Reading Modernism with Machines Digital Humanities and Modernist Literature-Palgrave Macmillan

Home , High modernism, Kenneth Fearing

READING MODERNISM WITH MACHINES DIGITAL HUMANITIES AND MODERNIST LITERATURE Edited by SHAWNA ROSS and JAMES O’SULLIVAN Reading Modernism with Machines Shawna Ross • James O’Sullivan Editors Reading Modernism with Machines

Digital Humanities and Modernist Literature Editors Shawna Ross James O’Sullivan Department of English Humanities Research Institute Texas A&M University University of Sheffield Department of English Sheffield, UK College Station, Texas, USA

ISBN 978-1-137-59568-3 ISBN 978-1-137-59569-0 (eBook) DOI 10.1057/978-1-137-59569-0

Library of Congress Control Number: 2016955951

© The Editor(s) (if applicable) and The Author(s) 2016 The author(s) has/have asserted their right(s) to be identified as the author(s) of this work in accordance with the Copyright, Designs and Patents Act 1988. This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.

Cover illustration: © Cultura Creative (RF)/Alamy Stock Photo Cover design by Oscar Spigolon

Printed on acid-free paper

This Palgrave Macmillan imprint is published by Springer Nature The registered company is Macmillan Publishers Ltd. The registered company address is: The Campus, 4 Crinan Street, London, N1 9XW, United Kingdom Preface

This volume is comprised of essays that are, methodologically, rooted in the digital, but the significance of this collection is to be found in the results of the studies—the literary interpretations that are supported, not dictated, by the machine. In taking such an approach, it is our hope that this collection represents precisely what digital literary studies should be: a field in which computers assist in the discovery of new forms of evidence, evidence which is in turn used to further existing critical arguments, while shining a light on new, previously unforeseen strands of enquiry worth pursuing. The method is just that—the method—and while we must strive to ensure that the techniques of this field remain valid, it is what we derive from the method, rather than the method itself, that should be, as literary and cultural scholars, our primary focus. A fascination with method is important to the development of more robust and sophisticated techniques, but any such research should always be conducted in the service of those disciplines and activities that constitute the Arts and Humanities. Saying this, the value of method should not be diminished, and thus, this collection also serves as a timely demonstration for those scholars who wish to see the validity of the Digital Humanities. These essays should act as a template for those seeking to juxtapose computer-assisted techniques with critical enquiry, particularly in a field such as this, where modernism’s central tenant, the desire to “make it new,” seems as readily applicable to the scholar’s method as it does the artifact’s content. While the Digital Humanities are comprised of various and sometimes dissonant activities, the methods that have emerged from this community of praxis are applicable to a multiplicity of literatures. The analysis of most,

v vi Preface if not all, literary movements, epochs, genres and styles can be assisted by a computer. Yet, while these techniques are the progeny of interdisciplinary efforts, and entirely transferable in their application, we should not lose sight of our own humanistic origins. Many of the Digital Humanities’ existing collections are broad in their focus, covering a variety of disciplines. Undoubtedly a product of the field being inherently interdisciplinary and collaborative, while this trend is to be encouraged, there is also a need for disciplinary focus. This collection offers an example of such: while there may be some appeal to a wider set of literary scholars intrigued by recent shifts in the way that scholarship is conducted, this is a collection about modernist literature, comprised of contributions by scholars who are humanists first, tech- nicians second. In being so, it is an example of precisely what the Digital Humanities promises: a robust interrogation of the literary, informed by methods which do not replace, but rather, supplement, existing modes of criticism. And in doing so, it does not render the long-established principles of modernist scholarship obsolete—it merely contributes to making them new.

James O’Sullivan Acknowledgements

The editors would like to acknowledge the pioneering work done by the Modernist Journals Project (Robert Scholes, Sean Latham, Susan Smulyan, Jeff Drouin, Clifford Wulfman and Mark Gaipta), the Orlando Project (particularly Susan Brown), the Modernist Versions Project (particularly Stephen Ross), and Editing Modernism in Canada (Dean Irvine). We would also like to thank April James, Ben Doyle, and Peter Cary at Palgrave Macmillan for their support of this collection, as well as our tire- less copyeditors. Portions of Chapter 1 have appeared in Digital studies/Le champ numérique, vol. 6 (2016). Shawna would like to thank Andrew for his attentive willingness to talk in perhaps excruciating detail about this project. She wants to dedicate her work on this book to her mother, Cynthia Jordan. James would like to thank Graham Allen and Órla Murphy, University College Cork, for their continued support and guidance.

vii Contents

1 Introduction 1 Shawna Ross

2 ModLabs 15 Dean Irvine

3 Modeling Modernist Dialogism: Close Reading with Big Data 49 Adam Hammond, Julian Brooke, and Graeme Hirst

4 Mapping Modernism’s Z-axis: A Model for Spatial Analysis in Modernist Studies 79 Alex Christie and Katie Tanigawa

5 Textbase as Machine: Graphing Feminism and Modernism with OrlandoVision 109 Kathryn Holland and Jana Smith Elford

6 Remediation and the Development of Modernist Forms in The Western Home Monthly 135 Hannah McGregor and Nicholas van Orden

ix x Contents

7 Stylistic Perspective Across Kenneth Fearing’s Poetry: A Statistical Analysis 165 Wayne E. Arnold

8 In the End Was the Word: A Computational Approach to T. S. Eliot’s Poetic Diction 185 Adam James Bradley

9 A Macro-Etymological Analysis of James Joyce’s A Portrait of the Artist as a Young Man 203 Jonathan Reeve

10 Body Language: Toward an Affective Formalism of Ulysses 223 Kurt Cavender, Jamey E. Graham, Robert P. Fox Jr., Richard Flynn, and Kenyon Cavender

11 “We Twiddle … and Turn into Machines”: Mina Loy, HTML and the Machining of Information 243 Andrew Pilsch

