. But I stated, I think in one of the Prefaces at the time it was published, that the scenes of the stories were not guaranteed to be solidly standing where shown on the map, but were only suggested more or less by places that stood there, & fully existed nowhere but in the novels themselves.64

This letter seems to me to give an unambiguous account of Hardy’s own position. He himself is under no illusion about the nature of realist representation which only exists ‘in the novels themselves’. But, at the same time, he is willing to allow for a different position being held by others (his readers) who may need a referential relationship between fictional place and real-world place. The same point is made in the ‘Preface’ to Far from the Madding Crowd when read in full (rather than taking the phrase ‘dream-country’ out of context): [T]he appellation which I had thought to reserve to the horizons and landscapes of a partly real, partly dream-country, has become more and more popular as a practical provincial definition; and the dream-country has, by degrees, solidified into a utilitarian region which people can go to, take a house in, and write to the papers from. But I ask all good and gentle readers to be so kind as to forget this, and to refuse steadfastly to believe that there are any inhabitants of a Victorian Wessex outside these volumes.65

Here Hardy again distinguishes between his personal imaginative conception of Wessex, as opposed to its real-world equivalent, emphasising its status as a literary representation. In his ‘General Preface to the Novels and Poems’ for the Wessex edition (1912) he at first seems to shift the emphasis towards a dangerously transparent understanding of realism: ‘the description of these backgrounds has been done from the real – that is to say, has something real for its basis, however illusively treated’ (46). However, later comments in that document make a clear distinction between writerly and readerly positions: Subject to the qualifications above given, that no detail is guaranteed, – that the portraiture of fictitiously named towns and villages was only suggested by certain real places, and wantonly wanders from inventorial descriptions of them – I do not

64

65

Thomas Hardy, Collected Letters of Thomas Hardy Volume IV 1909–1913, ed. R. L. Purdy and Michael Millgate (Oxford: Clarendon Press, 1984), 137. Thomas Hardy’s Personal Writings: Prefaces. Literary Opinions, Reminiscences, ed. Harold Orel (London: Macmillan, 1967), 9–10.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

Hardy and Trollope

265

contradict these keen hunters for the real; I am satisfied with their statements as at least an indication of their interest in the scenes. (47–8)

Hardy's depiction of ‘keen hunters for the real’ is surely partly ironic and perhaps implies a fruitless search, but (crucially) it is not one that he feels any need to oppose. In other words, it is not that Hardy’s ‘realism’ is of a particular order but that he is willing to allow the fictional and the actual to coexist without priority – or for the priority as understood by the writer (fictional over actual) to be reversed by the reader-tourist (actual over fictional).66 The same base position is summed up in a letter to Hermann Lea of 1904 concerning the possibility of a guidebook: ‘I do not really desire it to be done, but am quite indifferent whether it be done or not.’67

Hardy and Trollope Because of the publication of maps of his fictional region, Hardy has to engage directly with questions around referential representation in ways that Trollope is not required to do. When we compare Hardy’s mappingout of a fictional region to that of Trollope, we can see that there are certainly points of comparison: the role of narrator as a recorder of a certain way of life; the way that the imaginative region slowly emerges over time; the importance of there being a focal city within the fictional region. If Hardy has a Barchester then his is Casterbridge, but this does not function as powerfully as Trollope’s county capital (to generate the rest of the county around it) and even Simon Gatrell – who argues that it ‘becomes the focus of a circle of previous fictions’ –goes on to note of the next novel that: ‘It is somewhat surprising, in the light of this substantial contextualising of Wessex, that Hardy chose to return in The Woodlanders (1887) to an environment as enclosed and isolated as that of Return of the Native.’68 So, where Trollope gradually enlarged his regional focus from a tight core, allowing for the same characters to interact across the regional totality, Hardy operates a more discrete regionality. 66

67

68

In an essay that indirectly engages with realism, Hardy favours ‘thoughtful readers of mature age’ who are ‘weary of puerile inventions and famishing for accuracy; who consider that, in representations of the world, the passions ought to be proportioned as in the world itself’ (‘Candour in English Fiction’, The New Review, January 1890: 15–21; 16). Collected Letters of Thomas Hardy, Volume III 1902–1908, ed. R. L Purdy and Michael Millgate (Oxford: Clarendon Press, 1982), 140. Hardy does become less ‘indifferent’ over time, to be sure, but the base position holds true. Simon Gatrell, ‘Wessex’, in Cambridge Companion to Thomas Hardy, ed. Dale Kramer (Cambridge: Cambridge University Press, 1999), 19–37; 25.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

266

Fearing the Map

In his account of regionalism in Hardy’s writing, biographer Michael Millgate employs the metaphor of a map: ‘Hardy’s procedure as a regionalist was thus to scrutinise microcosmic samples, to blow up particular segments of the map into much larger scale, to insist upon the particular rather than upon the whole.’69 This results in ‘a sequence of separate exemplars’ (347). Similarly, Pite stresses the way in which across the entirety of the novels ‘Few places come up in more than one narrative’ (172), so that ‘the places marked on the map are connected with principally one novel or story’ (173). We might elicit from the comparison between Trollope and Hardy at least three kinds of spatial relationship between imagined regions and the real world for realism. The first could be described as ‘nested’ or ‘anchored’: the fictional world is created within a loose frame with reference to a real site that functions as a kind of anchor. That is, it signals proximity to a realworld location (usually a city), and even directly uses that location at times, but the fictional region remains entirely self-contained somewhere close by. Trollope uses London like this (he also references Oxford), and Jane Austen’s novels function in a similar way. In his first novel, Far From the Madding Crowd, Hardy uses Bath as the anchor – to which Bathsheba and Troy go to be married. The second spatial relationship might be described as ‘mosaic’ when the writer takes different elements of a real-world region and re-converges or alters them to make the whole – as in Ransome’s Lake District (which although a work of children’s fiction is still essentially a realist setting). The third spatial relationship is what we might call ‘overlay’ – as if the fictional map is laid on top of a directly corresponding real-world topography, or vice versa (the overlay model may also involve mosaic elements).70 Both Pite and Watson explicitly use the third term for Hardy’s mapping: [Hardy] overlays the physical reality of Dorsetshire with the parallel imagined physical reality of Casterbridge.71 It has the additional consequence of unsettling the usual hierarchy of real and fictional. The map overlays one on the other but neither is obviously prior.72

69

70

71 72

Michael Millgate, Thomas Hardy: His Career As a Novelist (London: The Bodley Head, 1971), 347. Cf. Hillis Miller’s description of his essays on literary place and space in Topographies: ‘all of them taken together may be thought of as like the transparencies superimposed in palimpsest on a map’ (Topographies, 6). As he makes clear: ‘the landscape as such is never given, only one or another of the ways to map it’ (6). Watson, Literary Tourist, 177. Pite, Hardy’s Geography, 174.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

The Earliest Map: Return of the Native

267

The overlay model involves the greatest degree of potential correspondence to real-world geography. In Hardy’s case this occurs ostensibly in terms of the outline of the land mass (corresponding to the coastline of south-west England); the presence of rivers; the use of real names for major cities in and around the region; and the identifiable fictional renaming of real towns. However, it is also worth noting the omission of rail and road and the fact that, as Pite points out, ‘the map is comparatively empty because it includes nothing which does not feature in the novels’ (171–2). With the obvious exception of Jude the Obscure, the novels are (for the most part) clustered pretty tightly in the same area of South-West Wessex. In spite of this, an effect of regional distinctiveness is still felt because Hardy chooses not to link the texts together at a level of character as Trollope did, resisting the sense of a unified whole in social terms, even as he asserts it cartographically. As Pite tells us: ‘the narratives coexist without significantly interlinking or forming a single world’ (173). This also reflects an allimportant shift of class focus from Trollope to Hardy. Trollope’s focus on houses and social events allowed for points of interconnection and encounter across the totality. In Barsetshire the middle classes have the money and leisure to wander freely across the county, while the rich can afford to move between London and the region. In Wessex the poor stay put, unless compelled to move. The contrast between Hardy and Trollope is felt most strongly in relation to Tess who, in quite a deliberate way by Hardy, wanders across the entire region, but does so alone.

The Earliest Map: Return of the Native Pite has pointed out that the dominant critical reading of Hardy in relation to his maps is negative, viewing his retrospective realignment of the fictional world to the Wessex map as proto-imperialist: Recent criticism of Hardy has, by and large, been hostile to ‘Wessex’, seeing it as a late addition, superimposed on the original variety of his novels and fitting them to the needs of his late nineteenth-century audience. (169)

In the last part of this chapter, I adopt a method of trying to re-view his mapping of place and space by largely overlooking the much-discussed later mapping of Wessex as a totality, in order to focus on earlier acts of mapping, and on alternative possible models for Hardy’s literary mapping that are stifled by increasing real-world referentiality (initiated by others). In a letter to Smith, Elder & Co. of October 1878 Hardy writes:

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

268

Fearing the Map I enclose for your inspection a Sketch of the supposed scene in which the ‘Return of the Native’ is laid – copied from the one I used in writing the story – & my suggestion is that we place an engraving of it as frontispiece to the first volume. Unity of place is so seldom preserved in novels that a map of the scene of action is as a rule impracticable: but since the present story affords an opportunity of doing so I am of opinion that it would be a desirable novelty, likely to increase a reader’s interest. I may add that a critic once remarked to me that nothing could give such reality to a tale as a map of this sort: & I myself have often felt the same thing. (CL, I, 61)

Hardy’s comments relating to this – his first-ever literary map – tell us quite a lot about his early ideas in relation to literary mapping. First, he immediately registers the difference between represented and actual place; and chooses to only focus on the former – the place of the story. Second, he makes it clear that the initial idea for publishing an explicitly readerly map (as a ‘desirable novelty’) comes from his own use of a writerly map, with some sense of a different role for the map in each case. Third, for Hardy the presence of a map exists to visually reinforce unity of place in a way that looks back to Aristotle’s Poetics (all action must occur within the same circumscribed region). The particular nature of literary place and space in Return of the Native (the centrality of the Heath) allows for such a map. And finally, Hardy views the localised map as adding verisimilitude in a way that runs entirely counter to the dominant realist position when we recall that this is being added to the first edition of a three-volume novel. Where did the idea for this sketch map – actually more original than the later regional map – come from? When we look at the illustrations by Arthur Hopkins published with the serial version of the novel in The Belgravia (1873), we find that many of these are set quite dramatically outdoors with a strong sense of atmosphere that powerfully conveys the setting of the novel. (See, for example, Fig. 7.5.)73 If we compare the illustration entitled ‘The Stakes Were Won By Wildeve’ with Hardy’s sketch map reproduced below it (see Fig. 7.6), then a similar kind of dark intensity is felt.74 That is to say, if the illustrations depict characters at key moments within the novel, out upon the heath, then the sketch map depicts the setting itself, which is to all intents and purposes a central brooding force with a character of its own: ‘the sombre scene of the story’ (‘Preface’, Thomas Hardy’s Personal Writings, 12). This first map then

73

74

For all of these images, see Phillip V. Allingham, ‘Arthur Hopkins’s Illustrations for Thomas Hardy’s The Return of the Native’, Victorian Web, November 2004: www.victorianweb.org/ authors/hardy/native/illustrations.html. Accessed 26 January 2019. Hardy’s note in the letter states: ‘In the drawing for the book it would be desirable to shade the hills more fully’ (I, 61).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

The Earliest Map: Return of the Native

269

Fig. 7.5. ‘The Stakes were won by Wildeve’, from Belgravia: A London Magazine 36.141 (July 1878)

seems to position itself somewhere between illustration and cartography – Hillis Miller describes it as ‘this map, which also seems vaguely to be a picture’ (25). The use of a low oblique angle for the hills and barrow and the pictogrammatic depiction of the houses is highly illustrative and almost allows the reader, looking from the bottom of the image upwards, to experience the heath from the same perspective as the characters: The scene before the reddleman’s eyes was a gradual series of ascents from the level of the road backward into the heart of the heath. It embraced hillocks, pits, ridges, acclivities, one behind the other, till all was finished by a high hill cutting against the still light sky. The traveller’s eye hovered about these things for a time, and finally settled upon one noteworthy object up there. It was a barrow.75

The presence of the map, then, works to draw attention to the tight bonds between movement, inhabitation, self-identity, self-situating and the surface upon which all of this occurs.76 At the same time, this map is

75 76

Thomas Hardy: The Return of the Native (New York: W.W. Norton & Co., 2006), 15. Cf. Hillis Miller who views the novel as: ‘made up of a multitude of objects and persons distributed on a topographical surface connected by a reticulation of lines’ (Topographies, 39). For analysis of journeys on foot mapped across the heath, see Sally Bushell, ‘Mapping Fiction: Spatialising the Literary Work’, in Literary Mapping in the Digital Age, ed. David Cooper, Christopher Donaldson and Patricia Murrieta-Flores (London: Routledge, 2016), 125–46. See also Scott Rodes, ‘Figuring the Map in The Return of the Native’, in his Reading and Mapping Hardy’s Roads (London: Routledge, 2008), 19–58.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

270

Fearing the Map

Fig. 7.6. Frontispiece from The Return of the Native

purely literary in a way that the later Wessex map is not; it seeks only to refer to the text to which it corresponds as a ‘Scene of the Story’, not to any supposed real-world space lying behind this. It implies a different position in terms of the relationship between literary representation and

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

The Earliest Map: Return of the Native

271

real-world places from the later Wessex map and it is presumably for this reason that it is published by Hardy in the first three-volume edition and then again in the one-volume edition of 1880 by Kegan Paul & Co., but never after this.77 One other important point suggests itself here. We know that Hardy’s publishing of the Wessex map in 1895 was at least partly defensive and concerned with copyright and that two unauthorised maps preceded Hardy’s own map of Wessex.78 Watson reminds us that it is the unauthorised map by Clive Holland that is ‘the first such map, in which real names are put in parenthesis under fictional names . . . and the whole is fitted into the outline of south-western England’ (Literary Tourist, 181). However, although critics have focused on these unauthorised maps as a stimulus to Hardy’s publishing of his own Wessex map, what we do not really know is how much effect unauthorised maps may have had on Hardy’s way of mapping or (a closely linked point) from what date he began to work directly with a map alongside the text. Gatrell suggests that it is only with the writing of Tess that Hardy starts to fully map out a larger Wessex totality (beyond that of South Wessex/Dorset) and he gives the most likely date for this: ‘we can say with some confidence that in September or October of 1890 Hardy first formulated for himself the idea of Wessex as twentieth-century readers have been accustomed to experience it’ (27). This does also mean, though, that without external pressures we can speculate that there might have been a ‘series of separate exemplars’ of a far more illustrative form reflecting the ‘discrete’ model of the mapping-out of place outlined above. Hardy’s original conception of literary mapping for Wessex back in the 1870s could have been more abstract (and more aligned to the sketch map model for Return of the Native) than the ‘overlay’ model onto a real coastline that he later adopts (following the work of others and in anticipation of the literary tourism that they initiate). In his analysis of both the general Wessex map published at the start of each novel, and the individual map for The Return of the Native in Topographies, Hillis Miller states that: ‘These maps may be thought of as the last element in the series – from the real landscapes to the maps or 77

78

See also Birgit Plietzsch, The Novels of Thomas Hardy As a Product of Nineteenth-Century Social, Economic and Cultural Change (Berlin: Tenea-Verlag Ltd, 2004), 17–18. Watson states: ‘As Keith, Kay-Robinson, Millgate, Widdowson and Pite have variously argued, to some extent Hardy’s hand was forced with regard to the need to copyright Wessex by competitor publications . . . [I]n this light the 1895 map of “The Wessex of the Novels” can seem the result of prudent defensiveness’ (Literary Tourist, 185).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

272

Fearing the Map

photographs of them, to the texts of the novels, to the maps based on the novels’ (20). In such a model, priority is strictly maintained – ‘any link presupposes the others as its determining cause’ (21) – and with the certainty of priority comes the demand for accuracy and real-world reference. However, as Miller goes on to explain: If the landscape is inside the novel, then it is determined by it and so cannot constitute its ground. The same thing may be said of the relation of any two members of the series: novel and map; real map and imaginary map; landscape and map. Each is both prior to the other and later than it. (21)

Once we accept that the model of priority is not true, that literary representation does not work in this way even if it appears to do so, then all forms are of equal status, ‘the relationship between landscape and story is rather that of figure and figure’ (52). The creation of space, making place meaningful, occurs twice over – for characters existing within that world and moving through it, and for the reader activating the text through the act of reading. Whilst the novel itself functions as a network of relations, for the reader, each reading is also a new mapping. What we find in literary works (which Hardy’s and Trollope’s novels fully exemplify) is a deeply spatialised form in which place and space are fully bound up with the central meanings and structures of the work but, more than this, in which we are potentially enabled to experience through forms of representation a richer spatiality in relation to a non-existent place than we can ourselves experience out in the world, or one that enriches our experience of that landscape in ways that would not otherwise be possible. Should we be using the term ‘spatialise’ rather than map? Spatialising texts involves mapping but extends far beyond this and allows for the possibility of mapping not just movement across and within place (geographical; topographical; physical) and space (imagined; interiorised; emotional; psychological) but also finding ways of exploring the significance of situated-being through language – which would seem to be ultimately what the mapping of literature is all about. To put it another way: literary place and space is experiential, not static: for writer, for character, for reader. The fact that there is no grounded reference point becomes liberating in terms of the two-way dynamic that opens up when we accept that, ‘The true ground, the “it,” is everywhere and nowhere. It can be located on no map . . . There is no attainable behind or within’ (52–3).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.008

8

Reading As Mapping, or, What Cannot Be Visualised

‘There is a great deal of unmapped country within us.’ George Eliot, Daniel Deronda But the map left a good deal to imagination, and she had not got any. E. M. Forster, Where Angels Fear to Tread

In his essay On Literature, J. Hillis Miller engages directly with the area of primary interest for this final chapter: the internal visualisation of the literary world – an invisible, unrealisable space. Miller states: If it is really the case, as I have argued, that each literary work opens up a singular world, attainable in no other way than by reading that work, then reading should be a matter of giving one’s whole mind, heart, feelings and imagination without reservation, to recreating that world within oneself, on the basis of the words.1

The previous chapter has shown that readerly visualisation is an essential part of literature as a communicative act, allowing words to ‘evoke’ in the minds of the readers. If we are to fully understand the spatial experience of literature we need to at least attempt to understand how readers internally visualise, map and move across that space.2 Such questions, of course, come out of a long tradition of post-Lockean philosophical thinking about the nature of ideas, but this final chapter focuses strongly on the actual (empirical) practice of cognitive reading-as-mapping for literature.3 It brings together a number of different approaches to try and illuminate the unrealisable map within. The first section is phenomenological, identifying the spatial dimensions of reader-response theory in order to re-interpret an influential account of the process of reading as a form of mapping; the

1 2

3

J. Hillis Miller, On Literature: Thinking in Action (London: Routledge, 2000), 115. The account given in this chapter tends to assume a base level of schematised internal spatialisation for the reader, but it is important to bear in mind that readers will visualise and spatialise to widely differing degrees and some readers of literature claim not to internally visualise at all (although they may still map using verbal forms). For useful accounts of earlier models see Tally, Spatiality (27–37) and John O’Keefe and Lynn Nadel, The Hippocampus As a Cognitive Map (Oxford: Clarendon Press, 1978), which provides a thorough review of prior ‘Histories of Space’ from the seventeenth century onwards (10–61).