12 CGI Monstrosities: Modernist Surfaces, the Composite and the Making of the Human Form 265 Eunsong Kim

Index 291 Notes on the Contributors

Wayne E. Arnold holds a PhD in English (2013) from The University of Louisiana at Lafayette, USA, and an MA in TESOL (2013) from the same university. Additionally, he has earned an MA in English (2007) from Western Kentucky University and an MBA (2001) from Wright State University. Prior to accepting a position as Associate Professor of American Studies at The University of Kitakyushu, Japan, he taught at Kansai Gaidai University, Japan, and Harvard University Summer Program. Research interests include Henry Miller and Kenneth Fearing. Adam James Bradley BA (McMaster) MA (Waterloo), is a PhD candidate in both the departments of English Language and Literature and Systems Design Engineering at the University of Waterloo. He is interested in the intersections between technology and traditional literary studies with a focus on early twentieth- century poetics. His current work focuses on digital tool design for literary criticism and investigations into how philology can still function within a technological context. Other interests include modernist literature, classical languages and ancient rhetoric. Julian Brooke is a McKenzie Postdoctoral Fellow in the Computing and Information System Departments at the University of Melbourne, Australia. The topic of his PhD thesis was computational analysis of lexical style. His published work in computational linguistics includes papers at major conferences in the field such as ACL and COLING as well as an article in the flagship journal,Computational Linguistics. He is co-creator, with Adam Hammond, of two websites for exploring modernist dialogism: He Do the Police in Different Voices (hedothepolice.org) and The Brown Stocking (brownstocking.org). He is co-developer, with Adam Hammond, of GutenTag (projectgutentag.org), a tool for computational text analysis in the Project Gutenberg corpus.

xi xii Notes on the Contributors

Kenyon Cavender is a freelance programmer with a BS in Mathematics from Texas A&M University, USA. He is interested in the application of free and open source software in both academia and the private sector. Kurt Cavender is the Andrew Grossbardt Graduate Fellow in the Department of English at Brandeis University, USA. His work focuses on the historical novel in twentieth and twenty-first century American literature. Alex Christie is Assistant Professor in Digital Prototyping at Brock University’s Centre for Digital Humanities, Canada. He completed his doctorate at the University of Victoria, where he conducted research on geospatial expression and scholarly communication for the Modernist Versions Project (MVP) and Implementing New Knowledge Environments (INKE) in the Electronic Textual Cultures Lab (ETCL) and the Maker Lab. He developed an open source toolkit for digital humanities pedagogy with grant funding from the Association for Computers and the Humanities. He is currently working on a modernist history of the mechanical production and interpretation of texts before the advent of digital computing. Jana Smith Elford is a SSHRC-funded doctoral candidate in the department of English and Film Studies at the University of Alberta, Canada, where she is also a Research Associate with the Orlando Project and the Canadian Writing Research Collaboratory (CWRC). Her research explores the lives and writings of several fin de siècle feminist reformers using network visualization software. With her collaborators, she helped develop the OrlandoVision prototype, conducting user testing, making recommendations for changes and drafting documentation. She is currently involved with the development of HuVis, an RDF visualization tool. Her research appears in Orlando: Women’s Writing in the British Isles from Beginnings to the Present (2006), the Victorians Institute Journal Annex, Victorian Review, and the Journal of Modern Periodicals Studies. Richard Flynn is a PhD candidate in English Literature at Brandeis University, USA. Robert P. Fox Jr. is a PhD candidate in English and American Literature at Tufts University, USA, who focuses on Renaissance literature with a particular interest in the intersections of literary and legal culture in Early Modern England. He is a graduate of Boston College Law School and Harvard College and was a partner at the Boston law firm of Nutter, McClennen & Fish, LLP specializing in real estate. Jamey E. Graham teaches Renaissance British Literature at Le Moyne College, USA. Previously, she taught History and Literature at Harvard University, where she earned her PhD in Comparative Literature. The author of articles on Shakespeare and Spenser, she is currently working on a book titled How Character Became Literary: Virtue and Example in Early Modern Poetics. Notes on the Contributors xiii

Adam Hammond is an Assistant Professor in the Department of English and Comparative Literature at San Diego State University, USA. He is the author of Literature in the Digital Age: A Critical Introduction (2016) and co-author, with Melba Cuddy-Keane and Alexandra Peat, of Modernism: Keywords (2014). He is co-creator, with Julian Brooke, of two websites for exploring modernist dialogism: He Do the Police in Different Voices (hedothepolice.org) and The Brown Stocking (brownstocking.org). He is co-developer, with Julian Brooke, of GutenTag (projectgutentag.org), a tool for computational text analysis in the Project Gutenberg corpus. Graeme Hirst is Professor in the Department of Computer Science at the University of Toronto, Canada. His research interests cover a range of topics in applied computational linguistics and natural language processing, including lexical semantics, the resolution of ambiguity in text, the analysis of authors’ styles in literature and other text, and the automatic analysis of arguments and discourse (especially in political and parliamentary texts). Hirst’s present research includes determining ideology in political texts; detecting markers of Alzheimer’s disease in language; and the identification of the native language of a second-language writer of English. Hirst is the editor of the Synthesis series of books on Human Language Technologies. He is the author of two monographs: Anaphora in Natural Language Understanding (1981) and Semantic Interpretation and the Resolution of Ambiguity (1992). Kathryn Holland is a Senior Research Associate with the Orlando Project and an instructor at MacEwan University, Canada. Her research is situated at the intersection of modernist literary history, feminist studies and digital humanities, with a focus on the place of the multigenerational family in modernist networks and syn- chronic approaches to literary history. Her current projects include the essay collection she is co-editing, Digital Diversity: Writing | Feminism | Culture. Her writing is published in Tulsa Studies in Women’s Literature, Modernism/modernity, Canadian Journal for the Scholarship of Teaching and Learning, and the Times Literary Supplement. She earned her doctorate in English literature as a Clarendon Scholar and SSHRC Doctoral Scholar at the University of Oxford. Dean Irvine is an associate professor at Dalhousie University, Canada. He is director of Editing Modernism in Canada and the open-source software and web- design company, Agile Humanities Agency. His publications include Editing Modernity: Women and Little Magazine Cultures in Canada, 1916–1956 (2008) as well as the edited collections The Canadian Modernist Meet (2005), Editing as Cultural Practice in Canada (2016), co-edited with Smaro Kamboureli, Making Canada New: Editing, Modernism, and New Media (2016), co-edited with Vanessa Lent and Bart Vautour, and Translocated Modernisms: Paris and Other Lost Generations (2016), co-edited with Emily Ballantyne and Marta Dvorak. He is the director and general editor of the Canadian Literature Collection published by the University of Ottawa Press. xiv Notes on the Contributors

Eunsong Kim is a doctoral candidate at the University of California, San Diego, USA. She works with local and national youth arts organizations like Urban Gateways to develop and teach critically based digital film programs. She has created video content for the Getty Center, the Culture Art and Technology Program at UCSD and the European Independent Film Festival. Her essays on literature, digital cultures and art criticism have appeared and are forthcoming in Scapegoat, Lateral, The New Inquiry, Model View Culture, The Margins, and in the forthcoming book anthologies, Global Poetics, Critical Archival Studies, and Forms of Education. Her prose has been published in Denver Quarterly, Seattle Review, Feral Feminisms, Minnesota Review, Iowa Review and Action Yes. She is the co- founder of contemptorary, an online arts platform dedicated to women of color artist and writers. Hannah McGregor is a researcher and full-time instructor in the Department of English and Film Studies at the University of Alberta, Canada. Her recently completed SSHRC postdoctoral fellowship, Modern Magazines Project Canada, is a collaborative initiative that takes up the call to read magazines as a form of new media technology that, alongside radio and film, helped to shape the emergent consumer-publics of the twentieth century. In collaboration with the University of Alberta Libraries and the Manitoba Legislative Library, McGregor has helped to facilitate the digitization of the full run of the Winnipeg-based magazine The Western Home Monthly (1899–1932). Her research takes advantage of this digitization to explore digital methods for the study of periodicals including topic modeling with MALLET, visualization with R, and interactive timelines. She has published on this research in English Studies in Canada, the Journal of Modern Periodical Studies, Archives and Manuscripts, the International Journal of Canadian Studies, and in the edited collection Editing as Cultural Practice in Canada (2015). Nicholas van Orden is a PhD student in the Department of English and Film Studies at the University of Alberta, Canada. His research focuses on the collision of digital spaces and fictional forms. He is interested in a range of digital humanities topics and methodologies and works to build DH projects into his undergraduate classes. James O’Sullivan is Digital Humanities Research Associate at the University of Sheffield’s Humanities Research Institute, United Kingdom. His research primarily focuses on electronic literature, though he is also concerned with computational approaches to criticism. His work has been published or is forthcoming in a variety of interdisciplinary journals, including Digital Scholarship in the Humanities, English Studies, Digital Humanities Quarterly, Leonardo, and the International Journal of Humanities and Arts Computing. His research was shortlisted for the Fortier Prize in 2014. James is Chair of the DHSI Colloquium at the University of Notes on the Contributors xv

Victoria, and a member of the Association for Computers and the Humanities’ Standing Committee on Affiliates. James is also a published poet, and the founder of New Binary Press. Further information on James and his work can be found at josullivan.org. Andrew Pilsch is Assistant Professor of English at Texas A&M University, USA. He researches and teaches rhetoric and digital humanities, with specific focus on post-digital ideas of embodiment, online utopianism and forms of digital rhetorical engagement. His book on transhumanism and contemporary notions of utopia, including additional material on Mina Loy’s digital afterlives, is currently under contract with The University of Minnesota Press. He tweets online at @ oncomouse. Jonathan Reeve is a graduate student in the Department of English and Comparative Literature at Columbia University, USA, where he works in computational literary analysis. He has worked as a web developer for the Modern Language Association, New York University, and the Greenwich Village Society for Historic Preservation. His current projects include the Macro-Etymological Analyzer; Annotags, a protocol for decentralized textual annotation; and Git-Lit, an initiative to version-control and publish electronic texts from the British Library. Find his blog at http://jonreeve.com. Shawna Ross is Assistant Professor of Modern British Literature and the Digital Humanities at Texas A&M University, USA. She is currently working on a book manuscript that argues that modernist literature theorized relations of leisure and labor, participating in the production of a comprehensive public discourse of leisure that challenged the Victorian work ethic and recognized the role of leisure in transnational economies and politics. Readings of Charles Dickens, G. K. Chesterton, Henry James, T. S. Eliot, E. M. Forster, Katherine Mansfield, Vita Sackville-West and others are juxtaposed by archive-based studies in the visual history of the leisure industry. She frequently writes on Henry James and on the digital humanities, and her work has been published in The Henry James Review, the Journal of Modern Literature, Digital Humanities Quarterly, and various edited collections, including Henry James Today (2014), Literary Cartographies (2014), and Utopianism, Modernism and Literature in the Twentieth Century (2013). Katie Tanigawa is a PhD candidate in English at the University of Victoria, Canada. She works on geospatial analyses of modernist texts for the Modernist Versions Project (MVP) and Implementing New Knowledge Environments (INKE), and she is the Project Manager for the Map of Early Modern London. Her past work includes marking up, versioning and visualizing textual differences in extant versions of Joseph Conrad’s Nostromo. Her current areas of research include representations of poverty in Irish modernist literature and exploring modernist approaches to digital humanities. List of Figures