273

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

274

Reading As Mapping

second section explores deliberate acts of mapping in ancient and medieval memory models; the third section is directly cognitive, drawing on a seminal account of the workings of the mind. Together, and cumulatively, all of these approaches point toward a double model for the internal mapping of literary place and space involving a combination of immediate ‘route-mapping’ with some form of underlying geometric base map. This map may take on a range of forms depending on the visual or verbal dominance of the individual imagination.4 The final part of the chapter emerges directly out of these earlier sections to consider the relevance of such a double model (route/base map) for the way we map literature in the digital domain. The concept of ‘cognitive mapping’ emerged strongly in the late 1970s and found expression in the humanities in part as a result of Fredric Jameson’s highly influential essay of that name. In ‘Cognitive Mapping’, Jameson sets out to initiate ‘a project for a spatial analysis of culture’ that outlines three stages of capitalism.5 The disoriented and alienated nature of the current social condition is understood to be a product of the third stage (‘late capitalism’ (350)). It is in order to explain the nature of this state that James invokes the concept of the cognitive map in relation to urban sprawl. He draws upon Kevin Lynch’s account of alienation in The Image of the City to claim that: ‘the mental map of city space explored by Lynch can be extrapolated to that mental map of the social and global totality we all carry around in our heads in variously garbled forms’ (353).6 Where, in prior stages of capitalism, there was still some overlap between individual experience and larger social and economic forces, the increasing distance between these means that ‘those fundamental realities are somehow ultimately unrepresentable’ (350) and that ‘the incapacity to map socially is as crippling to political expertise as the analogous incapacity to map spatially is for urban experience’ (353). In other words, the individual is no longer able

4

5

6

In relation to internal verbal rather than visual representation, Allan Paivio argues for a dualcoding model, suggesting that images and words are memorised differently, with a tree structure for the coding of language. He asserts two independent, but interacting, systems that work together (i.e. a verbal form can stimulate a non-verbal and vice versa). See Paivio, Mental Representations: A Dual Coding Approach (Oxford: Oxford University Press, 1986) and ‘The Mind’s Eye in Arts and Science’, Poetics 12 (1983): 1–18. Ellen J. Esrock also considers this issue in some detail across her study, The Reader’s Eye: Visual Imaging As Reader Response (Baltimore, MD, and London: Johns Hopkins University Press, 1994). ‘Cognitive Mapping’, in Marxism and the Interpretation of Culture, ed. Cary Nelson and Lawrence Grossberg (Basingstoke: Macmillan, 1988), 347–57; 348. See Kevin Lynch, The Image of the City (Cambridge, MA, and London: MIT Press, 1960). Chapter 3, ‘The City Image and Its Elements’, provides a full account of each of the five categories (46–90).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Reading As Mapping, or, What Cannot Be Visualised

275

to internally spatialise the larger forces in which he or she is caught up. Jameson takes the core problematic of the cognitive map as something ‘ultimately unrepresentable’ (350) but only as a ‘spatial analogue’ (353) for the larger societal point he seeks to make. Cognitive mapping functions figuratively in relation to a totalising force beyond the mind’s capacity to grasp it and with a primary focus on the effects of that inability. On the one hand, Jameson effectively initiates the (self-proclaimed) epistemic ‘spatial turn’ for the humanities – a shift from modernist emphasis on time to postmodern emphasis on space.7 As Marie-Laure Ryan sums up: Jameson’s concept of cognitive mapping paved the way for cultural and globalization studies, and it opened the floodgates to the mapping of anything that passes through the mind of the postmodern subject.8

The enlargement of the concept to the postmodern condition makes it potentially all-encompassing. On the other hand, this macro-concept does no favours at all to the exploration of the process of cognitive mapping itself, or to the cognitive mapping of artistic forms through inner visualisation and spatialisation. In relation to the actuality of cognitive mapping Jameson is at best patronising, at worse dismissive, as can be seen in his references to Lynch’s book as that which, ‘spawned the whole low-level subdiscipline that today takes the phrase “cognitive mapping” as its own designation’ (353). Jameson’s position (theoretical as opposed to empirical) also points to a similar division in relation to cognitive mapping in cultural and human geography where there is an ongoing divide ‘between the empiricist paradigm and the critical paradigm’.9 This is potentially resolved in a muchcited article by Kitchin, Perkins and Dodge that moves the field on from ‘a representational to a processual understanding of maps’, leading ultimately towards what has been termed a ‘post-representational’ model.10 7

8

9

10

Jameson states: ‘A certain spatial turn has often seemed to offer one or more productive ways of distinguishing postmodernism from modernism proper’ (Jameson, Postmodernism, or, the Cultural Logic of Late Capitalism (Durham, NC: Duke University Press, 1991)), 154. Marie-Laure Ryan, ‘Cognitive Maps and the Construction of Narrative Space’, in Narrative Theory and the Cognitive Sciences, ed. David Herman (Stanford, CA: CSLI Publications, 2003), 214–42; 215. See also Robert T. Tally Jr, ‘Jameson’s Project of Cognitive Mapping: A Critical Engagement’, in Social Cartography: Mapping Ways of Seeing Social and Educational Change, ed. R. Paulston (Levittown, PA: Taylor and Francis, 1996), 399–416; Tally, Spatiality, 39–42, 67–78; and Rossetto, ‘Theorizing Maps with Literature’, 513–30; 518–21. See Sébastien Caquard, ‘Cartography III: A Post-Representational Perspective on Cognitive Cartography’, Progress in Human Geography 39 (2015): 225–35; 226. Caquard, ‘Cartography III’, 229. See also Tania Rossetto’s discussion of the ‘postrepresentational’ in ‘Theorizing Maps with Literature’.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

276

Reading As Mapping

My position here is concerned with the actual practice of mapping the reading of texts in the mind and the usefulness of connecting theories of reading and mapping to empirical accounts in order to redetermine future digital models. I therefore follow Downs and Stea’s influential definition of cognitive mapping as: a process composed of a series of psychological transformations by which an individual acquires, codes, stores, recalls, and decodes information about the relative locations and attributes of phenomena in his everyday spatial environment . . .11

Or, more specifically in relation to literature (from Ryan): A cognitive map is a mental model of spatial relations . . . The space represented by the map can indeed be real or imaginary. The representation can be based on embodied experience . . . or on the reading of texts.12

What I am exploring in this chapter could also be contextualised, on the literary side, in relation to the field of cognitive literary studies and is assisted by Marco Carracciolo’s recent (2016) attempt to define the different approaches of this field, and the advantages and disadvantages of combining two such apparently disparate areas as literary analysis/interpretation and cognitive science.13 Carracciolo defines four different cognitive approaches to literature: analogical; thematic; processual; functional (202) – the last two of which would seem to be directly relevant to this chapter. The processual approach is described by him as: Quite simply, a reader-response theory revisited in the light of cognitive-scientific models and methods it consists of theorizing the act of reading in ways that are at least consistent with scientific paradigms (for instance, the embodied mind) and may be supported by experimental research. (194)

This provides an excellent summary of two of the approaches adopted in this chapter which reflect upon each other. Alongside this we can also place Carracciolo’s description of the functional approach:

11

12 13

Roger M. Downs and David Stea, ‘Cognitive Maps and Spatial Behaviour’, in Image and Environment: Cognitive Mapping and Spatial Behavior, ed. Roger M. Downs and David Stea (New Brunswick, NJ, and London: Aldine, 1973), 8–26; 9. Rob Kitchin and Scott Freundschuh also open their introductory essay to a useful edited collection on cognitive mapping with this quotation. See ‘Cognitive Mapping’, in Cognitive Mapping: Past, Present and Future, ed. Rob Kitchin and Scott Freundschuh (Abingdon: Routledge, 2002), 1–9. Ryan, ‘Cognitive Maps and the Construction of Narrative Space’, 215. Marco Carracciolo, ‘Cognitive Literary Studies and the Status of Interpretation: An Attempt at Conceptual Mapping’, New Literary History 47 (2016): 187–207.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Spatialising Reader-Response Theory

277

Rather than zooming in on readers’ experience and cognitive processing of literature, it seeks to shed light on how engaging with literary texts can play a role in broader psychological processes. (195)

Such an account is potentially relevant to the reflections on the digital domain in the final section, for which relations between an immersive environment and ways of reading and mapping texts begin to connect with theories of the mind and affect. To some extent, this heads in the direction of cognitive narratology – a newly emerging approach which essentially inverts the relationship between cognitive mapping and narrative to argue, not that stories can be mapped, but the inverse – that ‘Story is a basic principle of mind’ and that ‘Most of our experience, our knowledge, and our thinking is organized as stories.’14

Spatialising Reader-Response Theory One way in which we might be able to understand the reading process spatially is by connecting Hans-Georg Gadamer’s broader philosophical work on the nature of understanding to Wolfgang Iser’s model for readerresponse, both involving the spatial concept of an intentional horizon. Gadamer (after Heidegger) argues that the historicised nature of being means that understanding itself takes the conceptual form of a constantly anticipated new horizon: ‘This constant process of new projection constitutes the movement of understanding and interpretation.’15 In Gadamer’s open-ended model, the nature of understanding is such that it concerns leaping forward, anticipating, returning and redefining in order to advance. This larger model also applies to the act of understanding in reading, as he makes clear: A person who is trying to understand a text is always projecting. He projects a meaning for the text as a whole as soon as some initial meaning emerges in the text.

14

15

Mark Turner, The Literary Mind: The Origins of Thought and Language (New York and London: Oxford University Press, 1996), preface. See also David Herman, StoryLogic: Problems and Possibilities of Narrative (Lincoln and London: University of Nebraska Press, 2002) and Lisa Zunshine, Why We Read Fiction: Theory of Mind and the Novel (Columbus: Ohio State University Press, 2006). Cf. Robert T. Tally Jr, ‘Introduction’, in Literary Cartographies: Spatiality, Representation and Narrative, ed. Robert T. Tally (Basingstoke: Palgrave Macmillan, 2014): ‘[N]arratives are in some ways devices or methods used to map the real-and-imagined spaces of human experience’ (3). Hans-Georg Gadamer, Truth and Method [1960], 2nd ed., trans. Joel Weinsheimer and Donald G. Marshall (New York and London: Continuum, 2003), 267.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

278

Reading As Mapping

Again, the initial meaning emerges only because he is reading the text with particular expectations in regard to a certain meaning. (267)

The readers must be in a state of open receptiveness in order to allow for their preconceptions to be challenged: ‘The important thing is to be aware of one’s own bias so that the text can present itself in all its otherness and thus assert its own truth against one’s own fore-meanings’ (269). Such an account leads directly into the work of Wolfgang Iser.16 Iser seeks to get at the heart of what the reading act is by understanding reading as a shared intersubjective (dialectical) experience that is part of the communicative act for literature, to which the reader is central (in a Barthesian way): ‘it is in the reader that the text comes to life’ (19). The focus for meaning now shifts from that held within the language of the text to the act of reading as projection: ‘as meaning arises out of the process of actualization, the interpreter should perhaps pay more attention to the process than the product’ (18). The reader thus participates in both the understanding of the work and its production: ‘the reader “receives” it by composing it’ (24). The location of this meaning-making is internal: ‘ideated by the mind of the reader’ (38). The aspect of Iser’s account that makes him of such usefulness here is the spatial nature of his model. This is not a feature of which he seems to be self-consciously aware, but it is strongly present in all of the central concepts he articulates. Whilst he only writes about the dynamic inner process in terms of reading, what he describes might equally well be understood as a kind of mapping of the text: reading-as-mapping. Three core concepts at the heart of his account are all spatial: the shifting perspective; the enlarging horizon; the wandering viewpoint. Iser describes the reader’s role as: prestructured by three basic components: the different perspectives represented in the text, the vantage point from which he joins them together, and the meeting place where they converge. (36)

The role of the reader is conceived of in terms of a perspective that gradually merges through the process of internal visualisation (Iser calls this ‘ideation’). However, since the reader is not in control of the perspectives and cannot have a total grasp until the book has been read, Iser distinguishes between an immediate and larger perspective: 16

Wolfgang Iser, The Act of Reading: A Theory of Aesthetic Response, trans. Stephen Rendall (Baltimore, MD, and New York: Johns Hopkins University Press, 1978). Iser directly references Gadamer on p. 97 n. 28.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Conscious Memory Mapping

279

As perspectives are continually interweaving and interacting, it is not possible for the reader to embrace all perspectives at once, and so the view he is involved with at any one particular moment is what constitutes for him the ‘theme’. This, however, always stands before the ‘horizon’ of the other perspective segments in which he had previously been situated. (97)

The model is cumulative, with each previously read element capable of being changed by the introduction of a new element. Iser’s model of a ‘wandering viewpoint’ emerges out of this: The structure of theme and horizon constitutes the vital link between text and reader, because it actively involves the reader in the process of synthesizing an assembly of constantly shifting viewpoints. (97)

We can see, then, that the model is deeply spatio-temporal and strongly suggests the situatedness of the reader in terms of moving through a landscape that he or she creates: ‘the wandering viewpoint permits the reader to travel through the text, thus unfolding the multiplicity of interconnecting perspectives’ (118).

Conscious Memory Mapping An analysis of the spatial elements of Iser’s account begins to provide some of the elements potentially in play for reading-as-mapping by clarifying the nature of the process that informs the production of an internal readerly map. However, Iser’s account situates the process at a subconscious level (he describes the synthesis of perspectives as ‘pre-predicative and . . . subconscious, we continue to produce them throughout our reading’ (136)). It may be helpful, therefore, to compare this experiential ‘in the moment’ act of internalising place with deliberate, conscious, acts of spatial memorising. Frances Yates’s seminal work, The Art of Memory, as well as the work of Mary Carruthers in relation to medieval forms of internal mapping, prove indispensable here.17 In The Art of Memory, Yates takes us through different accounts of an ancient practice from classical times to the Renaissance, that seeks to ‘memorize through a technique of impressing “places” and “images” on memory’ (11) primarily for the purposes of rhetoric. As is well known, the origins of this idea are found in Simonides – who famously escaped a tragic

17

Frances Yates, The Art of Memory (Chicago, IL: University of Chicago Press, 1966).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

280

Reading As Mapping

feast at which the building collapsed and was able to tell relatives where their loved ones had been sitting by reconstructing the seating plan from his memory of their faces. The concept of mnemonic loci that emerged from this experience was later more fully articulated by Cicero in De Oratore (in book II of which, discussing memory, he retold the tale of Simonides) and then by Quintilian in Institutio Oratoria. The latter gives the fullest account of the core principles of such memory-placing (summarised by Yates): The first step was to imprint on the memory a series of loci or places. The commonest, though not the only, type of mnemonic place system used was the architectural type . . . The images by which the speech is to be remembered . . . are then placed in imagination on the places which have been memorized in the building. This done, as soon as the memory of the facts requires to be revived, all these places are visited in turn and the various deposits demanded of their custodians. (18)

A second key text for teaching rhetoric through the use of mnemonic loci was the Rhetorica ad Herennium which made a distinction between ‘natural’ and ‘artificial’ memory: ‘The natural memory is that which is engrafted in our mind, born simultaneously with thought. The artificial memory is a memory strengthened or confirmed by training’ (Yates, 20). This treatise (originally attributed to Cicero but actually written earlier) also makes a clear distinction between the loci (memory-places within the chosen site) and the objects placed within them: The formation of the loci is of the greatest importance. . . The images which we have placed on them for remembering one set of things fade and are effaced when we make no further use of them. But the loci remain in the memory and can be used again. (23)

This account makes it clear that the dominant memory structure (palace) and the sites within it remain fixed whilst the objects placed in those sites can be changed to meet different short-term needs. It is also worth noting that, according to Quintilian, though the method recommended the use of a real place as the site for memory-placing, later versions of memory mapping also allowed for fictional settings (‘“ficta loca” or imaginary places’ (117)) and for other settings beyond those of buildings: ‘What I have spoken of as being done in a house can also be done in public buildings, or on a long journey, or in going through a city’ (38).18 We can

18

Yates notes: ‘It is Quintilian alone, of the classical sources, who says that one may form memory places when on journeys’ (120).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Conscious Memory Mapping

281

see, then, that the human mind has always possessed the capacity to create strong internal spatial visualisations (essentially cognitive maps) of buildings and routes. Equally, although the dominant form of memory storage here is architectural (a palace; a theatre) it does not have to be. It could take the form of a journey or a stylised route through an imaginary place. In fact, Yates proposes that Dante’s Inferno potentially functions in just this way. She postulates that: ‘Inferno could be regarded as a kind of memory system for memorizing Hell and its punishments, with striking images on orders of places’ (104). Whilst Yates’s work first brought the spatial attributes of memory patterns for ancient rhetoric to public attention, her account has been developed more fully by Mary Carruthers in two related studies.19 The important early text that articulates a spatial memory-model most clearly, Rhetorica ad Herrenium, is again central, but now in terms of its revival in the late thirteenth century and as mediated through various scholastics.20 In The Craft of Thought, Carruthers re-engages with Yates’s account of memory-mapping from a different perspective. She addresses one key aspect of Yates’s account and challenges it: Yates herself believed that the goal of the art of memory was solely to repeat previously stored material: she characterized the medieval versions of the ancient art as ‘static,’ without movement, imprisoning thought. She could not have been more wrong. (9)

So, where Yates focuses primarily on the structure and building of the ‘memory palace’ itself, Carruthers shifts the focus to the use of that structure by the individual. This is possible because of the doubleness inherent even in the static architectural model, noted above – of the solid macro-space and the re-utilisation of the ‘niches’ within it in each journey round it. The base map (the plan of the building) is also a route map, as Carruthers puts it: ‘a readily reconstructable quasi-narrative scene of related figures, each of which cues a particular subject in the case’ (8). When she shifts the focus from locational memory to the mind’s engagement with the site of memory, her subject effectively becomes an early form of cognitive literary mapping.

19

20

Mary Carruthers, The Book of Memory: A Study of Memory in Medieval Culture (Cambridge: Cambridge University Press, 1990); Mary Carruthers, The Craft of Thought: Mediation, Rhetoric and the Making of Images 400–1200 (Cambridge: Cambridge University Press, 1998). Carruthers states: ‘In the late thirteenth century the architectural method enjoyed a revival, having been commended as the best method by both Albertus Magnus and Thomas Aquinas’ (155). Three appendices in The Book of Memory provide translations of central medieval scholastic texts (339–68).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

282

Reading As Mapping

The shift of focus historically as well – from the ancient to the medieval period – introduces a new dimension to memory-mapping that is highly pertinent here. As Carruthers reminds us: In ancient mnemotechnic, architecture was considered to provide the best source of familiar memory locations. Architecture also plays an essential role in the art of memory which is basic to my present study, but the monastic version of architectural mnemonic carries non-Roman resonances that make something rich and strange from the . . . set of memory ‘rooms’. These resonances, as one might have predicted, are Biblical. (16–17)

The spiritual focus of the medieval period shifts the spatial ground from a real-world setting to a textual one, thus ‘The monastic architectural mnemonic is founded, like a vast superstructure, on a key text from St Paul [1 Corinthians 3:10–17]’ (17). The medieval period takes the literal form of a building and textualises it. As Carruthers makes clear, this makes the passive model active, since it strongly personalises it, with value given to the distinctiveness of each individual’s inner mapping. Thus, just as I suggested in Chapter 2 that biblical maps could be considered as the earliest form of literary or fictional mapping so, here, early models for internalised re-mapping of the Bible offer a template for the cognitive mapping of literary texts. Carruthers describes a process of spatialising and visualising the text internally in order to ‘chart’ it, to internalise key locations represented by the images and revisit these as part of a meditation on meaning. This is, in effect, a form of internalised literary mapping, albeit for a single text – the Bible – and with a strong moral and spiritual purpose. What it shows is that we are more than capable of holding a complex internal map of fictional space in our long-term memories. We can also see that these early memory-plans lend credence to the claims of cognitive narratology since the mnemonic route with stopping places could be compared directly to a story-map. In fact, Carruthers reminds us that the base concept for cognitive narratology – that the mind employs the form of a narrative – is not entirely new. She gives the example of the thirteenth-century scholastic, Albertus Magnus, who: understood that mnemonic places are entirely pragmatic; they are cognitive schemata rather than objects . . . They should be thought of as fictive devices that the mind itself makes for remembering. (13)

There is, of course, a strong distinction to be made between conscious and deliberate acts of memory (‘fictive devices that the mind itself makes’) and

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Two Forms for the Internal Map: Route and Locale

283

an argument about the processes of cognition. Nevertheless, if one seeks to argue that the mind is structured like a story, then that story would surely have to be of a highly spatial nature, involving a locational structure that would compare, at least to some degree, to the ancient and medieval models described above.

Two Forms for the Internal Map: Route and Locale Alongside these considerations of (subconscious) reader-response and (conscious) memory-mapping we can introduce the work of French theorist, Michel de Certeau, and his account of the reader as ‘poacher’ encroaching on another’s territory in The Practice of Everyday Life. Certeau states: [T]he activity of reading has [on the contrary] all the characteristics of a silent production: the drift across the page, the metamorphosis of the text effected by the wandering eyes of the reader . . . He insinuates into another person’s text the ruses of pleasure and appropriation: he poaches on it, is transported into it, pluralizes himself in it . . . [T]his production is also an ‘invention’ of the memory . . . The readable transforms itself into the memorable . . . A different world (the reader’s) slips into the author’s place. This mutation makes the text habitable, like a rented apartment. It transforms another person’s property into a space borrowed for a moment by a transient.21

This is reading-as-mapping as illicit thrill-seeking. The account of the reader ‘insinuating’ him or herself into ‘another person’s text’ fits with Certeau’s desire to validate the ‘consumer’ as a creative agent in his own right, making meaning through the act of reading, and in so doing transforming the original. His description slips across external material space (the page) and internal mapping of it and asserts an active dynamic engagement or re-spatialising of the text by the reader. Elsewhere in the same book, Certeau draws on a fundamental distinction between two ways of understanding and moving through the world that maps onto his own distinction between two main ways of experiencing society and culture spatially, in terms of the totalising view (strategy) and the immersive street-level experience (tactic). He compares transcribed oral accounts of directions of the map-type as ‘a plane projection totalizing

21

Michel de Certeau, The Practice of Everyday Life, trans. Stephen Rendall (Berkeley, Los Angeles and London: University of California Press, 1984), xxi.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

284

Reading As Mapping

observations’ and the tour-type, ‘a discursive series of operations’ (119).22 Where the map-type is concerned with ‘seeing’ and with a static geometric representation of space from a position of total control, the narrative style of the tour or itinerary is concerned with ‘going’ (119); with enabling the individual to negotiate space and with an organisation of movement.23 Certeau defines these as ‘two symbolic and anthropological languages of space. Two poles of experience’ (119). Each means of conceptualising space is, however, bound up with the other. The tour contains within it a strong awareness of the map: ‘The chain of spatializing operations seems to be marked by references to what it produces (a representation of places) or to what it implies (a local order)’ (120). Equally, the map is strongly bound up with the itinerary for which it is required. In Narrating Space/Spatialising Narrative, Marie-Laure Ryan, Kenneth Foote and Maoz Azaryahu make the same distinction, although they draw upon two different underlying models from narratology (Stanzel; and Linde and Labov).24 The first distinguishes between ‘perspectivism and aperspectivism’ in which the tour ‘is a highly immersive representation of space’ (26), and the map ‘a position freely floating in space’ (26). The second uses exactly the same terms as Certeau (map/tour): ‘While in the tour strategy speakers look at the apartment from the inside, in the map strategy they occupy an external, elevated, static perspective’ (27). Ryan et al. conclude that far more participants use the tour than the map approach: ‘the tour is obviously the preferred strategy’ (28). If we compare these distinctions with the reader-response model outlined by Wolfgang Iser above, this affirms that the kind of mapping implicit in Iser’s model is also much closer to the tour than the map. The low perspective of the ever-enlarging ‘horizon’ suggests that the act of mapping during reading is one that is strongly immersed – involving movement

22

23

24

Michel de Certeau makes clear the nature of the two types: ‘The first is of the type: “The girls’ room is next to the kitchen.” The second: “You turn right and come into the living room.”’ (119). See also Jacobs and Schenk’s discussion of visual movement of eyes on the map in the light of Certeau’s map/tour distinction in terms of ‘the static synoptic gaze’ and ‘a mobile gaze’ (309–13). Marie-Laure Ryan, Kenneth Foote and Maoz Azaryahu, Narrating Space/Spatialising Narrative: Where Narrative Theory and Geography Meet (Columbus: Ohio State University Press, 2016), 25–9. Franz K. Stanzel, A Theory of Narrative (1979), trans. Charlotte Goedsche (Cambridge: Cambridge University Press, 1984); Charlotte Linde and William Labov, ‘Spatial Networks As a Site for the Study of Language and Thought’, Language 51 (1975): 924–39. See also Holly A. Taylor and Barbara Tversky, ‘Spatial Mental Models Derived from Survey and Route Descriptions’, Journal of Memory and Language 31 (1992): 261–92.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Two Forms for the Internal Map: Route and Locale