Fig. 3.1 Stylistic change curve over the abridged version of The Waste Land 58 Fig. 4.1 Z-axis map of Djuna Barnes’s Nightwood 81 Fig. 4.2 Paris Monumental et Métropolitain map (1932) 82 Fig. 4.3 Detail from map of Paris: Saint-Sulpice 83 Fig. 4.4 XML markup of text and map-based locations 86 Fig. 4.5 XSTL-transformed geo-data of Paris 86 Fig. 4.6 Displacement map of Paris 88 Fig. 4.7 3D map mesh of Paris 89 Fig. 4.8 Warped 3D map mesh of Paris 89 Fig. 4.9 Z-axis map of Djuna Barnes’s Nightwood 92 Fig. 4.10 The doctor’s city in Nightwood 94 Fig. 4.11 Barnes’s version of the carriage ride in Nightwood 96 Fig. 4.12 Rhys’s representation of the Observatoire in Quartet 98 Fig. 4.13 Map of Marya’s narrative engagement with Paris in Quartet 100 Fig. 4.14 Quartet map showing place Denfert-Rochereau 101 Fig. 4.15 Quartet map showing the Palais de Justice and Santé Prison 102 Fig. 5.1 Flyer for the United Procession of Women suffrage march, London, 9 February 1907 111 Fig. 5.2 Links screen for Vernon Lee entry in the Orlando Project 114 Fig. 5.3 Life document type definition (subtags not visible) 114 Fig. 5.4 Selection tools on Right, text box at bottom, and visualization pane in top left 115 Fig. 5.5 Initial network graph of the NUWSS 119 Fig. 5.6 Graph of NUWSS focused on Marsden 121 Fig. 5.7 Cluster 1 in OViz graph of Newnham College 123

xvii xviii List of Figures

Fig. 5.8 Cluster 2 in OViz graph of Newnham College, with view of Woolf, Harrison and Mirrlees nodes 124 Fig. 6.1 Advertisements, illustrations and pages in each year of the Western Home Monthly 144 Fig. 6.2 “A chat with our readers” in the WHM, July 1919 149 Fig. 6.3 The fall of the phonograph and the rise of radio in WHM advertisements 150 Fig. 6.4 Graph demonstrating sudden increase in fragmentation in the WHM in 1919 151 Fig. 6.5 A collage of text leaves more space for advertising in October 1919 issue of the WHM 152 Fig. 6.6 Advertisements, illustrations, pages and words (/1000) in the WHM 154 Fig. 6.7 Scatterplot showing distribution of new media topics in the WHM 157 Fig. 6.8 Two-page full-color advertisement for Fada Radios in the WHM, October 1925 159 Fig. 8.1 Percentage of total words used by century of first usage by Eliot and the Georgians 193 Fig. 8.2 Georgian poets’ versus T. S. Eliot’s Z-scores for total usage by century 194 Fig. 8.3 Comparison with Victorian Z-scores for total usage by century 195 Fig. 8.4 Comparison with Brown News Corpus Z-scores 195 Fig. 8.5 Full table of Z-scores for total usage by century 197 Fig. 8.6 Z-score vocabulary across centuries 199 Fig. 9.1 Latinate words in the Brown News Corpus 207 Fig. 9.2 Hellenic words in the Brown News Corpus 208 Fig. 9.3 L scores for chapters of Joyce’s Portrait of the Artist as a Young Man 209 Fig. 9.4 L scores for sections of Joyce’s Portrait of the Artist as a Young Man 210 Fig. 9.5 H scores for sections of Joyce’s Portrait of the Artist as a Young Man 214 Fig. 10.1 Affective density by episode in Ulysses 232 Fig. 11.1 Comparison of printed and HTML block stanzas for Songs to Joannes 258 Fig. 12.1 Pierre-Louis Pierson, “La Frayeur” (1861–67) 271 Fig. 12.2 CGI screenshot of John Adams showing initial shot of primary actors 276 List of Figures xix

Fig. 12.3 The compositing process of CGI animation 277 Fig. 12.4 Constructing extras in John Adams 278 Fig. 12.5 Crowd scene of composited extras in John Adams 279 Fig. 12.6 Idealized renderings of political figures inJohn Adams 279 Fig. 12.7 Interpolating human actors in a CGI crowd scene in John Adams 280 Fig. 12.8 A final composited scene inJohn Adams 280 List of Tables

Table 3.1 Stylistic profiles for various characters in The Waste Land 62 Table 3.2 Stylistic profiles for discourse types inTo the Lighthouse and “The Dead” 68 Table 3.3 Stylistic profiles for characters in To the Lighthouse and “The Dead” 70 Table 3.4 Stylistic profiles for various social groups in To the Lighthouse 71 Table 7.1 Means of first-person singular and first-person plural across Kenneth Fearing’s poetry 171 Table 7.2 Distribution of “you” and “your” across Fearing’s poetry 171 Table 7.3 Means of third-person pronouns across Fearing’s poetry 173 Table 7.4 Consistency percentages across Fearing’s poetry 174 Table 7.5 Type/token percentage across Fearing’s poetry 175 Table 7.6 N-gram distribution across Fearing’s poetry 176 Table 7.7 Mean sentence length across Fearing’s poetry 177 Table 7.8 Punctuation use across Fearing’s poetry 177 Table 10.1 Sentence length by episode in Ulysses 231

xxi CHAPTER 1

Introduction

Shawna Ross

In his 1922 constructivist manifesto, László Moholy-Nagy proclaims that technology is the “reality of this century.” In defi ning technology as “the invention, construction and maintenance of the machine” and proclaim- ing that to be “a user of machines is to be of the spirit of his century,” Moholy-Nagy characterizes avant-garde modernism in terms uncannily similar to our own twenty-fi rst century justifi cations of the timeliness and signifi cance of digital humanities. 1 After all, what unites the disparate projects of digital pedagogy, data collation and visualization, archive and tool construction, cultural studies of media and computation, and all of the other disparate projects under the umbrella term digital humanities (DH), is the common denominator of the machine known as the computer. If we accept Moholy-Nagy’s claim that machines create reality—“reality” being, according to his defi nition, that which “determines what we can grasp and what we cannot understand” (299)—then our computational machines constitute the foundation of our episteme as they did in Moholy-Nagy’s modernist episteme. In this episteme, technology determines what emerges as real, as palpable, as capable of producing truths, and what recedes, unreal, ungraspable and unrepresented. Digital humanists doing