285

within the environment and constant adjustment to what can be seen, rather than at a distance from it. The use of oral transcripts of instructions by Certeau and others to make the tour/map distinction is also highly relevant here, since such an activity is, effectively, an externalised expression of an internal map that is being recalled and then articulated verbally. It is time to turn to empirical accounts of cognitive mapping by the brain for a fuller understanding of all of this. In Beyond the Cognitive Map (1999), A. David Redish helpfully defines a full taxonomy for navigation (in rodents).25 He identifies five route-mapping strategies in hierarchical order of increasing sophistication: random navigation; taxon navigation (using cues); praxic navigation (following the same route, knowing it will work); route navigation – a ‘chaining sequence of taxon and praxic strategies’ (7); and finally locale navigation – ‘[the animal] can learn a map . . . it can plan a path’ (7). As Redish explains, only locale mapping actually requires a constructed cognitive map and this is an ‘all-or-none phenomenon . . . The animal either knows where it is on the map or doesn’t’ (19). As his synthesis of work in the field makes clear, within these five dominant forms, different scientists propose different combinations of models. The account drawn upon in this chapter works out of that made by O’Keefe and Nadel in The Hippocampus as a Cognitive Map (1978).26 I focus on this early account because it ‘not only [was] the first unified theory of hippocampal function but also has been the most influential’ and because the core hypotheses have been tested and refined but not radically overturned.27 In their seminal account, O’Keefe and Nadel distinguish between two main forms of mapping, one behavioural and responding to external cues (route), the second involving some kind of spatial structure held in the long-term memory (locale). While a sense of relative space is created through route-mapping they also argue for internalised absolute space in the form of a place or locale map: an innate ‘cognitive mapping system’ 25

26

27

A. David Redish, Beyond the Cognitive Map: From Place Cells to Episodic Memory (Cambridge, MA, and London: MIT Press, 1999). There are two supposedly competing theories about the role of the hippocampus: the cognitive map theory and a counter-theory arguing that it supports declarative memory (recall of events and episodic memory). For an attempt to integrate the two approaches, see Howard Eichenbaum and Neal Cohen, ‘Can We Reconcile the Declarative Memory and Spatial Navigation Views of the Hippocampal Function?’, Neuron 83.4 (2014): 764–70. See also Eleanor A. Macguire and Sinéad L. Mullally, ‘The Hippocampus: A Manifesto for Change’, Journal of Experimental Psychology 142 (2013): 1180–9. Lucia F. Jacobs and Françoise Schenk, ‘Unpacking the Cognitive Map: The Parallel Map Theory of Hippocampal Function’, Psychological Review 110 (2003): 285–315; 285.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

286

Reading As Mapping

(p. 2 note) located in the hippocampus. O’Keefe and Nadel’s own primary focus is concerned with arguing for the existence of an ‘a priori Euclidean spatial network’ (6), which goes further than I would choose to go.28 However, for the purposes of this study the value of their account lies in the fact that it allows the possibility of a combined model for cognitive mapping in which route-mapping feeds into and informs the locale map. We need to consider their account of these two main forms of mapping in greater detail. Although their primary interest is not in the route map – for which activity takes place outside the hippocampus – nonetheless O’Keefe and Nadel do give quite a detailed account of how this works. Route-mapping is egocentric and involves the use of different forms of guidance including: landmarks; the goal (destination) as landmark; actual guides; streams and trails (82–3). Such a form follows a Stimulus-Response-Structure (S-R-S) and relies upon external cues. Alongside the use of external topographical features, route-mapping also draws upon orientation using such elements as a compass and the sun. The more sophisticated elements of route-mapping occur when external factors are combined in ‘path integration’ as described by Redish. If we compare this neurological account to the phenomenological one of Iser, we can immediately see that the spatial dimensions of his model do align to this kind of route-mapping. When Iser describes the text as ‘a network of response-inviting structures, which impel the reader to grasp the text’, this strongly suggests that his model is concerned with an S-R-S structure outside the hippocampus and consisting of spatially significant content held within the text.29 The act of route-mapping is active and held in immediate and short-term memory. This corresponds to each (cumulative) act of reading of a text until the whole work is finished. During the act of reading, the horizon of expectation remains open. The ‘goal’ is the completion of the whole work, at which point all information has been given and the horizon closes. This also compares to Carruthers’s account of the rhetorical ductus as a form of flow, moving around the site of memory in ‘the way that a composition guides a person to its various goals’ (Book of Memory, 78). Essentially route-mapping involves a form of way-finding, relying upon the sighting of landmarks and beacons, wherever it is

28

29

My own position would be closer to that articulated by O’Keefe and Nadel in terms of a rather fence-sitting relativist for whom ‘The notion of an absolute spatial framework, if it exists at all, is held . . . to derive from prior concepts of relative space, built up in the course of an organism’s interaction with objects or with sensations correlated with objects’ (1). Iser, The Act of Reading, 34.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Two Forms for the Internal Map: Route and Locale

287

undertaken. This might be out in the world, across a text, in memory, or in the active workings of the mind. Despite accounting for route-mapping stimulated by external cues (and potentially relying solely on these), the focus for O’Keefe and Nadel was always on the activity of the hippocampus and on proving the existence of the primary ‘locale map’ in longer-term memory. In their account, the hippocampus locale map ‘can be viewed as a set of ordered, connected places’ but also involves both a place system and a ‘misplace system’.30 The hippocampus draws upon a range of external and internal cues (landmarks, beacons, paths, guides) as well as associative memory models (subjective past experience) in combination, in order to create an internal representation.31 Later work on a ‘Parallel Map Theory’ by Jacobs and Schenk effectively formalises this argument by articulating a theory of two parallel maps (a ‘bearings’ map and a ‘sketch’ map) that combine to make a third: The concept of the cognitive map is usually assumed to be a unitary mental representation. We propose instead that the cognitive map is constructed from two parallel maps that, when integrated, allow the navigator to calculate cognitive map shortcuts.32

In their account, short term route-mapping generates a 2D coordinate system which ‘becomes the scaffold’ for ‘encoding the relative positions’ of sketch maps (296). Pulling both of these together, the integrated map ‘[r]elies on the presence of two parallel map processes’ (293). The sketch maps can then be linked together as a series of mini-maps to give an overview of a larger space.33 Whether or not an actual inner map form is generated from each mapping activity, as Jacobs and Schenk theorise, the model confirms the point about the two main ways of mapping working together: ‘the encoding of space must link exploratory bouts within the same region of space’ (301). From these underlying neurological accounts, different theories about how humans negotiate an environment emerged from the late 1990s onward. Kitchin and Blades outline three main approaches (hierarchical

30 31 32

33

O’Keefe and Nadel, Hippocampus As a Cognitive Map, 93, 2. Redish, Beyond the Cognitive Map, 102. Jacobs and Schenk, ‘Unpacking the Cognitive Map’: Lucia F. Jacobs and Françoise Schenk, ‘Unpacking the Cognitive Map: The Parallel Map Theory of Hippocampal Function’, Psychological Review 110 (2003): 285–315; 291. Jacobs and Schenk describe the sketch map as ‘a sketch of a new locale. It can be thought of as a mini-map of a subspace rather than as the map of all space’ (301).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

288

Reading As Mapping

model; route-centred model; sequential vista model) which they summarise thus: In the first theory environmental cues, such as landmarks, are seen as the fundamental building blocks . . . to which subsequent information, such as paths, is added . . . The second theory hypothesizes that path-based information forms the initial framework of knowledge and that landmarks and other information are then placed in relation to this . . . [I]n contrast to these a third theory emphasizes that wayfinding can be dependent on memorizing a series of vistas . . . made up of ordered views or scenes rather than landmarks and paths.34

The first two models assume the formation of an underlying ‘spatial database’ (35), whereas the third is strongly visual-sequential assuming later translation of the ordered views into ‘metric information and configurational knowledge’ (40). The vista model essentially argues for a perceptual mapping experience: ‘When an individual moves along a particular path of travel, and thereby adopts a continuous series of observation points, she generates an optical flow of perspective structure’ (109).35 The vista model could be seen to correspond strongly to Iser’s model of enlarging horizons during reading and provides a cognitive spatial equivalent of the phenomenological model as a dynamic map. Whichever model we support, all three effectively combine the route-mapping account with an underlying geometric structure at some stage. When we map in an unknown environment it makes sense that we rely far more heavily on external cues and route-mapping to confirm our situation. It follows from this that it is when we return to a place previously 34

35

Rob Kitchin and Mark Blades, The Cognition of Geographic Space (London and New York: I.B. Tauris, 2002), 35. This book provides a thorough synthesis of models of cognitive mapping, acquisition and form. The three theories summarised above are (1) hierarchical model: A. W. Siegel and S. White, ‘The Development of Spatial Representation of Large-Scale Environments’, in Advances in Child Development and Behaviour, ed. H. Reese (New York: Academic Press, 1975), 9–55; anchor-point theory: R. G. Golledge, ‘Representing, Interpreting and Using Cognised Environments’, Papers and Proceedings of the Regional Science Association 41 (1978): 169–204 and ‘Place Recognition and Wayfinding: Making Sense of Space’, Geoforum 23 (1992): 199–214); (2) route-centred model: T. Garling, A. Book, E. Lindberg and T. Nilsson, ‘Memory for the Spatial Layout of the Everyday Physical Environment: Factors Affecting Rate of Acquisition’, Journal of Environmental Psychology 1 (1981): 263–77; (3) sequence of vistas model: E. H. Cornell and D. H. Hay, ‘Children’s Acquisition of a Route via Different Media’, Environment and Behaviour 16 (1984): 627–41. Harry Heft, ‘The Ecological Approach to Navigation: A Gibsonian Perspective’, in The Construction of Cognitive Maps, ed. Juval Portugali (Dordrecht: Springer Publishing, 1996), 105–32.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Two Forms for the Internal Map: Route and Locale

289

experienced and re-map it that we rely more on the pre-existing interior locale map, and that this map is therefore able to be understood in greater detail through the act of return. O’Keefe and Nadel’s original theory allowed for a repeated return to the hippocampus map: ‘The map itself can become richer and more distinct . . . but it is not altered in any fundamental sense with repeated exposures to the same environment’.36 Similarly, Redish argues that: The primary role of the hippocampus in the navigation domain seems to be to allow an animal to reset its internal coordinate system from external cues when it returns to a familiar environment.37

This is exactly what we would expect for reading-as-mapping. On an initial reading of a text, ‘route-mapping’ involves encountering all of the mappable elements (narrative, plot twists, key events and scenes, affective foci, relative human connections, character interactions etc.) for the first time in sequence, thus making it impossible to fully spatialise the whole until the end. It is only on rereading that we can make use of the long-term locale map in the hippocampus. The rereader can then rely on the cognitive map to navigate through the imaginative space, drawing on ‘external cues’ (events or key sites within the text) as well as on pre-existing internal cues – recall of what is going to happen or the significance of a particularly meaningful section. This is partly why, as readers, we are able to find rereading a satisfying process that does not merely repeat the first reading. Iser’s account fits with this. Although his model is centred implicitly on a first reading of a text, he does also allow for rereading: ‘a second reading of the text will never have the same effect as the first, for the simple reason that the originally assembled meaning is bound to influence the second reading’.38 Note, though, that what is being mapped is not necessarily merely place and space. We should also bear in mind that many readers rarely or never reread texts. This means that not everyone will be aware of the hippocampus map or want to enrich it. The two most easily identifiable groups of rereaders would be children and academics (particularly literary critics).39 In relation to the former, for children under the age of twelve, visualisation and spatialisation are often reinforced by the presence of illustrations and maps 36 37 38 39

O’Keefe and Nadel, The Hippocampus as a Cognitive Map, 95. Redish, Beyond the Cognitive Map, 216. Iser, The Act of Reading, 149. Cf. Ryan et al. ‘Subsequent readings are more specialised since it is mostly “professional” readers with a specific purpose who have the luxury to reread a text’ (Narrating Space, 79).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

290

Reading As Mapping

alongside the text (as we have seen in earlier chapters). This confirms the sense that the presence of an explicit map strengthens and reinforces the inner map in a way that bears directly on the reading experience. At the same time, it reduces the value of both acts of mapping (external and internal) which are able to be dismissed as only necessary for the young. As Ellen Esrock sardonically puts it: The very fact that visualization aids children in reading reinforces the presumption that imagery is an elementary, less-developed form of cognition. For those who hold this view, imaging is like moving one’s lips while reading – a remnant of an earlier stage of cognitive development. (The Reader’s Eye, 118)

In relation to the latter group – professional literary critics – these must provide the best resource for any experiments in terms of producing external evidence of long-term maps for literature held in the hippocampus, since they are the most likely to use them repeatedly and to hold the greatest range of internal, subjective maps.

The Cognitive Mapping of Literature How exactly do the two forms of mapping described above (immersive route map; geometric locale map) interconnect in the spatial and cognitive experience of mapping the literary text? If we assume that there is a hippocampus map for each fictional text read, then we can also speculate that these will share core characteristics found in almost all texts: narrative structures; the chronological ordering of events; generic characteristics.40 These remain unchanging elements of the literary work that may be enriched by rereading but will not fundamentally alter. We could compare these core elements to those articulated by Kevin Lynch (paths; edges; districts; nodes) or those identified by Iser as determining different themes within each perspective (narrator, character, plot, reader-position). Something along these lines must form the base locale maps for literature. In terms of the route map that informs and updates the hippocampus map we can assume that movement through the text will be made in terms of waypoints, landmarks and nodal points, as it would if way-finding in the world. This kind of active route-mapping will occur at a spontaneous level, concurrent with the act of reading. At the same time, the use of key nodes 40

Both O’Keefe and Nadel, Hippocampus As a Cognitive Map, and Redish assert the presence of multiple maps in the hippocampus (Redish, Beyond the Cognitive Map, 130).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

The Cognitive Mapping of Literature

291

as part of the active reading process means that these are likely to be laid down on the long-term cognitive map. Anchor-point theory in relation to way-finding is relevant here, as an attempt to offer a unified model for the cognitive mapping process. Whilst anchor-points function in a similar way to landmarks, they are understood to be more subjectively determined: [M]any anchors (such as the location of home and work) would be too personal to have any significance for other, unrelated individuals . . . Landmarks are concrete, visual cues, whereas anchor-points may be more abstract elements.41

The hierarchy of knowledge produced in anchor-point theory is not simply dependent on the most significant feature in the landscape. Instead, a more personalised order of importance can be established: ‘a hierarchy of cognitive salience rather than spatial scale’ (103). Such a concept is directly relevant to the internal mapping of literary texts, which must have a subjective element. Work by Ferguson and Hegarty on the construction of cognitive maps from texts takes us one step closer to the cognitive mapping of literature through anchor-points (although their work draws on real-world description, not literary texts).42 They hypothesise that ‘mental models constructed from text are also organized around anchors’ (455) and that: If anchors in a text also serve as anchors in the cognitive map that a reader constructs from this text then they should be learned first and be recalled more often and more accurately than other landmarks. (456)

Their summary of findings strongly confirms the importance of anchorpoint theory for internal mapping of a text: [w]hen people read a text describing a spatial environment they represent some landmarks in the environment more completely and more accurately than others. These landmarks are anchors, that is, landmarks that are used in the text as reference points for describing the locations of other landmarks. In our experiments, anchors were represented more accurately than details regardless of whether the text was accompanied by a map, was written from a route or survey perspective, or the information was presented in a linear or an anchors-first order. This was also true regardless of whether subjects drew sketch maps from memory, from a list of landmarks in the town, or from the entire text. (468)

41

42

H. Couclelis, R. G. Golledge, N. Gale and W. Tobler, ‘Exploring the Anchor-Point Hypothesis’, Journal of Environmental Psychology 7.2 (1987): 99–122; 102. E. L. Ferguson and M. Hegarty, ‘Properties of Cognitive Maps Constructed from Texts’, Memory and Cognition 22 (1994): 455–73.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

292

Reading As Mapping

This is a highly pertinent finding: whatever form the external act of reading and mapping took, the core model for the individual map relied on anchors. Even assuming that this account is correct, however, when we apply anchor-point theory to the spatial mapping-out of literature it proves more problematic. The whole point of literary place and space – and what makes it so challenging– is that it can and needs to be mapped and re-mapped in a range of ways. Anchor-points are not necessarily concerned simply with the relative situating of (represented) spatial elements as would be true for the inner mapping of real-world place, nor can they be solely focused on narrative structures, as previous accounts have tended to suggest.43 The reader might be mapping in terms of linguistic complexity with anchors generated out of the sections of the text that are the most demanding, or require the most intense interpretation (this is more likely for ambiguity of language in a poem than a novel, for example, suggesting also the possibility of different generic inner map forms). The reader might be mapping conceptually, so that what stands out are the core ideas, or the relations between them (like the ancient memory-map for a rhetorical argument, but more schematic). Alternatively, readers may be mapping affectively; recalling most strongly the sections of a text that elicited from them the strongest emotion on a previous reading.44 This suggests the possibility that inner cognitive maps for fictional works must either be layered to allow for the multiple ways in which we can read and spatialise such a text (with each layer representing a different kind of meaning) or that they function in the way suggested by Angus Wilson (see below) when he describes his ‘permanent personal mental map of London’ working as ‘on those French Metro maps[:] by pressing a button your route is plotted out in electric lights’.45 In this horizontal version, the totalising map highlights multiple possible routes across it, depending on the reader’s need. The cognitive map for a literary work, then, can be understood to take a semi-schematic form. On the whole, information will be simplified and diagrammatic, relying upon anchor-nodes, but it will also bring into play some real-world elements drawn from associative memory. We might 43

44

45

Ryan et al. state: ‘Readers need mental maps to follow the plot, but they construe these maps on the basis of the plot’ (Narrating Space, 99). For an attempt to map affective response to literary texts, see David S. Miall, ‘Affect and Narrative: A Model of Response to Stories’, Poetics 17 (1988): 259–72 and Literary Reading: Empirical and Theoretical Studies (New York: Peter Lang, 2006). Angus Wilson, The Wild Garden: or, Speaking of Writing (Berkeley, Los Angeles and London: University of California Press, 1963), 126.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

The Cognitive Mapping of Literature

293

envisage some kind of core spatial structure loosely linked to the spatial form of the novel (i.e. itinerary form for a novel about a journey; circular for a village narrative etc.). This structure involves a series of relative positions that are of significance to the novel and/or its characters. While there are likely to be anchor-points created in one way in relation to key events or moments that occur within the text (‘intra-diegetically’ in narratalogical terms), there will also be ‘extra-diegetic’ anchor-points: denser, richer sites within the schema that correspond to subjective knowledge, referential real-life experience or to rich intertextual contexts in the associative web of memory. Note that the inner map also entirely ignores representational priority as real-world memories of places and literary representations are held in the same plane and merge together.46 Material elements of publication may also bear upon mapping as part of the reading experience. For example, the breaks made by chapters, or even the place on a page, may also be memorised and shape the experience of the fictional world indirectly, particularly where there is a significant break between reading enforced by publication (e.g. for the novel published as a serial) and reading across different media. Many literary works are about a journey of some kind, or involve following a route. In such cases, as Ferguson and Hegarty point out, ‘reading is naturally a sequential process, so it might be easier to read a route description, which is also sequential, than to read a survey description, which is not’.47 So, for example, a book like Huckleberry Finn follows the passage of the river as the character journeys along it; novels which adopt an itinerary form (e.g. the picaresque novel) do the same for the road; a quest is strongly spatial and goal-directed and so on. In such cases, self-reflexivity reinforces the inherently spatial meaning of the text because route and locale forms are closely aligned and of a similar order. We can ground this speculative account more concretely with reference to an authorial attempt to describe his own internal mapping process, by Angus Wilson in The Wild Garden.48 Wilson identifies himself as someone who has a strong spatial imagination: ‘I am a person who easily visualises life . . . in terms of maps, time charts and genealogies’ (118). He then gives a full account of internal mapping, considered here in terms of different

46 47 48

My thanks to Debbie Stubbs for pointing this out. Ferguson and Hegarty, ‘Properties of Cognitive Maps’, 456. I am indebted to Yi-Fu Tuan’s article ‘Images and Mental Maps’, Annals of the Association of American Geographers 65 (1975): 205–13 for directing me towards this example. Tuan quotes from it at some length (210–11) but does not analyse the account.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

294

Reading As Mapping

stages of enlargement involving the semi-schematic form.49 He begins by describing an initial act of inner visualisation: If I remember at some moment a particular object that I have seen, say the Blue Mosque in Istanbul, then the natural tendency of my mind . . . is to place this building in relation to other famous Istanbul mosques that I have seen. (119)

The immediate memory initiates a process of relative internal situating by association in which the particular example (the Blue Mosque) is placed in relation to related types found in the memory (taxon processing). He continues: These in turn I see visually on what I remember from the whole map of Istanbul in my Blue Guide. If I am tired and idle the picture will begin automatically to expand. Istanbul will appear on a map of Turkey beside the other Turkish towns I have visited. (119)

At this point, Wilson transposes his own visual images (internally) onto an ‘accurate’ referential map form that has previously been remembered and merges with his inner mapping (which can only be relative and diagrammatic). Cartographic knowledge of a map in the real world provides an internal base structure. As he goes on, Wilson also identifies an internal hierarchical ordering of images on the basis of intensity of visual definition: ‘This map will also be marked with the towns I failed to see in feebler pictures of details that I have only read of’ (119). Direct experience is stronger than indirect in the memory. The map continues to enlarge and is now filled in to a greater or lesser degree depending on Wilson’s own level of personal knowledge of place for which black represents ‘real experience’, grey ‘imagination’ and ‘in between, varying shades to mark literary associations, historic events . . . and so on’ (119–20). He continues: Thus on my mental map the London area is a black splodge, Provence richly black, Antarctica . . . a heavy grey, Tehran lightly marked by my view of the airport . . . overshaded because it is the residence of an old friend, cross-shaded by the word Mussadiq and his pyjama-ed form. (120)

Finally, Wilson also factors time into his map, both his own subjective temporal experiences and those represented in the text: Above this world map with its overlays or shadings and collections of dramatis personae, time spirals upwards so that each place too has its historical chart either dating personal experiences or bringing into mind its historic past. (120)

49

The example given here is of a re-mapping of real-world space, rather than fictional – although Wilson then goes on to apply the model to his own novelistic spaces.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

The Cognitive Mapping of Literature

295

The flat map becomes almost three-dimensional (3D) and dynamic in his account, with time as a vertical model emanating from the horizontal. Wilson’s account (always assuming we take it at face value) confirms a semi-schematic model that is layered and partial, varying in density according to associative power and personal subjective experience of a particular place.50 I want to conclude this exploration by asking: What does the cognitive mapping of literature mean for the main subject of this book: the presentation of an empirical map alongside the text? Ferguson and Hegarty address the presence or absence of such a map directly: A growing body of research suggests that constructing a mental model is facilitated when the text is accompanied by a diagram . . . a map displays all the spatial relations between landmarks in an environment. In contrast, a text describes only some of these relations explicitly, so that the others must be inferred. (‘Properties of Cognitive Maps’, 456)

Although their account draws upon both verbal and visual representations of real-world topography, the same principle applies for a fictional map accompanied by a text. They continue: ‘If subjects read a text accompanied by a map, they can inspect the map to encode the spatial relations between objects directly, eliminating the need to make spatial inferences’ (456). We must assume that where a fictional map also exists for the text, this will act to shape, strengthen and fix the reader’s internal map by adding another layer of (authoritative) spatial understanding that gives that world far greater spatial definition (in a way similar to that of a real-world map that has been memorised). The first reading of a text cannot provide a totalising overview since information is only able to be sequentially accumulated. However, a map given alongside the text can provide this. It enables readers to situate the active practice of route-mapping within another structure. From this perspective then, the explicit map functions both as a proxy for real-world reference (fixing the space in a more universal way for all readers) and as a kind of anticipatory proxy for the individual hippocampus map that cannot yet have been formed. The danger of this is that the published external map (not generated out of the reader’s own spatial experience of the text) will then dominate the personal inner map. In this sense, fears about the fictional map dominating readerly imaginative 50

Wilson’s description of fading colours can be compared directly with Ryan et al.’s account of first mapping as partial: ‘This first pass over a text typically leaves readers with a vivid, though spotty visualization of the setting: something like those reconstructed Cretan frescoes made of pictorial fragments separated by blank areas’ (79).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

296

Reading As Mapping

engagement are justified. However, we need also to remember that this will be happening in a wide range of ways relating to all kinds of cultural, visual, geographic and cartographic forms of representation. If a reader is reading a text set in a well-known real-world landscape that he or she has visited, then personal experience as well as more formal cartographic layers (e.g. Ordnance Survey maps) would come into play and affect the schema. If the reader has seen a film of the book, this will bear upon it. Any visual-spatial experience relatable to the spatial world of the book is likely to inform the base model.