S. Ross ( ) Department of English, Texas A&M University, College Station, USA

© The Editor(s) (if applicable) and The Author(s) 2016 1 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_1 2 S. ROSS research in modernism are thus truly reading modernism with machines: more than simply means to an end, our machines underwrite the reality of our scholarship. Their processes and outputs infl uence what emerges as knowable and what counts as proof, bending modernist texts and modernism itself toward our contemporary machinic episteme. Of course, it is no more willfully anachronistic than any school of literary criticism—so long as we do not silently attribute to modernism itself our own contemporary revolutionary digital rhetorics of the new. This is the danger of our work, this all-too-satisfying equation of modernists’ fascinated celebrations and critiques of their technological present with our own . The best work in digital humanities addresses this spurious resemblance by refl ecting both on the conditions of its own production and on the world-making (and world-limiting) powers of our machines. Consider, for example, the refl exivity of certain DH schools of thought (such as critical code studies and minimal computing) and certain DH thinkers (particularly Johanna Drucker’s critique of techno-positivism and Alan Liu’s mandate that digital humanists engage in cultural criticism). 2 With modernist studies’ similar emphasis on the refl exivity, poetics and cultural criticism, it could still seem possible to stretch the modernist episteme of the machine comfortably over ourselves, making, temporally, a very big tent indeed. It would undoubtedly be a tent particularly fl attering to modernists. But Moholy-Nagy’s twentieth-century optimistic fervor cannot quite be our own, though his insight into the constitution of an episteme by the machine can be—with the addendum that a succession of technological presents has superseded the putatively universal machinic episteme posited by the modernist avant garde. We cannot claim, as he does, “There is no tradition in technology, no consciousness of class or standing. Everybody can be the machine’s master or its slave” (300). We can, now, point to such a tradition, the elaboration of which demonstrates the distance between our particular machinic episteme and the modernist machinic episteme. More alarmingly, Moholy-Nagy’s uncritical invocation of the master/slave relationship, which he seems to applaud for its apparently utopian potential, could itself be interrogated as a troubling, dystopian reinscription of imperial forms of cultural and economic production. The critical work of postcolonial and feminist digital humanities—critical in every sense of the term, as Jesse Stommel writes of critical digital pedagogy 3 —has been engaging in such interrogations. When these interrogations are done in the fi eld of modernism, they certainly point to parallels between modernism and the Digital Humanities, but they are parallels not at all desirable. INTRODUCTION 3

A CALL FOR FIELD-SPECIFIC DIGITAL HUMANITIES This principle applies to other DH subfi elds as well. Any proposed parallel between any literary period and the contemporary digital culture and scholarship needs to be specifi ed (is this stylistic, political, material, rhetorical, methodological or philosophical?), qualifi ed (acknowledging gaps, errors or hyperbole), and historicized (accounting for both material and discursive change). This is not to deny, outright, any contemporary recur- rences of certain avant-garde rhetorics and representational strategies. Such arguments have been persuasively cast at least since 1989 with the publication of Richard Lanham’s “The Electronic Word,” which posits that modernism and humanities computing share the avant-garde rhetoric he calls “experimental humanism.” 4 More recently, Jessica Pressman’s Digital Modernism observes recurrence of modernist representational strategies in contemporary electronic literature, and Stephen Ross and Jentery Sayers trace “productive convergences” between modernism and the Digital Humanities (625). 5 My own work has participated in the digital revival of modernist avant-garde forms, 6 although I have also argued that this neomodernism results from the shared methodology a modernist scholar brings to bear on modernism and on the Digital Humanities rather than from inherent resemblances between digital humanities methods and modernist literature. 7 By dwelling on this impulse to draw parallels, I mean not to reject these arguments wholesale but to affi rm that they are only successful when they are informed by and grounded in fi eld-specifi c literary criticism. This is true even when those arguments reject or revise the assumptions common to that critical fi eld. It is to affi rm that within disciplinary traditions lay a rich, vetted and productive method of avoiding the continually renewed claims that DH is undertheorized. 8 Certainly, other ways of debunking those claims exist: Natalia Cecire points to the greater signifi cance of mea- suring the ethical implications of choosing certain models for DH, and Steven E. Jones notes that “those outside DH often underestimate the theoretical sophistication of many in computing” (10). 9 Jean Bauer argues that work already undertaken is inherently theoretical, and Chris Forster, in responding to Bauer, deconstructs the “slippery grammatical place” of the term theory to reveal the fundamental incoherence of demands to be theoretical. 10 All of these defenses provide a solid defense against these claims, but our disciplines offer another way out. 4 S. ROSS

This way does not require another volume of debates or defi nitions but instead an explicit grounding within individual academic fi elds that have individually wrestled with theory since it became legible as a scholarly value. Institutionally, traditional disciplines may exist (to put it mildly) in tension with newer DH initiatives and practitioners, but this tension neither precludes the enrichment of DH by disciplinary knowledges nor nullifi es our obligation to learn and apply them. Admittedly, some varieties of DH may require more creativity in identifying their theoretical forebears (critical makers may have to turn to art and architecture), but in the case of digital literary studies, such indebtedness seems so obvious as to require no special acknowledgement. Yet it does need saying. As Brian Croxall observes in his Call for Papers for the Association of Computers and Humanities-sponsored panel at MLA 2014, “what is sometimes forgotten is that the output of digital analysis is not itself the goal; rather, such analysis is a means to an end, and that end is the interpretation of a text or corpus (understood widely).” Croxall’s panel was intended to “re- establish this understanding and conversation, defamiliarizing the conversation about the digital and making it re-familiar.” 11 Both of us answered Croxall’s CFP, and our presentations at MLA 2014 both argued that this interpretation, this focus on the literary ends rather than (only) the digital means, requires invoking, engaging with and ultimately contributing to discipline-specifi c arguments. This roundtable was in fact the starting point for Reading Modernism with Machines , which similarly asked contributors “to focus not on their methods but instead on the interpretations they have reached as result of their digital praxis,” as Croxall phrased it. The chapters in this volume therefore use digital methods to intervene critically in conversations current in modernist studies, foregrounding the interpretive signifi cance of their results rather than devote the larger portion of their argumentation to technical excursuses or methodological summaries. This signifi cance does not reference the statistical variety—though of course that is also necessary—but the literary-critical variety. Doing so avoids “the fetish of technology” that Hal Foster argues were typical of “machinic modernisms” of Futurism and Vorticism (and of Moholy-Nagy), under which “a machinic style was held out as the lure of a technological future to which people were asked, indeed com- pelled, to accede,” treating technology as “a force in its own right and/ or an emblem of ‘the modern spirit’” (7). 12 A properly modernist digital humanities will not fall into the traps that F. T. Marinetti and Wyndham Lewis fell into, but will instead use disciplinary norms to avoid fetishizing INTRODUCTION 5 technology—whether that is the technology of the modernists or the contemporary technology of the Digital Humanities—and will refl ect on the distance between our two machinic epistemes. As Foster warns of avant- garde machinic modernisms, the task of modernist digital humanities is not “to extrapolate the human toward the inorganic-technological” or “to trope the inorganic-technological as the epitome of the human” (15), but to identify when, how and why that happens, whether interpreting it as a trend in our object of study or observing it in other works of scholarship. This requires combining digital and traditional methods judiciously, allowing a space to refl ect critically on the aesthetic, political and philosophical ramifi cations of the ever-changing defi nitions and dependencies that connect humans, technology and the humanities. Teasing out these relations requires not a new platform or tool but renewed, sustained engagement in fi eld-specifi c conversations. It requires raising awareness within these individual literary fi elds, advocating for digital work by making explicit connections between these ongoing debates and by winning over scholars who do not identify as digital humanists or who show resistance or hostility to it. It requires creating formal and informal contexts for DH-minded scholars in the same fi eld to work with one another, encouraging conversations over social media and list-servs, during regional and national conferences, between faculty and students at the same institution or graduate students at different institutions. It requires addressing the canon—the canon that exists and the canon that does not exist. Making new contributions to the literary analysis of canonical works will help to ensure continued buy-in from the particular discipline; we do, indeed, always need another interpretation of Ulysses or Mrs. Dalloway . At the same time, though, DH is well suited to interrogating the canon, analyzing its origin and tracing its effects. Most importantly, it is, due to increased opportunities for electronic publishing, especially well suited to expanding the canon. It requires taking an inventory of the methods common to the individual fi eld and comparing these to DH methods in order to devise tailor-made DH processes appropriate to a specifi c fi eld. Devising these will in turn require returning to questions of style, to aesthetics and revisiting the period’s or fi eld’s non-fi ction, manifestoes, artist statements and contemporaneous reviews and criticism—particularly with an eye toward book history, media and technology. Over time, a fl exible collective identity, something close to a brand that is always close to disbanding, should emerge. And it should employ a different style and tone for each audience: inward toward other disciplinary DH specialists, 6 S. ROSS across to disciplinary specialists who do not identify with DH and outward to the broader DH community. The kind of fi eld-specifi c work outlined in these propositions has just begun to emerge, with publications such as Comparative Textual Media in new media studies, 13 Digital Rhetoric for rhetoric and composition, 14 Early Modern Studies after the Digital Turn 15 and Shakespeare’s Language and Digital Media 16 for the early modern period, and Virtual Victorians for Victorian literary and cultural studies. 17 Like Reading Modernism with Machines , these collections similarly examine the relationships between machinic epistemes—for example, “The virtual Victorian era that is emerging from our algorithmic searches, our digital editions and tools, and our data visualizations necessarily refl ects that era’s own navigation of changing media and information technology” 18 —without oversimplify- ing. Making Canada New: Editing, Modernism, and New Media , edited by Dean Irvine, Vanessa Lent and Bart Vautour, 19 does so through its laser- sharp focus and its refl exive analyses of the processes of digital editing. Reading Modernism with Machines is meant to participate in and advance this fl owering of discipline-specifi c work in the Digital Humanities. In this collection, we attempt to provide a context for a “modernists only” conversation, present new analyses of canonical works, use DH methods to revise and expand the modernist canon, refl ect on the intimacies and distances between our two machinic epistemes, test new digital methods of analyzing modernist texts and examine the limitations of certain DH strategies for understanding modernism. In the context of modernism, this attention to discipline-specifi c approaches toward literary interpretation arrives almost belatedly, considering the past twenty years of exciting modernist digital archives, from the maturation of early initiatives (such as the Modernist Journals Project, Modernist Versions Project and Editing Modernism in Canada) to the fl owering of second-generation initiatives (such as Linked Modernisms, the Modernist Commons and the Open Modernisms Anthology). On the other hand, it is because of these twenty years of modernist digital humanities initiatives that Reading Modernism with Machines can attempt to isolate a relatively coherent, specifi cally modernist set of approaches, while leaving room to evaluate (especially to critique) our characteristic digital methods. Our contributors’ professional experiences with modernist digital initiatives ensure that the collection not only takes stock of what digital modernism has looked like and has enabled over the past twenty years, but also projects new methods of interacting with these resources and suggests INTRODUCTION 7 what kind of modernist digital resources are still needed. Taken as a whole, these essays serve as a barometer for future forms of modernist digital humanities. They foreshadow more critical labor: work done to identify problems in the “big data” we generate about modernism while nevertheless continuing to experiment in quantitative methodologies. This work will likely reconsider and expand the canon, develop feminist digital modernisms and postcolonial digital modernisms, and consider the importance of pedagogy and student labor. Following trends in modernist criticism, it will also likely visualize transnational or global networks of modernism and engage with book history (particularly regarding copyright and publication history) and new media studies.