Mapping Literature in Digital Space The final part of this chapter attempts to offer future directions for the field of digital literary cartography, emerging out of the previous sections. To what extent might greater understanding of how we map literature internally bear upon how we map literature externally and virtually in the digital domain? The digital mapping of literary place and space only has a brief history but, even so, I think we can begin to identify a ‘first generational’ model based on one set of tools and principles that is ready to be, if not superseded, then significantly diversified and hybridised. For the first twenty years of the twenty-first century, GIS (Geographic Information Systems) has been in the ascendancy, combining vector data (using points, lines and polygons to determine coordinates) with raster data (individual grid cells with specific values recorded as an image) to create accessible visual and spatial ways of understanding information that have transformed human understanding of space and place.51 For the larger public, this has manifested itself in entirely new ways of perceiving and understanding place by means of Google Earth, Google World, Google Street View and so on.52 Increasingly, these forms are then democratised 51

52

GIS is a catch-all term that embraces a wide range of geo-technologies that underpin almost all digital systems to some degree (since by its nature a data-retrieval system requires spatial organisation). For the most part, though, it is understood to refer to ‘corporate “Big GIS” such as ESRI ArcGIS’ (Jeremy W. Crampton, Mapping: A Critical Introduction to Cartography and GIS (Oxford: Wiley-Blackwell, 2010), 102). For a full consideration and critique see Crampton, Mapping, particularly his account of the ‘GIS Wars’ in chapter 8. See David Cooper, Chris Donaldson and Patricia Murrieta-Flores, ‘Introduction’, in Literary Mapping in the Digital Age, ed. David Cooper, Chris Donaldson and Patricia Murrieta-Flores (London: Routledge, 2017), 4. This comprehensive collection of essays encompasses a range of current and future approaches including: networks; applied GIS; big data mapping; deep mapping.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Mapping Literature in Digital Space

297

through crowdsourcing such as the OpenStreetMap project (OSM) so that, as Sebastién Caquard points out, ‘Digitising, which was once considered as the most tedious part of any GIS project, has been turned by Google and Microsoft – as well as by OSM – into an enjoyable activity.’53 For academia, and for the spatial humanities in particular, it has meant the application of GIS tools (e.g. ArcGIS) that enable vector point mapping to specific coordinates, by the creation of underlying annotation tables for elements drawn from a textual database (corpus). In this way, as Caquard summarises for us: [P]laces appearing in narratives, such as novels . . . can now be located and mapped automatically . . . The increasing popularity of this process called geoparsing is directly linked to the development of digital gazetteers . . . in which place names are associated with geographic co-ordinates. (‘Cartography I’, 138)

Any text containing place names can easily be mapped onto real points corresponding to the realworld, using latitude and longitude. A number of high-profile literary projects have made good use of these technologies and sought to adapt them for the kind of knowledge and interpretation required in a subject such as Literary Studies. This is not the place to engage with them in great detail, but the following projects exemplify different approaches: the mapping of fictional onto real-world place and space (Atlas of Europe project); the mapping of social networks to reveal hidden connections (Six Degrees of Francis Bacon or Stanford’s Mapping the Republic of Letters project); the attempt to create a more qualitative model from within standard GIS technology (by means of a scale on which a particular level of emotion can be mapped) such as Mapping the Lakes and more recently Mapping Emotions in Victorian London.54 More generally, the use of GIS for the purposes of ‘text mining’ has encouraged mapping across large corpora and the ‘distant reading’ of literature in quantifiable ways – identifying patterns and meanings not previously easily seen, to be sure, but not meeting the more traditional needs of the discipline.55

53

54

55

Sébastien Caquard, ‘Cartography I: Mapping Narrative Cartography’, Progress in Human Geography 37 (2011): 135–44; 137. See www.atlas-of-literature.eu; www.sixdegreesoffrancisbacon.com; http://republicofletters .stanford.edu; www.lancaster.ac.uk/mappingthelakes/; David Cooper and Ian Gregory, ‘Mapping the English Lake District: A Literary GIS’, Transactions of the Institute of British Geographers 36 (2011): 89–108; R. Heuser, M. Algee-Hewitt, E. Steiner and V. Tran, ‘Mapping the Emotions of London, 1700–1900: A Crowdsourcing Experiment’, in Literary Mapping in the Digital Age, ed. Cooper, Donaldson and Murrieta-Flores (New York: Routledge, 2017), 25–46. The term ‘distant reading’ is defined by Franco Moretti in Graphs, Maps, Trees: ‘a process of deliberate reduction and abstraction. “Distant reading” I have once called this type of approach; where distance is however not an obstacle, but a specific form of knowledge’ (8).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

298

Reading As Mapping

What do all of the spatial projects for literature listed above have in common? They only map (or map onto) real-world places, people and specific geographic locations. The Atlas of Europe project, for example, takes as its central tenet the desire to answer the question ‘Where is fiction set?’ and is entirely premised upon an anchoring of fictional space by overlay in relation to real-world place and cartography.56 A recent paper on ‘The Spatial Analysis of Vague and Imaginary Place and Space’ that attempts to address the larger issue of mapping beyond real-world coordinates notes that: [A]lthough there has been some advance in addressing spatial uncertainty of real geographies in modern contexts, little has been done to fully study and address the interplay between real, vague, and imaginary places in historical literature.57

The paper allows for varying degrees of correlation to the real-world (37–8) and thus for the visualisation of more or less specific spaces and of spatial analyses across texts. Still, it relies on the use of GIS and complex annotation of a corpus of medieval romances, and what can be mapped in this way is restrictive in terms of the full range of spatial meaning held in literary forms. Two core problems for literary studies follow directly from over-reliance on referential GIS mapping of texts based upon specific locations in the world. First, counter to Piatti’s claim that ‘One of literary cartography’s traditional starting points is the assumption that a large part of fiction refers to the mappable world,’ I would argue that only quite a small proportion of literary works and genres have direct correspondence to real-world places (e.g. autobiography; the realist novel (arguably); travel writing) which means that the model is not applicable to many literary genres (sci-fi; fantasy; adventure; many forms of poetry etc.).58 Second, even where a literary text appears to correspond to the real world, this is a false correlation (as established in the previous chapter) and leads to the kind of referential confusion evidenced by literary-touristic desires for ‘accuracy’. It is not the case that a referential real-world model needs to be thrown out altogether. For some forms of literature, it remains highly appropriate (e.g. for a highly spatial writer for whom particular sites in the world are significant in relation to his or her writing, such as Balzac; 56

57

58

See B. Piatti, H. R. Bär, A. K. Reuschel, L. Hurni and W. Cartwright, ‘Mapping Literature: Towards a Geography of Fiction’, in Cartography and Art, ed. W. Cartwright, G. Gartner and A. Lehn (Berlin and Heidelberg: Springer, 2009), 1–16. This paper lays out the core principles of the project. Patricia Murietta-Flores and Naomi Howell, ‘Towards the Spatial Analysis of Vague and Imaginary Place and Space: Evolving the Spatial Humanities through Medieval Romance’, Journal of Map and Geography Libraries 13.1 (2017), 29–57. Barbara Piatti, ‘Mapping Fiction: The Theories, Tools and Potentials of Literary Cartography’, in Literary Mapping in the Digital Age, ed. Cooper, Donaldson and Murrieta-Flores, 88–101; 89.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Mapping Literature in Digital Space

299

Dickens; Wordsworth). The main problem is that it is determining almost the entire approach for literary cartography (and thus restricting the mapping of literature to authorial, cultural and historical approaches) whereas it should only apply to certain kinds of text, or acts of mapping. We can surely imagine a mixed model that combines GIS mapping with other forms of mapping to permit greater complexity. One emerging counter-model, in the form of ‘Deep Mapping’, potentially allows for a richer, more multiple map-form that enables the accretion of temporality and lived existence in place to be registered more deeply (although it still tends to be predicated on a real-world site).59 Such a model begins to challenge the assumptions of GIS for the Humanities, the difficulties of which David Bodenhamer et al. make clear: Chief among the issues are a mismatch between the positivist epistemology of GIS, with its demand for precise, measurable data, and the reflexive and recursive approaches favoured by humanists and some social scientists . . . who wrestle continually with ambiguous, uncertain, and imprecise evidence and who seek multivalent answers to their questions. The problem, it seems, is both foundational and technological: we do not yet have a well-articulated theory for the spatial humanities, nor do we have the tools sufficient to meet the needs of humanists. (171)

Such a statement suggests that we are at a point of shift – moving from early attempts to adapt the scientific model to the spatial humanities, towards more bespoke methods and tools that can answer the particular needs and questions of these disciplines. In a similar way, but from the perspective of 3D mapping, Gary Priestnall advocates ‘alternative forms of geo-referencing beyond the “pin in the map” approach’.60 In concluding remarks to a special issue of the Cartographic Journal dedicated to the ‘Mapping of Fictional Worlds’, Caquard calls for ‘the emergence of more exploratory and hybridized practices and concepts’ and concludes that: If we agree . . . that places are made of geographical layers and layers of stories, then providing solutions for merging these layers is an important challenge for cartographers engaged in this mapping process.61

59

60

61

See David J. Bodenhamer, John Corrigan and Trevor M. Harris, eds., Special Issue: ‘Deep Mapping and the Spatial Humanities’, International Journal of Humanities and Arts Computing 7 (2012): 170–5; David J. Bodenhamer, John Corrigan and Trevor M. Harris, eds., Deep Maps and Spatial Narratives (Bloomington, IN: Indiana University Press, 2015). Gary Priestnall, ‘Spatial Frames for Literature through Geospatial Technologies’, in Literary Mapping in the Digital Age, ed. Cooper, Donaldson and Murrieta-Flores, 240–55; 240. Sébastien Caquard, ‘Cartographies of Fictional Worlds. Concluding Remarks’, The Cartographic Journal, 48 (2013): 224–5; 225.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

300

Reading As Mapping

Such a position chimes directly with my own, here. It seems to me, that an entirely different conception of what we are doing when we map literature needs to inform the underlying digital model (and thus the kinds of tools created and the uses to which they can be put). Put simply, we need to create a non-referential model for the literary mapping of fictional works. Once we begin to think non-referentially (in terms of relative rather than absolute space), this releases us into an entire reconsideration of the model for digital literary mapping. What should we be mapping – what form it should take? How and Why? I want to address each of these questions here – doing so within the frame of the map/route distinction that (I have argued) underlies acts of individual readerly mapping. If we accept that there is no true referential ground for the mapping of literary place and space in the world, then a model that maps according to cognitive acts of mapping is at least as viable as the current one. The first question then: What should be on the base map and what form might it have? Here, the way is partially illuminated for us by the work of Ryan. It is clear that the complexity of literature requires some kind of multiple model that works in terms of layers of map that can be placed on top of one another and show through, or be selected according to the needs of a particular reading. In a sense, the number of these layers could be infinite (if one were to map for the needs of every possible theme required by every possible reader). In Narrating Space/Spatializing Narrative, Ryan and her fellow authors provide a blueprint: The ultimate goal of narrative cartography should be the design of dynamic models of plot that answer the types of questions formulated by Piatti. This project would require an interactive database comprising multiple interconnected diagrams, such as a geographic map of the storyworld, a time line showing the succession of events and diagrams representing the private worlds of characters, which includes their thoughts, emotions, feelings toward each other, goals and plans after every major event. (Narrative Space, 212)

As a narrative theorist, Ryan is interested in a ‘narrative cartography’ rather than a broader literary cartography, but a narratological model probably still needs to be at the heart of the latter. In an earlier paper, ‘Diagramming Narrative’, Ryan provides a full and thorough account of what is involved here.62 At the heart of her account is

62

Marie-Laure Ryan, ‘Diagramming Narrative’, Semiotica 165 (2007): 11–40.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Mapping Literature in Digital Space

301

the identification of four levels (given in reverse order here since 1 is the deepest layer): 4. 3. 2. 1.

‘The discourse level’ The ‘particular narrative structure’ Generic form or ‘the universal narrative structure’ The ‘universal deep structure’ (underlying schema of logical relations) (‘Diagramming Narrative’, 16)

The majority of her paper then focuses on the complexity of attempting to relatively map or ‘diagrammatise’ the individual text of level 3. Ryan asserts: Of these four levels of analysis, the third one is the most challenging for the narrative cartographer because of the complexity of the cognitive pattern that we call a story, but it is also the most stimulating, because it offers the greatest freedom to the visual imagination.(18)

Within this third level, Ryan explores three core dimensions of narrative that can be visualised: ‘spatial dimension’, ‘temporal and evenemential [event-centred] dimension’ and what she calls the ‘mental dimension’ (19) of characters. Of these dimensions it is the last – relating to the dynamics between characters externally and internally, the linkage in terms of negative and positive motivations, emotions and agency, that is the most problematic in any attempt at visual or spatial representation, not least because of the multiplicity of diagrams required. The ‘Parallel Map Theory’ for cognitive mapping by Jacobs and Schenk, (considered above) described a series of internal ‘sketch maps’ functioning in a connected way within the mind: ‘sketch maps are disconnected from absolute locations in space, unless they are linked to all other sketch maps through the bearing map’ (302). We could immediately adapt such a model to the mapping of literature if we envisage a series of mini-maps for each section of the narrative (or in a more formal structure, each chapter of a novel) that are then nested within a master-map of the whole. Thus, this (essentially narratological) map level might work as a central synchronic layer containing multiple nested layers for an individual text. One might then have a whole series of synchronic layers above and below it that create a similar series of nested maps for other meaningful ways of reading the text. I would take Ryan’s model, and offer an adapted account of spatialised map visualisation layers for any given text: Level 1: Communicative layer Literature as a communicative/dialogic act (forms of publication; paratext; narrator/narratee)

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

302

Reading As Mapping

Level 2: Relative toponymic layer Network of places created by use of place names within the text; actual or apparent referentiality of literary text to existing spatial representations (real-world map; fictional map; photographed or illustrated place) Level 3: Narratological layer 3a. Fabula (chronology of events) 3b. Syuzhet (narrative order of story) 3c. Time and event mapping (different forms of time; subjective time) Level 4: Predetermining form Generic; spatio-temporal or other form predetermining spatial relations Level 3 remains the core layer for the base map as in Ryan. Level 2 – currently the normative level for GIS-based mapping projects – now becomes both optional and subsidiary since it is generated out of one element within the text (place names) which may or may not be fictional, and may be mapped onto the real world or onto an internally consistent map representation of the fictional world. Let us turn to the second question: How might we attempt to map literary texts, especially those representing fictional realms? My own current AHRC-funded project (Chronotopic Cartographies 2017–2020) is undertaking exactly the kind of work outlined by Ryan and can thus begin to answer this question (while recognising that the ‘How’ will be constantly changing in relation to rapidly advancing digital tools).63 In order to try and retain some kind of integrity for the core data (concerning space and time in literary representation), 2D base maps are generated out of the text itself, using a schema that draws upon Mikhail Bakhtin’s model of the chronotope in ‘Forms of the Chronotope and the Novel’ and Gérard Genette’s account of time in Narrative Discourse.64 Space and time are coded separately and the base layers that can then be generated out of the code are those given above. In the context of this chapter, all of these layers could be understood to correspond to the hippocampus locale map for a literary text that creates a universal shared model at the core.65 At the same 63

64

65

See www.lancaster.ac.uk/chronotopic-cartographies. This project is funded by a major grant from the AHRC and none of the work undertaken would have been possible without their financial assistance. Gérard Genette, Narrative Discourse: An Essay in Method, trans. Jane E. Lewin (Ithaca, NY: Cornell University Press, 1979). For a detailed account of the method, see Sally Bushell, James Butler, Duncan Hay and Rebecca Hutcheon, ‘A Non-referential Model for Literary Mapping in the Digital Domain’

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Mapping Literature in Digital Space

303

time, and crucially, assuming a combined two-part model for mapping literary space and place, a more dynamic, individualised route-mapping across a text is enabled by means of 3D visualisations of various kinds. Individual modes of mapping are relevant here. In The Cognition of Geographic Space, Rob Kitchin considers the advantages and disadvantages of sketch maps as a form of externalised inner mapping for geography.66 He identified eight different mapping strategies by participants ‘including both image-based and non-image-based ones’ (75). These included various combinations of reliance upon prior knowledge of an Ordnance Survey map; active route-mapping ‘by constructing a map and then converting route knowledge into configural knowledge’ (76); a more visual model that involved looking at one point from another; an amalgam to create an aerial perspective; and purely verbal mapping, through a sequence of directions. The range of approaches to the task suggests that whilst there is a fairly consistent universal structure for internal mapping (that involves drawing upon and later combining route and locale map forms), that model can be put into practice by individuals in a wide range of ways – some of which are far more visual and immersive (vista-based), some of which might be verbal (or combine visual images and verbal symbols). In Narrating Space/Spatializing Narrative, Ryan, Foote and Azaryahu present the results of an experiment with sixty high school students who had read Gabriel Garcia Marquez’s novel Chronicle of a Death Foretold which ‘recommends itself for the study of mental maps through its meticulous attention to spatial configuration’.67 In terms of map style, five different forms emerged, ranging from the most objective (pure plan) through to highly subjective (pure picture) (95). This confirms a spectrum of internal map styles for individual mapping of literature as well as in the world. The 3D environment also introduces something entirely new to visualisation: the immersive environment, in relation to which we might consider that ‘not only do these technologies generate new dynamic spaces, they

66

67

(forthcoming, 2020). My deep thanks to James, Becca and Duncan for their conceptual and intellectual contribution to the project and to my summary of it here. Kitchin and Blades, The Cognition of Geographic Space. Kitchin draws upon his own earlier study: ‘Exploring Spatial Thought’, Environment and Behaviour 29 (1997): 123–56. The origins of interpreting such externalised maps can be traced back to Lynch’s Image of the City which was based on the amassing of data from sketch maps by inhabitants; and Peter Gould and Rodney White’s book Mental Maps (Harmondsworth: Penguin, 1974) which reproduced maps of black children for the Mission Hill area of Boston that revealed internalised demarcation of areas based on colour (14–15). Ryan et al., Narrating Space, 77.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

304

Reading As Mapping

[even] demand the development of new mapping strategies’ (39).68 The immersive environment for literature again leads in the direction of the theory of the mind and cognitive narratology.69 Route maps across or within such virtual worlds allow for ‘simulation’ or mimesis (following a character or re-enacting movement and event in a first-person role within a space, e.g. across the Heath of Hardy’s The Return of the Native; as Jim Hawkins on Treasure Island) but also permit creative engagement: the reader is free to wander, explore, adapt and transform the virtual environment. We must also bear in mind that any attempt to map and visualise literature – particularly for fictional works – needs to respect the nature of those texts and of the kind of spatiality they contain and project. The mapping of literature is not total, absolute or systematic – either at a level of representation or of internal mapping. It needs to allow for aporias and blindspots. It needs to find a way to represent what is not mapped, what resists mapping as well as what is mappable. In sum, the act of mapping literature has to be capable of challenging and questioning the act of mapping literature. Again, a two-part model for literary mapping can at least partly allow for this paradox. Whilst the automated base (locale) maps are necessarily ‘accurate’ to the spatial detail of the text (being generated directly out of it), other forms of visualisation are able to challenge or work against this. For example, in representing Oliver Twist’s London, where base maps create a relative positioning of key characters and events that spatialise the text as a whole, the 3D visualisation might represent only those real-world areas described in the text, leaving blank spaces for some streets and allowing only partial visualisation for regions not visited.70 Showing what is not mapped alongside what is mapped allows a new form of understanding that also respects the nature of literary representation and makes clear its distance, and necessary difference, from real-world space and place. Why do any of this? After all, could it not be argued that the desire to map out and visualise literature is wildly misplaced if the form is so resistant to it? Here we need to return to the rationale for this entire study in relation to its discipline. If literary studies is to remain relevant and

68

69 70

Christina Ljungberg, ‘Mapping Practices for Different Geographies’, in Mapping Different Geographies, Lecture Notes in Geoinformation and Cartography, ed. K. Kriz et al. (Berlin and Heidelberg: Springer-Verlag, 2010), 38–55. This is an area of forthcoming research but there is not space to enlarge upon it fully here. My thanks to Duncan Hay for making me aware of this point.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

‘Let Us Pretend That It Is the End’

305

vibrant as a discipline and to connect effectively in new ways with Geography, Cartography and GIS, then it has to be open to new ways and forms of understanding itself. If space and place is no longer background, if spatiality is central to who we are, to how we function, even to how we understand and acquire knowledge, then it is surely worth paying close attention to it and using the most recent tools to find new ways of engaging with it. The digital environment does offer us this in relation to place and space and it is the responsibility of scholars to engage with it and make it work for us and for the larger public in accessible ways.

‘Let Us Pretend That It Is the End’71 I want to conclude not only this chapter, but the book as a whole, by returning to a work of fiction – Peter Pan and Wendy – and to the ability of the literary work to play with all forms of the map in rich and complex ways. In a direct address to the reader, the narrator of Peter Pan describes the ‘map of a child’s mind’: I don’t know whether you have ever seen a map of a person’s mind. Doctors sometimes draw maps of other parts of you, and your own map can become intensely interesting, but catch them trying to draw a map of a child’s mind, which is not only confused, but keeps going round all the time. There are zigzag maps on it . . . and these are probably roads in the island; for the Neverland is always more or less an island, with astonishing splashes of colour here and there. (73)

The narrator takes the unrealisable nature of the cognitive map and manipulates it freely to ‘draw’ his own imaginative version that resists and defies description because of its constantly dynamic nature. He acknowledges the complexity created by the necessary connection of the subjective to other real-world or read experiences in an image of layered meaning: ‘either these are part of the island or they are another map showing through’ (74). As the narrator moves away from the individual and towards the collective map filled with imaginative generic elements of children’s play, we come to realise that ‘Neverland’ is both an analogue, or figure, for childhood creativity and (somehow, by the magic of literary representation) an externalisation of the inner cognitive map that now becomes an actual place. 71

‘Slightly tried to tell a story that night, but the beginning was so fearfully dull that it appalled even himself and he said gloomily, “Yes, it is a dull beginning. I say, let us pretend that it is the end”’ (Peter Pan and Wendy, 163).

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

306

Reading As Mapping

Within the self-consistent terms of the fictional world, Neverland is able to be fully externalised because of the aberrant existence of Peter, suspended forever in a state of make-believe (as we saw in Chapter 5) and generating it from his own inner space. For this reason it only fully activates when he is in it: ‘Feeling that Peter was on his way back, the Neverland had again woken into life . . . In his absence things are usually quiet on the island’ (112). At the same time, Neverland is a layered, merged totality made up of each of the children’s own individual inner versions: ‘John’s, for instance, had a lagoon with flamingoes flying over it . . . while Michael, who was very small, had a flamingo with lagoons flying over it’ (74). The fully externalised map is a ‘nicely crammed’ (74) composite so that when the children see the ‘actual’ island of Neverland – from above in literally aerial view – it is immediately familiar to each of them: Strange to say, they all recognized it at once, and until fear fell upon them they hailed it, not as something long dreamt of and seen at last, but as a familiar friend . . . ‘I say, John, I see your flamingo with the broken leg.’ ‘Look, Michael, there’s your cave.’ (105)

The reader’s cognitive map would have to be: a map of Peter Pan; that contains a map of Neverland; that is generated from Peter Pan’s existence, in and through the mental map of three particular children; who represent the creative-collective child mind; of which the reader may once have been a part. Barrie’s depiction of the map within the text is so anticipatory of the process of inner visualisation that it brings into being the object it describes: relative; multiple; layered; associational; subjectively anchored; dynamic; changeable; individual and shared. A non-existent map, of a nonexistent place.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.009

Bibliography

Adams, Richard. Watership Down. London: Rex Collings Ltd, 1972. Akerman, James R., ed. The Imperial Map: Cartography and the Mastery of Empire. Chicago, IL; London: University of Chicago Press, 2009. Allingham, Marjorie. More Work for the Undertaker. London: William Heinemann, 1948. Allingham, Phillip V. ‘Arthur Hopkins’s Illustrations for Thomas Hardy’s The Return of the Native’. Victorian Web, November 2004: www.victorianweb .org/authors/hardy/native/illustrations.html. Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso, 1983. Aseguinolaza, Fernando Cabo. ‘The Spatial Turn in Literary Historiography’. Comparative Literature and Culture 13.5 (2011): http://docs.lib.purdue.edu/ clcweb/vol13/iss5/5. Baker, Deirdre F. ‘What We Found on our Journey through Fantasy Land’. Children’s Literature and Education 32 (2006): 237–51. Bakhtin, M. M. ‘Forms of Time and of the Chronotope in the Novel’. In The Dialogic Imagination: Four Essays by M. M. Bakhtin. Ed. Michael Holquist, Trans. Caryl Emerson and Michael Holquist. Austin: University of Texas Press, 1981. 84–258. Barnard, R. A Talent to Deceive: An Appreciation of Agatha Christie. London: William Collins Sons & Co. Ltd, 1980. Barrell, John. ‘Geographies of Hardy’s Wessex’. Journal of Historical Geography 8 (1982): 347–61. The Idea of Landscape and the Sense of Place, 1730–1840. Cambridge: Cambridge University Press, 1972. Barrie, James M. The Little White Bird. London: Hodder & Stoughton, 1902. Peter Pan in Kensington Gardens. London: Hodder & Stoughton, 1906. Peter and Wendy. London: Hodder & Stoughton, 1911. Peter Pan, or the Boy Who Would Not Grow Up. London: Hodder & Stoughton, 1928. Barrow, Ian J. Making History, Drawing Territory: British Mapping in India 1756–1905. Oxford: Oxford University Press, 2003. Baucom, Ian. Out of Place: Englishness, Empire, and Locations of Identity. Princeton, NJ: Princeton University Press, 1999.

307

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

308

Bibliography

Baudrillard, Jean. Simulacra and Simulation [1981]. Trans. Sheila Farer Glaser. Ann Arbor: University of Michigan Press, 1994. Bavidge, Jenny. ‘Stories in Space: The Geographies of Children’s Literature’. Children’s Geographies 4.3 (2006): 319–30. Belyea, B. ‘Images of Power: Derrida/Foucault/Harley’. Cartographica 29 (1992): 1–9. Benjamin, Walter. ‘The Work of Art in the Age of Mechanical Reproduction’. In Illuminations. Ed. Hannah Arendt. Trans. Harry Zohn. London: Collins/ Fontana, 1973. 211–35. Billman, Carol. ‘Reading and Mapping: Directions in Children’s Fantasy’. Children’s Literature Association Quarterly 7 (1982): 40–6. Bishop, M. ‘Ambrosius Holbein’s Memento Mori for Sir Thomas More’s Utopia. The Meanings of a Masterpiece of Early Sixteenth Century Graphic Art’. British Dental Journal 199 (2005): 107–12. Blake, William. Blake’s Poetry and Designs. Ed. Mary Lynn Johnson and John E. Grant. New York; London: W.W. Norton & Co. Ltd, 1979. Blewett, David. The Illustration of Robinson Crusoe: 1719–1920. London: Colin Smythe Ltd, 1995. Bloom, Clive. Spy Thrillers from Buchan to Le Carré. Basingstoke: Palgrave Macmillan, 1990. Bodenhamer, David J., John Corrigan and Trevor M. Harris, eds. Special Issue: ‘Deep Mapping and the Spatial Humanities’. International Journal of Humanities and Arts Computing 7 (2012). eds. Deep Maps and Spatial Narratives. Bloomington: Indiana University Press, 2015. Bogstad, Janice M. and Philip Kaveney, eds. Picturing Tolkien: Essays on Peter Jackson’s The Lord of the Rings Film Trilogy. Jefferson, NC: McFarland and Co., 2011. Boyle, Andrew. The Riddle of Erskine Childers. London: Hutchinson, 1977. Bracher, Frederick. ‘The Maps in Gulliver’s Travels’. Huntington Library Quarterly 6.1 (1944): 59–74. Bray, Patrick. The Novel Map: Space and Subjectivity in Nineteenth Century French Fiction. Chicago, IL: Northwestern University Press, 2013. Brogan, Hugh. The Life of Arthur Ransome. London: Jonathan Cape, 1984. Brosseau, Marc. ‘Geography’s Literature’. Progress in Human Geography 18.3 (1994): 333–53. Brotton, Jerry. A History of the World in Twelve Maps. London: Penguin, 2013. Brown, Howard. ‘Tips, Traps and Tropes: Catching Thieves in Post-Revolutionary Paris’. In Police Detectives in History, 1750–1950. Ed. Clive Emsley and Haia Shpayer Makov. Aldershot: Ashgate, 2006. 33–60. Buchan, John. The Island of Sheep. London: Hodder & Stoughton, 1936. Prester John. London: Thomas Nelson & Sons Ltd, 1910.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

309

Bulson, Eric. Novels, Maps, Modernity: The Spatial Imagination, 1850–2000. New York; London: Routledge, 2006. Bunyan, John. John Bunyan’s The Pilgrim’s Progress. Ed. Cynthia Wall. W.W. Norton & Company, 2009. The Miscellaneous Works of John Bunyan. Vol XII. Ed. W. R. Owens. Oxford: Clarendon Press, 1994. The Pilgrim's Progress ... Complete in Two Parts. A New Edition. Embellished with a More Superb and Elegant Set of Copperplates than Was Ever Given With Any Former Edition. London: H. Trapp, 1778. Bushell, Sally. ‘Mapping Fiction: Spatialising the Literary Work’. In Literary Mapping in the Digital Age. Ed. David Cooper, Christopher Donaldson and Patricia Murrieta-Flores. London: Routledge, 2016. 125–46. ‘Paratext or Imagetext? Interpreting the Literary Map’. Word and Image 32 (2016): 181–94. ‘The Slipperiness of Literary Maps: Literary Cartography and Critical Cartography’. Cartographica 47.3 (2012): 149–60. Bushell, Sally, James Butler, Duncan Hay and Rebecca Hutcheon. ‘A Nonreferential Model for Literary Mapping in the Digital Domain’. Forthcoming, 2020. Campbell, Tony. ‘Portolan Charts from the Late Thirteenth Century to 1500’. In History of Cartography, Vol. 1. Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean. Ed. J. B. Harley and David Woodward. Chicago, IL: University of Chicago Press, 1987. 371–463. Caquard, Sebastién. ‘Cartographies of Fictional Worlds: Conclusive Remarks’. The Cartographic Journal 48.4 (2013): 224–5. ‘Cartography I: Mapping Narrative Cartography’. Progress in Human Geography 37 (2011): 135–44. ‘Cartography III: A Post-representational Perspective on Cognitive Cartography’. Progress in Human Geography 39 (2015): 225–35. ‘Narrative Cartography: From Mapping Stories to the Narrative of Maps and Mapping’. The Cartographic Journal 51 (2014): 101–6. Carracciolo, Marco. ‘Cognitive Literary Studies and the Status of Interpretation: An Attempt at Conceptual Mapping’. New Literary History 47 (2016): 187–207. Carroll, Lewis. The Complete Illustrated Works of Lewis Carroll. London: Chancellor Press, 1982. Carruthers, Mary. The Book of Memory: A Study of Memory in Medieval Culture. Cambridge: Cambridge University Press, 1990. The Craft of Thought: Mediation, Rhetoric and the Making of Images 400–1200. Cambridge: Cambridge University Press, 1998. Case, Arthur E. Four Essays on Gulliver’s Travels. Gloucester, MA: Peter Smith, 1958.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

310

Bibliography

Certeau, Michel de. The Practice of Everyday Life. Trans. Stephen Rendall. Berkeley; London: University of California Press, 1984. Chandler, Daniel. Semiotics: The Basics. London: Routledge, 2002. Childers, Erskine. The Riddle of the Sands: A Record of Secret Service Recently Achieved [1903]. Harmondsworth: Penguin, 1978. Christie, Agatha. An Autobiography. London: W. Collins Sons & Co. Ltd, 1977. The Murder at the Vicarage. London: W. Collins and Sons Co. Ltd, 1930. The Murder of Roger Ackroyd. London: W. Collins Sons & Co. Ltd, 1926. Murder on the Orient Express. London: W. Collins Sons & Co. Ltd, 1934. The Mysterious Affair at Styles. London: W. Collins Sons & Co. Ltd, 1921. They Do It With Mirrors. London: William Collins, 1952. [Clark, Samuel]. Reuben Ramble’s Travels through the County of England. London: Darton & Clark, 1845–50. Coleridge, Samuel Taylor. The Collected Works of Samuel Taylor Coleridge Vol. VII. Biographia Literaria, or Biographical Sketches of My Life and Literary Opinions. Ed. James Engell and W. Jackson Bate. Princeton, NJ: Princeton University Press, 1983. Collé Bak, Nathalie. ‘The Role of Illustrations in the Reception of Pilgrim’s Progress. In Reception, Appropriation, Recollection: Bunyan’s Pilgrim’s Progress. Ed. W. R. Owens and Stuart Sim. Frankfurt/Main: Peter Lang, 2007. 81–98. Conley, Tim. The Self-Made Map. Cartographic Writing in Early Modern France. Minneapolis; London: University of Minnesota Press, 1996. Connolly, Daniel. ‘Imagined Pilgrimage in the Itinerary Maps of Matthew Paris’. Art Bulletin 81.4 (1999): 598–622. Cooper, David. ‘Critical Literary Cartography: Text, Maps and a Coleridge Notebook’. In Mapping Cultures: Place, Practice, Performance. Ed. Les Roberts. Basingstoke: Palgrave Macmillan, 2012. 29–52. Cooper, David, Christopher Donaldson and Patricia Murrieta-Flores, eds. Literary Mapping in the Digital Age. London: Routledge, 2017. Cooper, David and Ian Gregory. ‘Mapping the English Lake District: A Literary GIS’. Transactions of the Institute of British Geographers 36 (2011): 89–108. Cooper, David and Gary Priestnall. ‘The Processual Intertextuality of Literary Cartographies: Critical and Digital Practices’. The Cartographic Journal 48.4 (2011): 250–62. Cornell, E. H. and D. H. Hay. ‘Children’s Acquisition of a Route via Different Media’. Environment and Behaviour 16 (1984): 627–41. Cosgrove, D. Mappings. Ed. D. Cosgrove. London: Reaktion Books, 1999. Couclelis, H., R. G. Golledge, N. Gale and W. Tobler. ‘Exploring the Anchor-Point Hypothesis’. Journal of Environmental Psychology 7.2 (1987): 99–122. Crampton, Jeremy W. Mapping: A Critical Introduction to Cartography and GIS. Chichester: Wiley-Blackwell, 2010.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

311

Crampton, Jeremy W. and J. Krygier. ‘An Introduction to Critical Cartography’. ACME: An International E-Journal for Critical Geographies 4.1 (2006): 11–33. Crispin, Edmund. The Case of the Gilded Fly. London: Victor Gollancz, 1944. Daiches, David. Literary Landscapes of the British Isles: A Narrative Atlas. London: Paddington Press, 1979. Davis, Colin. ‘Psychoanalysis, Detection, and Fiction: Julia Kristeva’s Detective Novels’. The Journal of Twentieth Century/Contemporary French Studies 6.2 (2010): 294–306. Defoe, Daniel. Robinson Crusoe. Ed. Michael Shinagel. New York; London: W.W. Norton & Co., 1975. Delano-Smith, Catherine and Roger J. P. Kain. English Maps: A History. London: The British Library, 1999. English Maps: A History. Toronto: University of Toronto Press, 1969. Derrida, Jacques et al. ‘The Purveyor of Truth’. Yale French Studies 52 (1975): 31–113. Dickens, Charles. Martin Chuzzlewit. Oxford: Clarendon Press, 1982. Didicher, Nicole E. ‘Mapping the Distorted Worlds of Gulliver's Travels’. Lumen 16 (1997): 179–96. Dodge, Martin, Rob Kitchin and Chris Perkins, eds. Rethinking Maps: New Frontiers in Cartographic Theory. London: Routledge, 2009. Doherty, Peter. ‘Mappaemundi, Maps and the Romantic Aesthetic in Children’s Books’. Children’s Literature in Education 48.1 (2017): 89–102. Donaldson, Chris. ‘The Depth of Walter Scott’s Geography’. In Romantic Cartographies. Ed. Sally Bushell, Julia Carlson and Damian Walford-Davies. Cambridge: Cambridge University Press, 2020. Downs, Roger M. and David Stea. ‘Cognitive Maps and Spatial Behaviour’. In Image and Environment: Cognitive Mapping and Spatial Behavior. Ed. Roger M. Downs and David Stea. New Brunswick, NJ; London: Aldine, 1973. 8–26. Doyle, Arthur Conan. The Original Illustrated Strand. Sherlock Holmes. The Complete Facsimile Edition. Ware: Wordsworth Editions, 1996. Drabble, Margaret. A Writer’s Britain. London: Thames & Hudson, 1979. Dunne, Jörg. ‘Map Line Narratives’. In Literature and Cartography: Theories, Histories, Genres. Ed. Anders Engberg-Pedersen. Cambridge, MA:MIT Press, 2017. 352–79. Eco, Umberto. ‘On the Impossibility of Drawing a Map of the Empire on a Scale of 1:1’. In How to Travel with a Salmon and other Essays. Trans. William Weaver. London: Secker & Warburg, 1994. Edney, Matthew H. ‘The Irony of Imperial Mapping’. In The Imperial Map. Ed. James R. Akerman. Chicago, IL; London: University of Chicago Press, 2009. 11–45. Mapping an Empire: The Geographical Construction of British India, 1765–1843. Chicago, IL; London: University of Chicago Press, 1990.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

312

Bibliography

Edson, Evelyn. Mapping Time and Space: How Medieval Mapmakers Viewed Their World. London: British Library, 1997. Edwards, Jess. ‘Study, Marketplace and Labyrinth: Geometry As Rhetoric’. New Formations 57 (2005–6): 126–44. Writing, Geometry and Space in Seventeenth-Century England and America: Circles in the Sand. London: Routledge, 2006. Edwards, Philip. The Story of the Voyage: Sea Narratives in Eighteenth Century England. Cambridge: Cambridge University Press, 2004. Eichenbaum, Howard and Neal Cohen, ‘Can We Reconcile the Declarative Memory and Spatial Navigation Views of the Hippocampal Function?’ Neuron 83.4 (2014): 764–70. Eidam, Laura. ‘Re-examining Illustration’s Role in Treasure Island: Do Images Pirate Texts?’ English Literature in Transition 1880–1920 55.1 (2012): 45–68. Ekman, Stefan. Here Be Dragons: Exploring Fantasy Maps and Settings. Middletown, CT: Wesleyan University Press, 2013. Eliot, George. Adam Bede. Ed. Carol A. Martin. Oxford: Clarendon Press, 2001. Daniel Deronda. Ed. Barbara Hardy. Harmondsworth: Penguin, 1967. Emsley, Clive. Police Detectives in History, 1750–1950. Aldershot: Ashgate, 2006. Engberg-Pedersen, Anders. Empire of Chance: The Napoleonic War and the Disorder of Things. Cambridge, MA: Harvard University Press, 2015. Literature and Cartography: Theories, Histories Genres. Boston, MA: MIT Press, 2017. Erskine-Hill, Howard. Jonathan Swift: Gulliver’s Travels. Cambridge: Cambridge University Press, 1993. Escott, T. H. S. Anthony Trollope: His Work, Associates and Literary Originals. London: John Lane, 1913. Esrock, Ellen J. The Reader’s Eye: Visual Imaging As Reader Response. Baltimore, MD; London: Johns Hopkins University Press, 1994. Felix, Charles [Charles Warren Adams].The Notting Hill Mystery. London: Saunders & Otley, 1865. Ferguson, E. L. and M. Hegarty. ‘Properties of Cognitive Maps Constructed from Texts’. Memory and Cognition 22 (1994): 455–73. Finch, Alison. ‘Reality and Its Representation’. In Cambridge Companion to the French Novel from 1800 to the Present. Ed. Timothy Unwin. Cambridge: Cambridge University Press, 1997. Findlay, Kirsty. Arthur Ransome’s Long-Lost Study of Robert Louis Stevenson. Boydell: Woodbridge, 2011. Forster, E. M. Where Angels Fear to Tread. Ed. Oliver Stallybrass. London: Edward Arnold, 1975. Foucault, Michel. This is Not a Pipe. Trans. and ed. James Harkness. Berkeley: University of California Press, 1987. Fraser, Robert. Victorian Quest Romance: Stevenson, Haggard, Kipling, and Conan Doyle. Plymouth: Northcote House, 1998.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