FIELD-SPECIFIC DIGITAL HUMANITIES IN ACTION: CHAPTER SUMMARIES Appropriately, then, Reading Modernism with Machines opens with Chap. 2 , Dean Irvine’s “ModLabs,” which makes great strides in this latter project by adapting the methods of new media studies to the specifi c context of modernist literature and culture. Irvine recontextualizes twentieth- century laboratory environments—from Bell Labs to the Stanford Literary Lab—as sites of modernist creation and critique. Identifying a core concept, the “modular,” that links this century of laboratories, Irvine histori- cizes, and thereby demythologizes, current work in the Digital Humanities that appears to be sui generis . Irvine’s corresponding remediation of the modernist avant garde links modernism and digital cultures methodologically, revealing commonalities through histories of laboratory practice. In doing so, Irvine not only reveals the modernist practices surviving in digital laboratories and DH centers (including a compelling reading of Bruno Latour and Steve Woolgar’s Laboratory Life ), but also advances a useful critique of labor practices inside these labs. One motif in Irvine’s contribution—a rejection of the oversimplifi ed perspective that pits distant reading against close reading—swells into the dominant strain with Chap. 3 , “Modeling Modernist Dialogism: Close Reading with Big Data,” by Adam Hammond, Julian Brooke and Graeme Hirst. This essay fl exibly combines an impressive range of texts, seamlessly incorporating modernist criticism, DH theory, modernist literature and the data they produced into two different projects—one designed to auto- detect the many speaker transitions in T. S. Eliot’s The Waste Land and the other designed to identify free indirect discourse in texts by Virginia Woolf 8 S. ROSS and James Joyce. Refl ecting on the “hybrid approach” they developed to combine close and distant reading methods, Hammond, Brooke and Hirst link these two projects under the rubric of “dialogism,” managing both to leverage the power of big data and to deliver a sophisticated revision of the dialogic theories of Mikhail Bakhtin and Erich Auerbach. Chapter 4 , Alex Christie and Katie Tanigawa’s “Mapping Modernism’s Z-axis: A Model for Spatial Analysis in Modernist Studies,” completes a classic DH methodological trifecta by adding digital mapping to Irvine’s cultural studies analysis and Hammond, Brooke and Hirst’s big data collection. Christie and Tanigawa’s pioneering work on the “z-axis” methods of mapping literary texts, developed collaboratively as an arm of the Maker Lab in the Humanities at the University of Victoria, distorts historical maps according to the frequency and depth of the text’s engagement with a particular locale. By highlighting the discrepancies, exaggerations and elisions that differentiate a modernist text’s spatiality from that produced by contemporaneous maps, Christie’s map of Djuna Barnes’s Nightwood and Tanigawa’s map of Jean Rhys’s Quartet avoid a spuriously objective approach, instead introducing anti-realist style of mapping. This subjective geography enables Christie and Tanigawa to develop innovative queer and feminist readings of these modernist novels and to posit new connections among gender, sexuality and space in modernist literature. Chapter 5 , Kathryn Holland and Jana Smith Elford’s “Textbase as Machine: Graphing Feminism and Modernism with OrlandoVision ,” continues this emphasis on using digital humanities tools to advance feminist approaches to modernism. Rather than visualize primary texts, however, they visualize modernist feminist scholarship by analyzing the scholarly articles comprising the digital resource Orlando . A case study on suffrage societies sheds new light on Dora Marsden’s crucial mediation of feminist politics through her editorship of The Freewoman , while a case study on Newnham College, Cambridge, recovers the college’s centrality as a site of modernist production, weaving a rich cultural and intellectual history that, among its other insights, reveals an important new context for Virginia Woolf’s A Room of One’s Own . Beyond reconstructing feminist networks in modernism, their analysis makes explicit the networks implicitly created by the Orlando database’s metadata, showing how digital resources and archives themselves, not simply their contents, constitute fertile subjects for research. Chapter 6 , Hannah McGregor and Nicholas van Orden’s “Remediation and the Development of Modernist Forms in The Western Home Monthly, ” continues this focus on digital archives. Their analysis advances theories INTRODUCTION 9 of mediation by interrogating it as not only an operation performed by the digital archive on its objects of representation, but also as a mechanism at work in the layout of early twentieth-century magazines, and as a specifi cally modernist technique. Focusing on a middlebrow Canadian magazine, McGregor and van Orden pivot between close and distant reading to question long-standing assumptions about the differences between middlebrow periodicals and modernist little magazines. Their fi ndings therefore substantiate recent scholarship reconsidering the middlebrow, and as McGregor and van Orden engage substantively with recent developments in new media studies—particularly through a fascinating discussion of the remediation of the phonograph and radio in The Western Home Monthly —they forge a new method that combines strategies from new media studies and modernist studies. In Chap. 7 , Wayne Arnold’s “Stylistic Perspective Across Kenneth Fearing’s Poetry: A Statistical Analysis,” also adjusts the modernist canon by reappraising neglected Canadian literature. Best known today for his 1946 novel The Big Clock , Fearing presents a fascinating test case for stylometry because, as Arnold explains, critics have cited a lack of stylistic change or development to explain the decline of his reputation as a poet. Arnold tests this claim by statistically analyzing 142 poems to establish usage patterns for words that, in other digital methods, are often thrown out as stop words or simply not retained as useful data. This includes pronouns, question marks and repetition, all of which Arnold traces