313

Furst, Lilian R., ed. Realism. New York: Longman, 1992. Gadamer, Hans-Georg. Truth and Method [1960], 2nd ed. Trans. Joel Weinsheimer and Donald G. Marshall. New York; London: Continuum, 2003. Garling, T., A. Book, E. Lindberg and T. Nilsson. ‘Memory for the Spatial Layout of the Everyday Physical Environment: Factors Affecting Rate of Acquisition’. Journal of Environmental Psychology 1 (1981): 263–77. Garlitz, Barbara. ‘The Immortality Ode: Its Cultural Progeny’. Studies in English Literature 6 (1966): 639–49. Garner, Alan. The Weirdstone of Brisingamen. London: Collins, 1960. Elidor. London: Collins, 1965. Gatrell, Simon. ‘Wessex’. In Cambridge Companion to Thomas Hardy. Ed. Dale Kramer. Cambridge: Cambridge University Press, 1999. Genette, Gérard. Narrative Discourse: An Essay in Method. Trans. Jane E. Lewin. Ithaca, NY: Cornell University Press, 1979. Gerould, Winifred Gregory and James Thayer Gerould, eds. A Guide to Trollope. Princeton, NJ: Princeton University Press, 1968. Goldman, Paul. Victorian Illustrated Books 1850–1870: The Heyday of WoodEngraving. London: British Museum Press, 1994. Golledge, R. G. ‘Place Recognition and Wayfinding: Making Sense of Space’. Geoforum 23 (1992): 199–214. ‘Representing, Interpreting and Using Cognised Environments’. Papers and Proceedings of the Regional Science Association 41 (1978): 169–204. Goodey, Brian. ‘Mapping “Utopia”: A Comment on the Geography of Sir Thomas More’. Geographical Review 60.1. (1970): 15–30. Gould, Peter and Rodney White. Mental Maps. Harmondsworth: Penguin, 1974. Grahame, Kenneth.The Wind in the Willows [1908]. London: Methuen & Co., 1931. Graves, Matthew. ‘Maps and Texts: Reading Literary Maps – The Case of The Riddle of the Sands’, Trans 16.9.6 (July 2006). Internet-Zeitschrift für Kulturwissenschaften (Knowledge Networking in Cultural Studies): www .inst.at/trans/16Nr/09_6/graves16.htm. Griffith, John. ‘Making Wishes Innocent: Peter Pan’, The Lion and the Unicorn 3 (1979): 28–37. Haggard, Henry Rider. King Solomon’s Mines. Cassell & Co., 1885. Haggard, Lilias Rider. The Cloak That I Left: A Biography of the Author Henry Rider Haggard. Ipswich: Boydell Press, 1951. Hall, Joseph. Mundus Alter et Idem (An Old World and a New). Ed. H. J. Andersen. London: George Bell & Sons, 1908. Hamer, Mary. Writing by Numbers: Trollope’s Serial Fiction. Cambridge: Cambridge University Press, 1987. Hammond, Wayne G. & Christina Scull, eds. The Art of The Hobbit. London: HarperCollins, 2011.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

314

Bibliography

The Art of The Lord of The Rings. London: HarperCollins, 2015. J. R. R. Tolkien: Artist & Illustrator. Boston, MA: Houghton Mifflin, 2000. Hardy, Thomas. ‘Candour in English Fiction’. The New Review, 2 (January 1890): 15–21. Collected Letters of Thomas Hardy, Volume I 1840–1892. Ed. R. L. Purdy and Michael Millgate. Oxford: Clarendon Press, 1979. Collected Letters of Thomas Hardy, Volume II 1893–1901. Ed. R. L. Purdy and Michael Millgate. Oxford: Clarendon Press, 1980. Collected Letters of Thomas Hardy, Volume III 1902–1908. Ed. R. L Purdy and Michael Millgate. Oxford: Clarendon Press, 1982. Collected Letters of Thomas Hardy Volume IV 1909–1913. Ed. R. L. Purdy and Michael Millgate. Oxford: Clarendon Press, 1984. Thomas Hardy: Tess of the D’Urbervilles. London; New York: W.W. Norton & Co., 1991. Thomas Hardy: The Return of the Native. London; New York: W.W. Norton & Co., 2006. Hardyment, Christina. Arthur Ransome & Captain Flint’s Trunk. London: Jonathan Cape, 1984. Harley, J. B. ‘Deconstructing the Map’. Cartographica 26.2 (1989): 1–20. ‘Maps, Knowledge and Power’ [1988]. In The New Nature of Maps. Ed. Paul Laxton. Intro. J. H. Andrews. Baltimore, MD: Johns Hopkins University Press, 1997. 277–312. ‘Power and Legitimation in the English Geographical Atlases of the Eighteenth Century’ [1997]. In The New Nature of Maps. Ed. Paul Laxton. Intro. J. H. Andrews. Baltimore, MD: Johns Hopkins University Press, 1997. 110–47. ‘Silences and Secrecy: The Hidden Agenda of Cartography in Early Modern Europe’ [1988]. In The New Nature of Maps. Ed. Paul Laxton. Intro. J. H. Andrews. Baltimore, MD: Johns Hopkins University Press, 1997. 84–107. ‘Texts and Contexts in the Interpretation of Early Maps [1990]’. In The New Nature of Maps. Ed. Paul Laxton. Intro. J. H. Andrews. Baltimore, MD: Johns Hopkins University Press, 1997. 34–49. Harley, J. B and David Woodward. ‘Preface’. In History of Cartography, Vol. 1. Cartography in Prehistoric, Ancient and Medieval Europe and the Mediterranean. Chicago, IL: University of Chicago Press, 1987. Harmon, Katharine. You Are Here: Personal Geographies and Other Maps of the Imagination. New York: Princeton Architectural Press, 2004. Harpold, Terry. ‘Verne’s Cartographies’. Science Fiction Studies 32.1 (March 2005): 18–42. Harvey, John. Victorian Novelists and Their Illustrators. London: Sidgwick and Jackson, 1970. Harvey, P. D. A. The History of Topographical Maps: Symbols, Pictures and Surveys. London: Thames & Hudson, 1980.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

315

Mappa Mundi: The Hereford World Map. London: Hereford Cathedral and The British Library, 1996. Medieval Maps. London: British Library, 1991. Harwood, Jeremy. To the Ends of the Earth: 100 Maps that Changed the World. London: Marshall Editions, 2006. Haslam, Sara. ‘Wessex Literary Pilgrims, and Thomas Hardy’. Literary Tourism and Nineteenth Century Culture. Ed. Nicola Watson. London: Palgrave Macmillan, 2010. 164–74. Haycraft, Howard.The Art of the Mystery Story: A Collection of Critical Essays. New York: Carroll & Graf, 1974. Heft, Harry. ‘The Ecological Approach to Navigation: A Gibsonian Perspective’. In The Construction of Cognitive Maps. Ed. Juval Portugali. Dordrecht: Springer Publishing, 1996. 105–32. Herman, David. StoryLogic: Problems and Possibilities of Narrative. Lincoln; London: University of Nebraska Press, 2002. Heuser, R., M. Algee-Hewitt, E. Steiner and V. Tran. ‘Mapping the Emotions of London, 1700–1900: A Crowdsourcing Experiment’. In Literary Mapping in the Digital Age. Ed. David Cooper, Christopher Donaldson and Patricia Murrieta-Flores. London: Routledge, 2017. 25–46. Hewitt, Rachel. Map of a Nation: A Biography of the Ordnance Survey. London: Granta, 2010. Higgins, D. S. The Great Storyteller. London: Cassell, 1981. Higonnet, Margaret R. ‘The Playground of the Peritext’. Children's Literature Association Quarterly 15.2 (1990): 47–9. Hodgkiss A. G. and A. F. Tatham. Keyguide to Information Sources in Cartography. London: Mansell, 1986. Hones, Sheila. Literary Geographies: Narrative Space in Let the Great World Spin. Basingstoke: Palgrave Macmillan, 2018. ‘Text As It Happens: Literary Geography’. Geography Compass 2.5 (2008): 1301–17. Hopkins, Martha and Michael Buscher. Language of the Land: The Library of Congress Book of Literary Maps. Washington, DC: Library of Congress, 1999. Horace. Horace: Satires, Epistles and Ars Poetica. Trans. H. R. Fairclough. London: Loeb Classical Library. 1936. Hughes, Felicity A. ‘Children's Literature: Theory and Practice’. ELH 45 (Autumn 1978): 542–61. Hunt, Peter. Approaching Arthur Ransome. London: Jonathan Cape, 1991. ‘Landscapes and Journeys, Metaphors and Maps: The Distinctive Feature of English Fantasy’. Children’s Literature Association Quarterly 12.1 (Spring 1987): 11–14. ‘Ransome Revisited: A Structural and Developmental Approach’. Children’s Literature in Education 12.1 (1981): 24–33.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

316

Bibliography

Hunt, Peter and Millicent Lenz. Alternative Worlds in Fantasy Fiction. London; New York: Continuum, 2001. Iser, Wolfgang. The Act of Reading: A Theory of Aesthetic Response. Baltimore, MD; New York: Johns Hopkins Press, 1978. Jacob, Christian. The Sovereign Map: Theoretical Approaches to Cartography through History. Ed. Edward H. Dahl. Trans. Tom Conley. Chicago, IL; London: University of Chicago Press, 2006. Jacobs, Lucia F. and Françoise Schenk. ‘Unpacking the Cognitive Map: The Parallel Map Theory of Hippocampal Function’. Psychological Review 110 (2003): 285–315. James, Henry. ‘Preface to The Golden Bowl’ [New York ed., 1909]. In Literary Criticism: French Writers; Other European Writers; The Prefaces to the New York Edition. New York: Library of America, 1984. Jameson, Fredric. ‘Cognitive Mapping’. In Marxism and the Interpretation of Culture. Ed. Cary Nelson and Lawrence Grossberg. Basingstoke: Macmillan, 1988. 347–57. Postmodernism, or, the Cultural Logic of Late Capitalism. Durham, NC: Duke University Press, 1991. Johnson, Barbara. ‘The Frame of Reference: Poe, Lacan, Derrida’. Yale French Studies 55/56 (1977): 457–505. Keith, W. J. ‘Thomas Hardy and the Literary Pilgrims’. Nineteenth Century Fiction 24 (1969): 80–92. Kendall-Price, Claire. In the Footsteps of the Swallows and Amazons. Dorchester: Wild Cat Publishing, 1993. Kendrick, Walter M. ‘Balzac and British Realism: Mid-Victorian Theories of the Novel’. Victorian Studies 20 (1976): 5–24. Kitchin, Rob. ‘Exploring Spatial Thought’. Environment and Behaviour 29 (1997): 123–56. Kitchin, Rob and Mark Blades. The Cognition of Geographic Space. London; New York: I.B. Tauris, 2002. Kitchin, Rob and Scott Freundschuh. Cognitive Mapping: Past, Present and Future. London; New York: Routledge, 2000. Kitchin, Rob, Chris Perkins and Martin Dodge. ‘Thinking about Maps’. In Rethinking Maps: New Frontiers in Cartographic Theory. Ed. Rob Kitchin, Chris Perkins and Martin Dodge. London: Routledge, 2009. 1–25. Knox, R. A. ‘A Ramble in Barsetshire’. The London Mercury 5 (1922): 378–85. ‘Introduction’, The Best English Detective Stories of 1928. London: Faber, 1929. Kolb, Margaret. ‘Plot Circles: Hardy’s Drunkards and Their Walks’. Victorian Studies 56 (Summer 2014): 595–623. Korzybski, Alfred. Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics [1933]. 5th ed. New York: Institute of General Semantics, 2000.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

317

Krishnamurth, Sailaja. ‘Reading between the Lines: Geography and Hybridity in Rudyard Kipling’s Kim’. Victorian Review 28.1 (2002): 47–65. Krygier, J. and D. Wood. ‘Critical Cartography’. In International Encyclopedia of Human Geography. Ed. Rob Kitchin and Nigel Thrift. Amsterdam: Elsevier Ltd (2009), 340–4. Lacan, Jacques. ‘Seminar on the Purloined Letter’. Yale French Studies 48 (1972): 39–72. Lando, Fabio. ‘Fact and Fiction: Geography and Literature’. Geojournal 38.1 (1996): 3–18. Larsen, Reif. The Selected Works of T. S. Spivet. London: Vintage, 2010. Lesnik-Oberstein, Karen. Children’s Literature: Criticism and the Fictional Child. Oxford: Clarendon Press, 1994. Lessing, Gotthold Ephraim. Laocoön: An Essay on the Limits of Painting and Poetry [1766]. Trans. Edward Allen McCormick. Baltimore, MD; London: Johns Hopkins University Press, 1984. Lewes, George. ‘Realism in Art: Recent German Fiction’. Westminster Review 70 (1858): 488–518. Lewis, Suzanne. The Art of Matthew Paris in the ‘Chronica majora’. Berkeley: University of California Press, 1987. Liddle, Dallas. ‘“Anatomy of a Nine Days’ Wonder”: Sensation Journalism in the Decade of the Sensation Novel’. In Victorian Crime, Madness and Sensation. Ed. Andrew Maunder and Grace Moore. Aldershot: Ashgate, 2016. 89–104. Linde, Charlotte and William Labov. ‘Spatial Networks As a Site for the Study of Language and Thought’. Language 51 (1975): 924–39. Ljungberg, Christina. ‘Cartographic Strategies in Contemporary Fiction’. In Orientations: Space / Time / Image / Word. Ed. Claus Clüver, Leo Hoek, Véronique Plesch and Peter de Voogd. Amsterdam: Rodopi, 2005. 155–72. Creative Dynamics: Diagrammatic Strategies in Narrative. Amsterdam: John Benjamins, 2012. ‘Dynamic Instances of Interaction: The Performative Function of Iconicity in Literary Texts’. Sign Systems Studies 38 (2010): 271–93. ‘Mapping Practices for Different Geographies’. In Mapping Different Geographies:Lecture Notes in Geoinformation and Cartography. Ed. K. Kriz et al. Berlin; Heidelberg: Springer-Verlag, 2010. 38–55. Lynch, Kevin. The Image of the City. Cambridge, MA; London: MIT Press, 1960. Macguire, Eleanor A. and Sinéad L. Mullally. ‘The Hippocampus: A Manifesto for Change’. Journal of Experimental Psychology 142.4 (2013): 1180–9. Macleod, Jeffrey J. and Anna Smol. ‘Visualizing the Word: Tolkien As Artist and Writer’. Tolkien Studies 14 (2017): 115–31. Maine, Fiona and Alison Waller. ‘Swallows and Amazons Forever: How Adults and Children Engage in Reading a Classic Text’. Children’s Literature in Education, 42 (2011): 354–7.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

318

Bibliography Makinen, M. ‘Agatha Christie 1890–1976’. A Companion to Crime Fiction. Ed. Charles J. Rzepka and Lee Horsley. Oxford: Wiley-Blackwell, 2010. 415–26. Manguel, Alberto and Gianni Guadalupi. The Dictionary of Imaginary Places. New York: Macmillan Publishing Co. Ltd, 1980. Maupassant, Guy de. Oeuvres complètes de Guy de Maupassant, Vol. 19: Pierre et Jean. Paris: Louis Conard, 1929. McClintock, Anne. Imperial Leather: Race, Gender and Sexuality in the Colonial Context. London: Routledge, 1995. McManis, Douglas. ‘Places for Mysteries’. Geographical Review 68.3 (1978): 319–34. Mendlesohn, Farah. Rhetorics of Fantasy. Middletown CT: Wesleyan University Press, 2008. Mezei, Kathy. ‘Spinsters, Surveillance and Speech: The Case of Miss Marple, Miss Mole and Miss Jeckyll’. Journal of Modern Literature 30.2 (Winter 2007): 102–20. Miall, David S. ‘Affect and Narrative: A Model of Response to Stories’. Poetics 17 (1988): 259–72. Literary Reading: Empirical and Theoretical Studies. New York: Peter Lang, 2006. Miller, J. Hillis. Illustration. London: Reaktion Books, 1992. On Literature: Thinking in Action. London: Routledge, 2000. Topographies. Stanford, CA: Stanford University Press, 1995. Miller, J. Hillis and David Borowitz. ‘The Fiction of Realism’. In Charles Dickens and George Cruikshank. Los Angeles, CA: William Andrews Clark Memorial Library Publication, 1971. 1–69. Millgate, Michael. Thomas Hardy: His Career As a Novelist. London: The Bodley Head, 1971. Milne, A. A. Winnie-the-Pooh. London: Methuen & Co. Ltd, 1926. Moll, Herman, A New and Correct Map of the Whole World Shewing the Situation of Its Principal Parts ... (London: H. Moll, T & J. Bowles, P. Overton and J. King, 1719). Monmonier, Mark. How to Lie with Maps. Chicago, IL; London: University of Chicago Press, 1991. Mapping It Out: Expository Cartography for the Humanities and Social Sciences. Chicago, IL; London: University of Chicago Press, 1993. Moody, Ellen. ‘Mapping Trollope or Geographies of Power’. Victorian Web, July 2013: www.victorianweb.org/authors/trollope/moody3.html. Moore, John R. ‘The Geography of Gulliver’s Travels’. Journal of German and English Philology 40 (1941): 214–28. More, Thomas. Utopia. Ed., rev. and trans. George M. Logan. New York: W.W. Norton, 2011. Moretti, Franco. Atlas of the European Novel: 1800–1900. London: Verso, 1998. Distant Reading. London: Verso, 2013.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

319

Graphs, Maps, Trees: Abstract Models for a Literary History. London; New York: Verso, 2004. Morris, Pam. Realism. London: Routledge, 2003. Muehrcke, P. C. with J. O. Muehrcke. Map Use: Reading, Analysis, and Interpretation. Madison, WI: J. P. Publications, 1978. Murietta-Flores, Patricia and Naomi Howell. ‘Towards the Spatial Analysis of Vague and Imaginary Place and Space: Evolving the Spatial Humanities through Medieval Romance’. Journal of Map and Geography Libraries 13.1 (2017): 29–57. Ogborn, Miles. ‘Mapping Words’. New Formations 57 (2005–6): 145–9. O’Keefe, John and Lynn Nadel. The Hippocampus As a Cognitive Map. Oxford: Clarendon Press, 1978. Oliver, Richard. Ordnance Survey Maps: A Concise Guide for Historians. London: The Charles Close Society, 1993. Osbourne, Lloyd. ‘Note’. The Works of Robert Louis Stevenson. Tusitala Edition. London: William Heinemann Ltd, 1923. Padrón, Ricardo. ‘Mapping Imaginary Worlds’. In Maps: Finding Our Place in the World. Ed. James Akerman and Robert Karrow. Chicago, IL: University of Chicago Press, 2007. 256–87. The Spacious Word: Cartography, Literature and Empire in Early Modern Spain. Chicago, IL; London: University of Chicago Press, 2004. Pageaux, Daniel-Henri. ‘Eléments pour une géosymbolique. Littérature générale et comparée et géographie’. In Savoirs et Litterature / Literature, the Humanities and the Social Sciences. Ed. Jean Bessière. Paris: Sorbonne Nouvelle, 2002, 75–92. Painter, G. D. ‘The Matter of Authenticity’. The Geographical Journal 140.2 (1974): 191–4. Paivio, Allan. Mental Representations: A Dual Coding Approach. Oxford: Oxford University Press, 1986. ‘The Mind’s Eye in Arts and Science’. Poetics 12 (1983): 1–18. Pavlik, Anthony. ‘A Special Kind of Reading Game: Maps in Children’s Literature’. International Research in Children’s Literature 3.1 (2010): 28–43. Peirce, Charles. Collected Papers of Charles Sanders Peirce. Vol. I Principles of Philosophy and Vol. II Elements of Logic. Ed. Charles Hartshorne and Paul Weiss. Cambridge, MA: The Belknap Press of Harvard University Press, 1960. Peluso, N. L. ‘Whose Woods Are These? Counter-Mapping Forest Territories in Kalimantan, Indonesia’. Antipode 27 (1995): 383–406. Peraldo, Emmanuelle. Literature and Geography: The Writing of Space throughout History. Newcastle: Cambridge Scholars Publishing, 2016. Pettigrew, Dulcie. ‘Swallows and Amazons Explored: A Reassessment of Arthur Ransome’s Books for Children’. New Review of Children’s Literature and Librarianship 15 (2009): 1–20.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

320

Bibliography

Phillips, Richard. Mapping Men and Empire: A Geography of Adventure. London; New York: Routledge, 1997 Piatti, Barbara. ‘Mapping Fiction: The Theories, Tools and Potentials of Literary Cartography’. Literary Mapping in the Digital Age. Ed. D. Cooper, C. Donaldson and P. Murrieta-Flores. London: Routledge, 2017. 88–101. Piatti, Barbara, H. R. Bär, A. K. Reuschel, L. Hurni and W. Cartwright. ‘Mapping Literature: Towards a Geography of Fiction’. In Cartography and Art. Ed. W. Cartwright, G. Gartner and A. Lehn. Berlin; Heidelberg: Springer, 2009. 1–16. Piatti, Barbara and Lorenz Hurni. ‘Editorial: Cartographies of Fictional Worlds’. The Cartographic Journal 48.4 (2011): 218–23. Piper, Leonard. Dangerous Waters: The Life and Death of Erskine Childers. Hambledon: Continuum, 2003. Pite, Ralph. Hardy’s Geography: Wessex and the Regional Novel. Basingstoke: Palgrave Macmillan, 2002. Plietzsch, Birgit. The Novels of Thomas Hardy As a Product of Nineteenth-Century Social, Economic and Cultural Change. Berlin: Tenea-Verlag Ltd, 2004. Pocock, Douglas. Humanistic Geography and Literature: Essays on the Experience of Place. London: Croom Helm, 1981. Poovey, Mary. ‘Trollope’s Barsetshire Series’. In Cambridge Companion to Anthony Trollope. Ed. Carolyn Dever and Lisa Niles. Cambridge: Cambridge University Press, 2010. 31–43. Post, J. B. An Atlas of Fantasy [1973], rev. ed. New York: Ballantyne Books, 1979. Priestnall, Gary. ‘Spatial Frames for Literature through Geospatial Technologies’. In Literary Mapping in the Digital Age. Ed. D. Cooper, C. Donaldson and P. Murrieta-Flores. London: Routledge, 2017. 240–55. Pyrhönen, Heta. ‘Criticism and Theory’. In A Companion to Crime Fiction. Ed. Charles J. Rzepka and L. Horsley. Oxford: Wiley-Blackwell, 2010. 43–56. Ransome, Arthur. The Autobiography of Arthur Ransome. London: Jonathan Cape, 1976. ‘A Letter to the Editor’. The Junior Bookshelf 1: 4 (1937): 2–5. The Picts and the Martyrs: or Not Welcome at All. London: Jonathan Cape, 1943. Pigeon Post. London: Jonathan Cape, 1936. Swallowdale. London: Jonathan Cape, 1931. Swallows and Amazons. London: Jonathan Cape, 1930 We Didn't Mean to Go to Sea. London: Jonathan Cape, 1937. Winter Holiday. London: Jonathan Cape, 1933. Redish, A. David. Beyond the Cognitive Map: From Place Cells to Episodic Memory. Cambridge, MA; London: MIT Press, 1999. Richardson, Alan. ‘Romanticism and the End of Childhood’. Nineteenth Century Contexts 21 (2008): 169–89. Ristow, Walter W. ‘Lithography and Maps: 1789–1850’. In Five Centuries of Map Printing. Ed. David Woodward. Chicago, IL: University of Chicago Press, 1975. 77–112.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

321

Robinson, Arthur H. with Randall. D. Sale, Joel L. Morrison and Phillip. C. Muehrcke. Elements of Cartography, 5th ed. New York: John Wiley & Sons, 1984. Rodes, Scott. ‘Figuring the Map in The Return of the Native’. In Reading and Mapping Hardy’s Roads. London: Routledge, 2008. 19–58. Rose, Jacqueline. The Case of Peter Pan or The Impossibility of Children’s Fiction. Philadelphia: Penn State Press, 1990. Rosebury, Brian. Tolkien: A Cultural Phenomenon. London: Palgrave Macmillan, 2003. Rossetto, Tania. ‘Theorizing Maps with Literature’. Progress in Human Geography 38 (2014): 513–30. Rousseau, Jean-Jacques. Émile or On Education. Trans Alan Bloom. Harmondsworth: Penguin Books, 1991. Oeuvres complètes IV: Émile. Ed. Bernard Gagnebin and Marcel Raymond. Paris: Editions Gallimard, 1969. Russell, William. The Recollections of a Detective Police Officer by ‘Waters’. London: W. Kent & Co., 1849. Ryan, Marie-Laure. ‘Cognitive Maps and the Construction of Narrative Space’. In Narrative Theory and the Cognitive Sciences. Ed. David Herman. Stanford, CA: CSLI Publications, 2003. 214–42. ‘Diagramming Narrative’. Semiotica 165 (2007): 11–40. ‘Toward a Visual Narratology’. In What Is Narratology? Questions and Answers Regarding the Status of a Theory. Ed. T. Kindt and H. H. Muller. Berlin: Walter De Geuyer and Co., 2003. 333–64. Ryan, Marie-Laure, Kenneth Foote and Maoz Azaryahu. Narrating Space / Spatialising Narrative: Where Narrative Theory and Geography Meet. Columbus: Ohio State University Press, 2016. Sachuko-Cecire, Maria, Hannah Field, Kavita Mudan Finn and Malini Roy. Space and Place in Children’s Literature, 1789 to the Present. Aldershot: Ashgate, 2015. Sadleir, Michael. Trollope: A Commentary. London: Constable & Co. Ltd, 1927. Said, Edward W. Culture and Imperialism. New York: Arnold A. Knopf, 1993. ‘Yeats and Decolonisation’. In Nationalism, Colonialism and Literature. Ed. Terry Eagleton, Fredric Jameson and Edward Said. Minneapolis: University of Minnesota Press, 1990. 69–95. Saunders, Angharad. ‘Literary Geography: Reforging the Connections’. Progress in Human Geography 34 (2010): 436–52. Sayers, Dorothy L. Five Red Herrings. London: Victor Gollancz, 1931. The Letters of Dorothy L. Sayers: 1899–1936: The Making of a Detective Novelist. Ed. Barbara Reynolds. London: Hodder & Stoughton, 1995. Scott, Walter. The Letters of Sir Walter Scott, 12 vols. Ed. H. J. C. Grierson. London: Constable & Co., 1832–6. Seed, David. ‘Spy Fiction’. In The Cambridge Companion to Spy Fiction. Ed. Martin Priestman. Cambridge: Cambridge University Press, 2003. 115–34.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

322

Bibliography Sharp, Joanne P. ‘Towards a Critical Analysis of Fictive Geographies’. Area 32.3 (2000): 327–34. Sharp, William. Literary Geography. London: Pall Mall Publications, 1904. Sheehan, Lucy. ‘Anthony Trollope’s “English Tale, on English Life with Clerical Flavour”’. Fortnightly Review (18 September 2011): http://fortnightlyreview .co.uk/2011/09/trollope-sheehan/. Shippey, T. A. The Road to Middle-Earth. London: George Allen & Unwin, 1928. Shpayer-Makov, Haia. The Ascent of the Detective: Police Sleuths in Victorian and Edwardian England. Oxford: Oxford University Press, 2011. Siegel, A. W. and S. White. ‘The Development of Spatial Representation of LargeScale Environments’. In Advances in Child Development and Behaviour. Ed. H. Reese. New York: Academic Press, 1975. 9–55. Simpson, J. A. and E. S. C. Weiner. The Oxford English Dictionary. 2nd ed. Oxford: Clarendon Press, 1989. Sipe, Lawrence R. and Caroline E. McGuire. ‘Picturebook Endpapers: Resources for Literary and Aesthetic Interpretation’. Children's Literature in Education 37 (2006): 291–304. Smith, David. Victorian Maps of the British Isles. London: B. T. Batsford Ltd, 1985. Snell, K. D. M. ‘“A Drop of Water from a Stagnant Pool?” Inter-War Detective Fiction and the Rural Community’. Social History 35.1 (2010): 21–50. Sorum, Eve. ‘Hardy’s Geography of Narrative Empathy’. Studies in the Novel 43 (2011): 179–99. Squire, Shelagh. ‘The Cultural Values of Literary Tourism’. Annals of Tourism Research 21 (1994): 103–20. ‘Landscapes, Places and Geographic Spaces: Texts of Beatrix Potter as Cultural Communication’. GeoJournal 38.1 (1996): 75–86. Stafford, David A. T. ‘Spies and Gentlemen: The Birth of the British Spy Novel, 1893–1914’. Victorian Studies 24.4 (Summer 1981): 489–509. Stanzel, Franz K. A. Theory of Narrative [1979]. Trans. Charlotte Goedsche. Cambridge: Cambridge University Press, 1984. Steele, C. R. ‘From Hakluyt to Purchas’. In The Hakluyt Handbook, 2 vols. Ed. D.B. Quinn. London: Hakluyt Society, 1974. Steiner, Wendy. The Colors of Rhetoric: Problems in the Relation between Modern Literature and Painting. Chicago, IL; London: University of Chicago Press, 1982. Stendhal, The Life of Henry Brulard. Trans. John Sturrock. New York: New York Review Books, 1995. Oeuvres intimes de Stendhal: La Vie de Henry Brulard. Ed. Henri Martineau. Paris: Librairie Gallimard, 1955. Stevenson, Robert Louis. The Letters of Robert Louis Stevenson. Vol. 4. Oct.1882– June 1884. Ed. Bradford A. Booth and Ernest Mehew. New Haven, CT; London: Yale University Press, 1994.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

323

‘My First Book: “Treasure Island”’. In Essays in The Art of Writing. London: Chatto & Windus, 1911. 111–34. Treasure Island. London; Paris; New York: Cassel & Co., Ltd, 1883. Treasure Island [1883]. Ed. Emma Letley. Oxford: Oxford University Press, 1985. Stiebel, Lindy. Imagining Africa: Landscape in H. Rider Haggard’s African Romances. Westport, CT: Greenwood, 2001. Stockhammer, Robert. Kartierung der Erde, Macht und Lust in Karten und Literatur. Munich: Wilhelm Fink Verlag, 2007. Stoker, Bram, Dracula. London: Archibald Constable & Co., 1897. Stott, Rebecca. ‘The Dark Continent: Africa As Female Body in Haggard’s Adventure Fiction’. Feminist Review 32 (1989): 69–89. Strachey, Barbara. Journeys of Frodo: An Atlas of J. R. R. Tolkien’s The Lord of the Rings. London: Unwin Paperbacks, 1981. Sundmark, Bjorn. ‘Mapping Middle Earth: A Tolkienian Legacy’. In Maps and Mapping in Children’s Literature: Landscapes, Seascapes and Cityscapes. Ed. Nina Goga and Bettina Kümmerling-Meibauer. Amsterdam: John Benjamins Publishing Co., 2017. 221–37. Sutherland, John. Literary Landscapes: Charting the Real-Life Settings of the World’s Favourite Fiction. London: Modern Books, 2018. Swift, Jonathan. Gulliver’s Travels. Ed. Albert J. Rivero. New York: W.W. Norton & Co., 2002. Symons, Julian. Bloody Murder: From the Detective Story to the Crime Novel: A History. London: Faber and Faber, 1972. Tally, Robert T. Jr. ‘Jameson’s Project of Cognitive Mapping: A Critical Engagement’. Social Cartography: Mapping Ways of Seeing Social and Educational Change. Ed. R. Paulston. Levittown, PA: Taylor and Francis, 1996. 399–416. Literary Cartographies: Spatiality, Representation and Narrative. Basingstoke: Palgrave Macmillan, 2014. Spatiality. London: Routledge, 2013. Taylor, Holly A. and Barbara Tversky. ‘Spatial Mental Models Derived from Survey and Route Descriptions’. Journal of Memory and Language 31 (1992): 261–92. Terry, R. C., ed. Oxford Reader’s Companion to Trollope. Oxford: Oxford University Press, 1999. Thacker, Andrew. ‘The Idea of a Critical Literary Geography’. New Formations 57 (2005–2006): 56–73. Moving through Modernity: Space and Geography in Modernism. Manchester: Manchester University Press. 2003. Timpf, Sabine. ‘Insights into Mapping the Imagined World of J. R. R. Tolkien’. Binding Them All: Interdisciplinary Perspectives on J. R. R. Tolkien and His

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

324

Bibliography

Works. Ed. Monika Kirner-Ludwig, Stephan Köser and Sebastian Streitberger. Zollikofen: Walking Tree Publishers, 2018. 231–52. Tingay, Lance O. ‘Mapmaking in Barsetshire’. Trollopian 3 (1948): 19–32. Todorov, T. ‘The Typology of Detective Fiction’. In The Poetics of Prose. Trans. R. Howard. Oxford: Blackwell, 1977. 42–52. Tolkien, Christopher, ed. J. R. R. Tolkien, The Lost Road and Other Writings. Vol. V The History of Middle-Earth. Boston, MA: Houghton Mifflin Co., 1987. ed. J. R. R. Tolkien, The Return of the Shadow. Vol. VI The History of MiddleEarth. Boston, MA: Houghton Mifflin Co., 1988. ed. J. R. R. Tolkien, The Treason of Isengard: The History of the Lord of the Rings Part Two. Boston, MA: Houghton Mifflin Co., 1989. ed. J. R. R. Tolkien, The War of the Jewels. Vol. XI The History of Middle-Earth. Boston, MA: Houghton Mifflin Co., 1994. Tolkien, J. R. R. The Fellowship of the Ring: Being the First Part of The Lord of the Rings. London: George Allen & Unwin, 1954. The Hobbit or There and Back Again. London: George Allen & Unwin, 1937. Letters of J. R. R. Tolkien: A Selection. Ed. Humphrey Carpenter with the assistance of Christopher Tolkien. London: HarperCollins, 2006. The Return of the King: Being the Third Part of The Lord of The Rings. London: George Allen & Unwin, 1955. The Silmarillion. Ed. Christopher Tolkien. London: George Allen & Unwin, 1977. Tree and Leaf; Smith of Wooton Major; The Homecoming of Beorhthoth Beorhtheln’s Son. London: George Allen & Unwin, 1975. Tolman, E. C. ‘Cognitive Maps in Rats and Men’. Psychological Review 55 (1948): 189–208. Tooke, Adrianne. Flaubert and the Pictorial Arts: From Image to Text. Oxford: Oxford University Press, 2000. Trollope, Anthony. Anthony Trollope: An Autobiography and Other Writings. Ed. Nicholas Shrimpton. Oxford: Oxford University Press, 2014. Barchester Towers. Harmondsworth: Penguin, 1983. Dr Thorne. Harmondsworth: Penguin, 1991. Framley Parsonage. Harmondsworth: Penguin, 1984. Tuan, Yi-Fu. ‘Images and Mental Maps’. Annals of the Association of American Geographers 65 (1975): 205–13. ‘Literature and Geography: Implications for Geographical Research’. In Humanistic Geography: Prospects and Problems. Ed. David Ley and Marwyn S. Samuels. Chicago, IL: Maaroufa Press, 1978. 194–206. Tucker, Nicholas. ‘Arthur Ransome and Problems of Literary Assessment’. Children’s Literature in Education 26.2 (1995): 97–105. Turchi, Peter. Maps of the Imagination: The Writer As Cartographer. San Antonio, TX: Trinity University Press, 2004.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Bibliography

325

Turner, Mark. The Literary Mind: The Origins of Thought and Language. New York; London: Oxford University Press, 1996. Van Dine, S. S. ‘Twenty Rules for Writing Detective Stories’. The American Magazine, September 1928. Verner, Coolie. ‘Copperplate Printing’. In Five Centuries of Map Printing. Ed. David Woodward. Chicago, IL: University of Chicago Press, 1975. 51–75. Wagoner, M. S. Agatha Christie. Boston, MA: G.K. Hall & Co., 1986. Walford Davies, Damian. Cartographies of Culture: New Geographies of Welsh Writing in English. Cardiff: University of Wales Press, 2012. Walker, R. E. ‘The Cartography of Fantasy’. Mythlore: A Journal of J. R. R. Tolkien, C. S. Lewis, Charles Williams and the Genres of Myth and Fantasy Studies 7.4 (1981): 33–8. Walker, Ronald G. and June M. Frazer. The Cunning Craft: Original Essays on Detective Fiction and Contemporary Literary Theory. Macomb: West Illinois University Press, 1990. Wall, Barbara. The Narrator’s Voice: The Dilemma of Children’s Literature. Basingstoke: Palgrave Macmillan, 1991. Wall, Stephen. ‘Trollope, Balzac, and the Reappearing Character’. Essays in Criticism 25 (January 1975): 123–44. Wallis, Helen M. and Arthur H. Robinson, eds. Cartographical Innovations: An International Handbook of Mapping Terms to 1900. Tring: Map Collector Publications Ltd, 1987. Wardale, Roger. In Search of Swallows and Amazons. Wilmslow: Sigma Press, 1996. Watson, Nicola J. The Literary Tourist: Readers and Places in Romantic and Victorian Britain. Basingstoke: Palgrave Macmillan, 2006. Watson, Victor. Reading Series Fiction: From Arthur Ransome to Gene Kemp. New York; Oxford: Routledge, 2000. Wegner, Phillip E. ‘“Life As He Would Have It”: The Invention of India in Kipling’s Kim’. Cultural Critique 26 (Winter 1993–4): 129–59. Wilson, Angus. The Wild Garden: or, Speaking of Writing. Berkeley; London: University of California Press, 1963. Wojcik-Andrews, Ian. ‘The Family As an Ideological Construct in the Fiction of Arthur Ransome’. The Lion and the Unicorn 14.2 (1990): 7–15. Wood, Denis with John Fels. The Power of Maps. New York; London: Guilford Press, 1992. Wood, Denis and John Krygier. Making Maps: A Visual Guide to Map Design for GIS. 2nd ed. New York; London: Guilford Press, 2011. Woodward, David. ‘Medieval Mappaemundi’. In The History of Cartography, Vol 1. Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean. Ed. J. B. Harley and David Woodward. Chicago, IL: University of Chicago Press, 1987. ‘Techniques of Map Engraving, Printing, and Coloring in the European Renaissance’. The History of Cartography, Vol. 3. Cartography in the

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

326

Bibliography

European Renaissance. Ed. David Woodward. Chicago, IL: University of Chicago Press, 2007. 591–610. Woolf, Virginia. ‘Literary Geography’ [Times Literary Supplement, 1905]. In The Essays of Virginia Woolf, Vol 1: 1904–1912. Ed. A. Mcneillie. London: Hogarth Press, 1986. 32–6. Wordsworth, William. William Wordsworth: The Major Works. Ed. Stephen Gill. Oxford: Oxford University Press, 1984. Wyatt, John. The Use of Imaginary, Historical and Actual Maps in Literature. Lampeter: Edwin Mellen Press, 2013. Wynn Fonstad, Karen. The Atlas of Middle-Earth. Boston, MA, and New York: Houghton Mifflin Co., 1991. ‘Writing “TO” the Map’. Tolkien Studies 3 (2006): 133–6. Wynne Jones, Diana. The Tough Guide to Fantasyland. London: Gollancz, 1996. Yates, Frances. The Art of Memory. Chicago, IL: University of Chicago Press, 1966. Zornado, Joseph. L. Inventing the Child: Culture, Ideology and the Story of Childhood. New York; London: Garland, 2007. Zunshine, Lisa. Why We Read Fiction: Theory of Mind and the Novel. Columbus: Ohio State University Press, 2006.

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.010

Index

absence of map, 80 danger of usurpation of verbal by visual, 242, 245, 248 fear of map, 248, 261, 263 from canonical Realist texts, 200, 239, 241, 253 intrinsic visualisation of words, 245–8, 251 photography, 245, 251 resistance to illustration, 241 accuracy (literary maps) agency, 95, 99, 113 real-world correspondence vs correspondence to text, 94 referential confusion, 98 referential reality, 100 relative priorities, 100 truthfulness of map (over- or undervalued), 99, 102–4 usefulness of map relies upon its accuracy, 95–101 accuracy (of maps ), 29, 46, 53, 62, 75–9, 81, 88, 94, 203, 233, 304, See also map use; referential relations; truth actual maps of crime journalistic vs fictional, 128–9, 131, 134 map as part of method /part of manipulation, 132, 134–5 Adams, Charles Warren Notting Hill Mystery, The, 135–7 Adequinolaza, Fernando Cabo, 19 adventure fiction, 110, 114, 164, See also imperialism Akerman, James, 81 Allingham, Margery, 14, 143 More Work for the Undertaker, 160–2 Aristotle Poetics, 268 Atlas of Europe, The, 297–8 Austen, Jane, 266 authorial intention, 18, 63

authorial map, 6–7, 99, 112, 176, 182, 203, 229, 268, 271 authorial map-makers, 4 authorial mapping, 1, 203, 221 withholding of, 113, 261 authorial Realist mapping Anthony Trollope, 259–61 alignment of power, class, space, 257–9, 267 model of accretion/enlargement, 256, 265 Balzac, 254–5 George Eliot, 262 spatial models for Realism, 266–7, 271 Thomas Hardy readerly needs, 264 referential relationship between map and world, 268–72 referential relationship between real and fictional, 263–5 writerly position, 265 authority of map, 43, 162 authenticity of map, 94, 103, 108–10, 123, 199 duplicity of map, 110, 113, 123, 143 map misleads, 56, 65, 98–9, 105, 123, 146, 149–51, 243 map withholds, 35, 56, 106, 108, 146 Azaryahu, Maoz, 6, 284, 303 Bakhtin, Mikhail chronotope, 22, 302 Balzac, Honoré de, 298 La Comédie humaine, 254–5 Le Pére Goriot, 249 Barrie, J. M. Little White Bird, The, 192 Peter Pan, 166, 192, 197 Peter Pan: A Fantasy in Five Acts, 195–6 Peter Pan and Wendy, 193–6, 305–6 Peter Pan in Kensington Gardens, 192, 194, 196

327

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

328

Index

Barrow, Ian, 85 Barthes, Roland, 33, 278 Baucom, Ian, 86 Baudrillard, Jean Simulacra and Simulations, 238, 252 Bavidge, Jenny, 168 Baynes, Pauline, 229 Belgravia, The, 268 Belyea, Barbara, 100 Bible, the, 45, 282 Billman, Carol, 167 Blades, Mark, 287 Blake, William Songs of Innocence and Experience, 190 Bloom, Clive, 114 Bodenhamer, David, 299 Boer War, 77, 114, 117 Borges, Jorge Luis, 238 Bracher, Fredrick, 60–3 Bray, Patrick, 247 Brigade de Sûreté (Security Brigade), 129–30 British Empire, 84, 87 Brogan, Hugh, 177 Brosseau, Marc, 19 Brotton, Jerry, 58, 93 Brown, Howard, 129 Buchan, John, 116 Nelson’s History of the War, 77 Prester John, 114–16 Bunyan, John Pilgrim’s Progress, The, 44, 67, 69–74 Buscher, Michael, 2 Campbell, Tony, 88 Caquard, Sebastién, 297, 299 Carracciolo, Marco, 276 Carruthers, Mary, 279, 281–2, 286 cartographic creativity maps in process, 203 multiple maps, 203, 221 readerly published maps, 203, 205, 221–5, 253, 268 writerly working maps, 99–100, 203–4, 215–18, 247, 253, 268 Tolkien’s process integration of visual and verbal creativity, 203, 207, 218–20 itinerary structure, 217 need to externalise, 205, 207 use of road and river, 208–13, 215–17 cartographic imagination authorial need for visualisation, 247 map precedes text, 199, 203

Stevenson, 204 Tolkien, 14, 204–6, 216, 236 cartography. See also history of cartography; map production cartographic turn, 28 Case, Arthur E., 62 Cassini family, the, 75–6 Certeau, Michel de, 20, 283–5 Chaucer, Geoffrey Canterbury Tales, The, 67 Childers, Erskine Riddle of The Sands, The, 94–5, 114, 116–26 children’s fiction, 164 difficult to define, 165 as doubled space, 168 doubled audience, 183 and national identity, 168 narrative structure, 167 as spatial genre, 164 Christie, Agatha, 14 And Then There Were None, 151 Autobiography, An, 152–3 The Murder at the Vicarage, 141–54 The Murder of Roger Ackroyd, 146–51 Murder on the Orient Express, 151 The Mysterious Affair at Styles, 144–6, 149, 157 Chronotopic Cartographies, 302 Cicero, 280 De Oratore, 280 cognitive mapping, 275 anchor-point theory, 291–2 base map (locale) vs route map, 285–7, 289–90, 302 combined cognitive model, 286, 288, 303 cultural geographic models, 287–8 individual mapping strategies, 303 mind as structure of story, 282–3 parallel maps, 287, 301 processual, 276 relative vs absolute mapping, 285, 300–1, 304 route navigation models, 285, 294 spatial relations, 276 way-finding, 286, 288 Collé Bak, Natalie, 72 Collins, Wilkie, 130 colonialism. See imperialism Cooper, David, 28, 34 Cosgrove, Denis, 20 Crampton, Jeremy, 29–30

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

Index

Crispin, Edmund, 143 critical cartography, 6, 14, 16, 28, 30, 34–5, 43, 94, 99 cartographic silence, 94, 104–7 critical literary mapping, 14, 16, 33–6, 261 Cruikshank, George, 242 cultural geography, 16, 20, 275 Daiches, David, 18 Dampier, William, 58–60 Dante Inferno, 6, 281 deconstructive approach (children’s fiction) denial of genre of children's fiction, 165–6 Freudian reading, 165, 186 Jacqueline Rose, 165–8, 189–90 Karen Lesnik-Oberstein, 190–1 Defoe, Daniel, 57, 59 Robinson Crusoe, 57, 60–1, 65–6, 184 Delano-Smith, Catherine, 47, 68, 80 Derrida, Jacques, 140–1 Detective Branch, 130 detective fiction, 39, 42, 164 doubled audience, 163 doubled narrative, 143, 145, 162 doubled readership, 143 emergence of, 128–9 models of, 142 origins in crime reporting, 14, 132, 243 self-referentiality of, 140–2 as spatial genre, 151 Dickens, Charles, 130, 254, 299 Household Words, 243 Martin Chuzzlewit, 243 Oliver Twist, 243, 304 Pickwick Papers, The, 242 ‘Sketches by Boz’, 243 Didicher, Nicole, 63 digital mapping. See also GIS (Geographic Information Systems) 3D immersive environment, 277, 303–4 deep mapping, 299–300 distant reading, 297 literary cartography, 25–8, 299–300 mapping onto real-world coordinates, 296–9 narrative cartography, 300–2 Dodge, Martin, 27, 30, 275 Donkin and Fourdrinier, 240 double-facedness of map, 5–13, 93, 109, 121, 138, 162 map copies, 98, 109 map in narrative (internal facing), 41, 80, 125–6

329

readerly map (external facing), 10, 41, 106, 112, 126, 151, 155 Downs, Roger M., 276 Doyle, Arthur Conan, 139 Hound of the Baskervilles, The, 79–80 Study in Scarlet, A, 139 Drabble, Margaret, 18 East India Company, 82 Eco, Umberto, 237–8 eco-criticism, 79 Edney, Matthew, 82–4, 87, 92, 95, 103 Edwards, Jess, 29 eighteenth century, the, 10, 57–8, 67, 74–5, 84, 87, 94 Ekman, Stefan, 201–2, 208, 224 Eliot, George Adam Bede, 249 Daniel Deronda, 273 Middlemarch, 262 Engberg-Pedersen, Anders, 3–4, 74, 76, 247 Enlightenment, the, 29–30, 66, 75 Esrock, Ellen, 290 expressive aesthetics, 10 Ferguson, E. L., 291, 293, 295 fifteenth century, the, 45, 49–50 Flaubert, Gustave, 245–6, 248, 253 Foote, Kenneth, 6, 284, 303 Forster, E. M. Where Angels Fear to Tread, 273 Foucault, Michel, 20, 100 fourteenth century, the, 67 Gaboriau, Émile Monsieur Lecoq, 137–8 Gadamer, Hans-Georg, 277–8 García Márquez, Gabriel Chronicle of a Death Foretold, 303 Garner, Alan Elidor, 197–8 The Weirdstone of Brisingamen, 196–7 Gatrell, Simon, 265, 271 Genette, Gérard, 14, 36–42, 302 Gilmour, Robin, 258 GIS (Geographic Information Systems), 25, 296–9, 302, 305 ArcGIS, 297 Goldman, Paul, 241 Google, 296 Graham’s American Magazine, 129 Graves, Matthew, 7, 119, 121

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

330

Index

Great Trigonometrical Survey of India, 82, 86 Griffith, John, 196 Haggard, H. Rider, 114, 253 King Solomon’s Mines, 14, 83, 96, 101–4, 107–12 Hakluyt, Richard, 58 Hall, Joseph Mundus Alter et Idem, 56 Hammond, Wayne G., 206, 216 Hardy, Thomas, 15, 43, 239, 250, 253, 257–9, 262, 272 Collected Letters, 264–5, 267 Far from the Madding Crowd, 264, 266 Jude the Obscure, 267 Return of the Native, 265, 268–72, 304 Tess of the d’Urbervilles, 267, 271 Wessex maps, 95 Woodlanders, The, 265 Works and Prose (Wessex Edition), 264–5 Harley, J. B, 5, 30–3, 35–6, 92, 94, 100, 104–5 Harmon, Katherine, 2 Harvey, John, 241 Harwood, Jeremy, 46–7, 93 Hegarty, M., 291, 293, 295 Heidegger, Martin, 20, 277 Hewitt, Rachel, 75 history of cartography, 14, 29, 44, See also map production; surveying; triangulation and map projection, 58 imperialism, 81–3 military activity, 76–7 Holbein, Ambrosius, 50 Holland, Clive, 271 Hones, Sheila, 27 Hopkins, Arthur, 268 Hopkins, Martha, 2 Horace, 10, 245, 315 human geometry, 14, 24, 128, 142, 151–2, 255 intersubjective play of people and actions (relative mapping), 153, 161–2 lines of sight, 157–9 surveillance in village space, 152, 154 visual patterns of connection, 155, 158 Hunt, Peter, 167, 169, 200 Hurni, Lorenz, 24–5 Illustrated London News, 127, 130–1, 134 Illustrated Police News, 127, 130, 133–4 imperialism, 96, See also history of cartography and adventure fiction, 92–3 in Edwardian novel, 114–15 and mapping, 95, 103, 107–8

interdisciplinarity, 1–2, 13, 16, 18–21, 25, 27–8, 34, 36, 44 Iser, Wolfgang, 277–9, 284, 286, 288–90 itinerary map, 66–8, 70 Jackson, Peter Lord of the Rings, The, 234–5 Jacob, Christian, 4, 6, 31, 33 Jacobs, Lucia, 287, 301 James, Henry, 242 Golden Bowl, The, 246, 250 Jameson, Fredric, 20, 29, 274–5 Jeffries, Richard Bevis, 189 Jerusalem, 47, 69 Johnson, Barbara, 141 Jonathan Cape, 176 Joyce, James, 24 juxtaposition of map and text, 95 endpaper, 38–9, 175–8, 182 expository relationship, 8–11, 42, 49, 105, 120, 179, 251 map exceeds text, 57, 90, 97–8, 201, 251 paratext, 11, 14–15, 38, 40–3, 183 for readerly visualisation, 295–6 Kain, Roger J. P, 47, 68, 80 Kipling, Rudyard Kim, 83–7 Kitchin, Rob, 20, 27, 30, 275, 287, 303 Knox, Ronald Ten Commandments for Detection, 140, 146 Korzybski, Alfred, 238 Krishnamurth, Sailaja, 86 Krygier, John, 28–9 Lacan, Jacques reading of Poe, 140–3 Lambton, William, 82 Latour, Bruno, 31 Lesnik-Oberstein, Karen, 190–1 Lessing, Gotthold Ephraim, 10–11 Lewes, George, 249 Liddle, Dallas, 128 literary cartography. See also digital mapping literary geography, 16–21, 24–7 literary map and authorial intention, 40 as chronotopic, 47 characteristics of, 8–13 definition of, 6 and double-facedness, 8, 11, 41

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

Index

and genres, 92, 164 and historicity, 35 and sequential narrative, 8, 10, 40, 179 literary mapping, 13, 15, 23, 43, 56, 77, 116, 182, 239, 267–8, 272 non-referentiality, 300 literary place, 26, 28, 58, 203, 205, 235, 254, 263, 268, 272, 274, 292, 296, 300 literary space, 26–8, 40, 58, 91, 205, 235, 263, 268, 272, 274, 292, 296, 300 literary tourism, 17–18, 181, 235, 271 and accuracy, 25, 298 Ljungberg, Christina, 5, 11, 65 Lynch, Kevin, 274–5, 290 Macleod, Jeffrey J., 207 Magnus, Albertus, 282 Maine, Fiona, 185 map and code, 9, 31–2, 36, 93, 123 as icon, 13–14 and symbols, 77 map in detective fiction appears trustworthy/ actively misleads, 146, 149–51 connotative vs denotative meanings, 127, 138, 163 dual intentions of map-maker, 138, 147 manipulation of map, 151, 162 map at multiple scales, 144, 149, 157 map production copperplate engraving, 60, 67, 77 lithography (nineteenth-century), 240 paper-making (nineteenth-century), 239–40 woodcut technique, 49–50, 70, 108 map use geography vs geometry, 23 pictorial, 45–9, 74, 205 processual, 27, 275 referentiality, 6 relative meaning rather than absolute, 46, 75, 158 map-makers, 26, 45, 58, 60, 82, 109 map-making, 24–6, 30, 49–50, 74–5, 77–8, 84, 105, 204–5, 220, 230 mappaemundi, 45–9, 88 Hereford Map, 47 Mapping Emotions in Victorian London, 297 mapping in war danger of accurate mapping, 118–19, 123 doceme, 118 double meanings, 114, 120, 123

331

falsification of geography — two sets of maps, 117, 123 map vs chart, 125 Realist maps of coastline, 119 Mapping the Lakes, 297 Mapping the Republic of Letters, 297 maps and empire. See history of cartography; imperialism double audience of imperialist map, 95 European control, 81, 92, 95, 103, 106 extension of imperialism into fictional mapping, 82, 92 gendered landscape, 107–8 map linked to quest narrative, 93, 101, 103 maps for fantasy both reductive and suggestive, 200 key element of genre, 200–2 primary and secondary worlds, 200–2, 224, 229–30, 233, 235 release readerly imagination, 200 Marsh, Ngaio, 143 Maupassant, Guy de Pierre et Jean, 250 McCarthy, Cormac The Road, 34, 67 McClintock, Anne, 107 McGuire, Caroline, 175 medieval maps, 47, 49, 68, 108 medieval period, 2, 4, 49–50, 66, 87, 282 Mendlesohn, Farah, 201–2 Mercator, Gerard, 60 Mercator Projection, 58, 60 Merleau-Ponty, Maurice, 20 Metropolitan Police Force, 130 Mezei, Kathy, 152–5 Microsoft, 297 Middle Earth, post-authorial mapping cultural power, 233–5 internal consistency, 230, 233 literary cartography, 231 style, 231–3 Middle Earth, published maps Christopher Tolkien as map-maker, 220–1 colour on maps, 208, 224 materiality of maps, 225 style of maps, 223, 225 Miller, J. Hillis, 43, 237, 242, 248, 269, 272–3 Millgate, Michael, 266 Milne, A. A., 182 Winnie-the-Pooh, 47, 169, 192, 253

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

332

Index

mimesis, 10, 13, 245, 249, 252, 304 Moll, Herman, 60–2 New and Correct Map of the Whole World, 64 Monmonier, Mark, 8–9, 30, 38, 117, 123 Montgomerie, Thomas, 84 Moody, Ellen, 255 Moore, John R., 62 More, Thomas Utopia, 11, 50–7, 63–4, 118 Moretti, Franco, 2, 18, 21–7, 33, 67, 254–5 Morris, Pam, 248–9 Muehrcke, P.C and J.O, 23, 30 Nadel, Lynn, 285–7, 289 Napoleonic Wars, 74, 247 National Geographic, 94 nautical chart, 96, 98, 118–19, 122, 183 Nesbit, Edith, 189 nineteenth century, 44, 57, 67, 74, 81, 91, 117, 128, 239–40, 244, 248 late, 2, 13–14, 92–3, 152 mid, 14, 77, 129, 164 O’Keefe, John, 285–7, 289 Ogilby, John Britannia, 70 Once a Week, 135 OpenStreetMap, 297 Ordnance Survey, 4, 70, 75–7, 79, 82, 125, 231, 240, 303 Padrón, Ricardo, 251 paratext, 7–8, See also Genette, Gérard; juxtaposition of map and text performative nature of, 41 two-faced nature of, 41 Paris, Matthew Historia Anglorum, 68–70 Paris Police, 130 Pavlik, Anthony, 200, 251–2 Peraldo, Emanuelle, 3 Perkins, Chris, 20, 27, 30, 275 perspective, 49, 107, 134, 155, 176, 178, 207, 224–5, 231, 233, 269, 303 in act of reading, 278, 284 phenomenology, 79, 273, 286, 288 Philips, Richard, 90 Piatti, Barbara, 3, 24–5, 298, 300 pilgrimage, 67, 85–6 literary, 17 Piper, Leonard, 117

Pite, Ralph, 262–3, 266–7 playspace multi-temporal and inter-generational, 185–6 necessary distance of adult, 168, 190 nostalgic landscape (present of child is past of adult), 168, 183 positive adult–child dynamic, 168 positive alternative (to deconstructive model), 164, 166 shared space, 165 playspace (negative model) Alan Garner hidden threat from adults, 197 inversion of safe model, 197 map participates in, 197 Peter Pan no boundaries, 195 rejection of adults, 193–4 threat to adults, 193 Pocock, Douglas, 19 Poe, Edgar Allan Murders in the Rue Morgue, The, 129, 139 Purloined Letter, The, 141 portolan charts, 50, 88, 108 Post, J. B, 2 post-authorial map, 6–7, 67, 203, 229 post-authorial mapping, 14, 228–35 postcolonialism, 82, 106 Priestnall, Gary, 299 Propp, Vladimir Morophology of the Folktale, 254 psychogeography, 79 Purchas, Samuel, 58–9 Pyrhönen, Heta, 139, 143, 146 Quintilian, 280 Institutio Oratoria, 280 Ransome, Arthur, 95, 189, 197, 203, 266 Lake District Series, 14, 165, 169 Picts and The Martyrs, The, 170–3 Pigeon Post, 170–1, 174, 180, 182 Swallowdale, 170, 172–4, 176–85 Swallows and Amazons, 170, 173, 176–80, 183–6 Winter Holiday, 170, 174, 178–81, 183–4 Ransome’s endpaper maps, 179–81 child as map-maker, 191 narrative content, 175 perspective, 176 use of colour, 178

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

Index

Ransome’s endpapers child as map-maker, 183 Ransome’s playspace, 170 adult-free play, 170–1 arc-narrative, 170, 184 boundaries of playspace, 171, 173 imaginative vs realist geographies, 183–4 Lake District mosaic, 182 real danger, 169, 174 re-mapping, 183–5 substitute parent (older child), 173 readerly visualisation ideation, 278–9 inner horizon, 277–9, 284, 286, 288 for literature (semi-schematic form), 292–6, 306 maps vs tour, 284–5 memory mapping memory palace, 280–3 mnemonic loci, 280 reader-response theory, 273, 276–9, 283–4 reading and mapping, 231 dynamic model of, 4, 9, 13, 22, 278, 283, 288, 303, 305 reading-as-mapping, 273, 276, 278–9, 283–4, 289 performative nature of, 5 processual, 36 real-world geography, 1, 47, 55, 267 Redish, David, 285–6, 289 referential relations accuracy between real and fictional, 185, 248, 253, 263–4, 267, 272 direct correspondence, 25, 67, 237–8, 248, 298 map defies its own fictionality, 253 map self-replicates, 238, 252 map vs territory, 238–9, 252 non-referentiality, 43, 72 representational priority, 43, 248, 253, 293 Relph, Edward, 19 Rhetorica ad Herennium, 280–1 Richardson, Alan, 189 Robinson, Arthur, 30, 35, 109, 183 Romantic period, 10, 77 Rose, Jacqueline, 165–8, 189–90 Rossetto, Tania, 26–7, 34 Rousseau, Jean-Jacques, 164 and education, 78, 167 false model of innocent child, 166

333

sceptical reading against, 188, 190–1 well-regulated freedom, 186–8 route survey, 84 Roy, William, 75–6 Royal Geographical Society, 82, 84 Russell, William Recollections of a Detective Police Officer, 130 Ryan, Marie-Laure, 6, 8, 275, 284, 300–3 Sadleir, Michael, 257, 260–1 Said, Edward, 83, 86 Culture and Imperialism, 82 Sant, Van, 94 Saunders, Angharad, 26 Sayers, Dorothy L., 127, 143 Schenk, Françoise, 287, 301 Scott, Walter, 255 Minstrelsy of the Scottish Border, 78 Scull, Christina, 206, 216 Seed, David, 114 semiotics, 5, 12, 31 and code, 11 and icon, 12 serial form, 80, 135, 239, 241–3, 268, 293, See also Dickens, Charles seventeenth century, the, 67, 75, 90 Sharp, Joanne, 19 Sharp, William, 17 Shepard, E. H., 47, 192 Shields, Rob, 20 Shpayer-Makov, Haia, 128 silence on the map material object of the map, 105 two-faced, 105, 108 unconsciously speaks, 107–8 Sipe, Lawrence, 175 Six Degrees of Francis Bacon, 297 sixteenth century, the, 50, 54, 58, 67, 70 sketch map, 77, 110, 118, 121, 220, 242, 260, 268, 271, 287 cognitive, 301, 303 Smith, David, 240 Smol, Anna, 207 Snell, K. D., 152, 155, 158 spatial humanities, 297, 299 Spurrier, Stephen maps for Swallows and Amazons, 176–9, 183 spy fiction, 94, 114, 164 Stafford, David, 116 Stanford LitLab, 3 Stea, David, 276

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

334

Index

Stendhal (Marie-Henri Beyle) Vie de Henry Brulard, 246–8 Stevenson, Robert Louis, 182, 203, 205 Letters, 112–13 map and authorial myth-making, 112–13 map in creative process, 99–100 ‘My First Book’, 99–100, 112, 204 Treasure Island, 10, 14, 41, 43, 87–91, 96–9, 101–6, 184, 304 Stiebel, Lindy, 107 Stockhammer, Robert, 7 Stoker, Bram Dracula, 79 Stott, Rebecca, 107–8 Strachey, Barbara, 231–3 Strand Magazine, The, 80 strip map, 67, 70, 72–3, 215, 217 Sundmark, Bjorn, 200 surveying, 70, 75–6, 85 Swift, Jonathan, 56–7 Gulliver’s Travels, 56–7, 59–65, 77, 229 Symons, Julian, 129

Barsetshire, 95 Chronicles of Barsetshire, 255, 261 Dr Thorne, 256, 258–9 Framley Parsonage, 255–6, 259–61 Warden, The, 256 truth and accuracy, 246–8 and authorial mapping, 215 and cartographic map, 30 and detective fiction, 127, 141, 143, 162 and fictional map, 14, 95–6, 127, 145 and literary map, 125 and reading, 278 Tuan, Yi-Fu, 18–20 Twain, Mark Huckleberry Finn, 293 twelfth century, the, 44–5 twentieth century, the, 14, 18, 20, 30, 44, 47, 62, 90, 92, 140, 143, 152, 202 twenty-first century, the, 1–2, 20, 26, 30, 67, 235, 296 ut pictora poesis, 244

Tally, Robert T., 3, 26, 29 Thacker, Andrew, 33–4 Thackeray, William Makepeace, 255 thematic mapping, 21, 23, 93, 232 thirteenth century, the, 281 Thrift, Nigel, 20 Times Literary Supplement, 120 Timpf, Sabine, 205, 207 Todorov, Tzvetan, 142–3, 145, 147 Tolkien, Christopher, 203, 205, 208, 224 History of Middle Earth, The, 206, 212–13, 217, 221 Tolkien, J. R. R., 14, 90, 95, 253, See also cartographic creativity; Middle Earth Hobbit, The, 43, 205, 207–9, 213–16 Letters, 204–5 Lord of the Rings, The, 43, 201, 204–7, 220–5, 227–33 Silmarillion, The, 230, 232 Tree and Leaf, 206, 212, 229 Tooke, Adrianne, 245–6 travel narratives, 59–60 travel writing, 1, 298 triangulation, 70, 75–6, 82, 84 Trignometrical Survey (1791), 75–6 Trollope, Anthony, 15, 239, 241, 253, 262, 266, 272 Autobiography, 257–9 Barchester Towers, 239, 256–8

Van Dine, S. S Twenty Rules for Writing Dectective Stories, 140 Verne, Jules, 182 Vespucci, Amerigo, 55 Victorian period, 77, 93, 101, 110, 123, 239, 264 Vidocq, Francois, 129–30 Mémoires de Vidocq, 130 voyage narrative, 58–9, 64 Walker, R. E., 201 Wall, Barbara double addressee, 168 Waller, Alison, 185 Wallis, Helen, 109 Watson, Nicola, 18, 185, 266, 271 Wegner, Phillip E., 83 Westminster Review, 249 Wilson, Angus, 292 Wild Garden, The, 293–5 Wilson, Edmund ‘Who Cares Who Killed Roger Ackroyd?’, 140 Wood, Denis, 32, 94 Woodward, David, 5, 45–6, 49 Woolf, Virginia, 16–17, 24 Wordsworth, William, 78, 167, 299

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

Index

writerly imagination vs readerly active readerly imagination, 238–51 evocation of word, 246, 248 map needs to be interpreted, 238–51

335

Wyatt, John, 4 Wynn Fonstad, Karen, 203, 230, 232–3 Wynne Jones, Diana, 199 Yates, Frances, 279–81

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011

Downloaded from https://www.cambridge.org/core. University of Virginia Libraries, on 21 Jul 2021 at 00:44:13, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108766876.011