to verify the existence of a stylistic change in Fearing’s oeuvre. Arnold’s chapter therefore demonstrates the play of scale invoked by DH—scanning hun- dreds of poems for variations in words of two or three letters—enables signifi cant interventions in understanding the course of a poet’s career. Chapter 8 , Adam James Bradley’s “In the End Was the Word: A Computational Approach to T. S. Eliot’s Poetic Diction,” similarly focuses on the large-scale analysis of small-scale choices made by a single poet, but unlike Arnold’s contribution, Bradley’s chapter explores Eliot’s penchant for exotic, archaic or self-coined terms. With the precision and patience of a linguist, Bradley outlines the precautions, extensive research and subtle reasoning necessary for preparing poetic data for submission to computational processes. By running comparisons of Eliot’s diction with the Oxford English Dictionary and with contemporaneous journalism, Bradley tests the accuracy of Eliot’s and Ezra Pound’s claims about modern poetic diction and of modernist scholars’ generalizations about Georgian poetry (and its differences from Victorian poetry and from high modernist 10 S. ROSS poetry). The results, which are, unsurprisingly, mixed—some truisms are confi rmed, others disconfi rmed—should be required reading for anyone engaging with Eliot’s criticism, particularly “Tradition and the Individual Talent” and “The Metaphysical Poets.” Equally as attentive to the nuances of diction and etymology, Chap. 9 , Jonathan Reeve’s “A Macro-Etymological Analysis of James Joyce’s The Portrait of the Artist as a Young Man ” also sheds new light on high modernism through macroanalysis. Reeve’s deep familiarity with and engagement with contemporary criticism of Joyce, along with his focused close readings of signifi cant passages in Portrait , demonstrates how comfortably digital and analog forms of reading can be interwoven. Though Reeve primarily uses his statistical results to confi rm the scholarly consensus regarding Stephen’s characterization, his individual discussions of each etymological origin in the narrative—the changing proportions of Latinate, Germanic, Hellenic, Celtic and unidentifi ed words—unearth new facets of Stephen’s religion, sexuality and politics. For Joyceans, Reeve’s hypotheses regarding Joyce’s attitude toward Irish nationalism will be of especial interest, while the more general audience can attend to Reeve’s broader thesis that macro-etymological analysis reveals that etymology and narrative structure are fundamentally related. Chapter 10 , “Body Language: Toward an Affective Formalism of Ulysses ,” by Kurt Cavender, Jamey E. Graham, Robert P. Fox, Jr., Richard Flynn and Kenyon Cavender, similarly tackles Joycean style, while also intervening in critical conversations about sentiment analysis (in DH) and affect theory (in literary theory). Critiquing machine sentiment analysis for their hard-wired conceptual slippages, Cavender’s team builds a “bodily lexicon” from Ulysses itself to compensate for the shortcomings of existing sentiment lexicons, then close read passages in the text suggested by results from their custom program, Affectcrawler. These close readings are solid examples of DH analysis successfully incorporating literary theory, yielding an ingenious theory accounting for Bloom’s centrality in the text and a comparison of modernist and Victorian affect (by running a comparison with George Eliot’s Middlemarch ). Perhaps most excitingly, they generate a reading of disability in the “Calypso” episode, demonstrating the unintuitive power of using algorithms to explore bodies in modernism. Bodies also matter in Andrew Pilsch’s Chap. 11 , “‘We Twiddle … and Turn into Machines’: Mina Loy, HTML and the Machining of Information.” Pilsch explains the diffi culties he encountered while coding INTRODUCTION 11 poems such as “Parturition” (Loy’s physically intense poem about childbirth) for inclusion in his archive, Mina Loy Online. Linking Loy’s powerful representations of modern bodies under the stresses of sex, childbirth, industrial labor and world war to her critique of androcentric futurism, Pilsch establishes the signifi cance of Loy’s idiosyncratic typography as a mode of feminist embodiment. Insistent on the duty of digital archives to preserve the precise typography of modernist poetry, Pilsch uses the resistance of HTML against Loy’s typography as a context for comparing Karl Marx’s and Alan Turing’s different concepts of the machine and for unpacking the history of hypertext. Arguing that the theory of textuality inhering in HTML is regressive, non-modernist and anti-feminist, Pilsch ends by advocating the creation of a more fl uid, fragmented and feminist markup language. A similarly strong critique of contemporary digital culture motivates our fi nal Chap. 12 , Eunsong Kim’s “CGI Monstrosities: Modernist Surfaces, the Composite and the Making of the Human.” Also like Pilsch, Kim uses the modernist methods of critique that motivate the other essays in order to pursue the political ramifi cations of modernist concepts inhering in our contemporary machinic episteme. Kim identifi es the perpetuation of modern philosophies of positivism in criticism on contemporary fi lm editing, then adapts modernist methods to provide an alternate reading of CGI (Computer Generated Imagery). Deconstructing the visual techniques used in CGI animation for the HBO series John Adams and the South Korean horror fi lm The Host , Kim reveals the same colonial tropes that are at also at work in literary modernism. CGI’s compositing of the human body, its erasure of labor through editing, its collapse of three-dimensional fi gures into two dimensions, and its neocolonial logic that produces monstrous others and monstrous crises while ignoring the monstrosity of its own methodical nanoscopic processes—all of these paradoxes inhering in the fabrication of the human body by algorithms, Kim shows, are the end- point of a process begun in modernism. Unlike the other essays, Kim’s essay does not read a modernist text, but rather teaches us how to read Reading Modernism with Machines ; it anato- mizes the procedures of digital representation to remind us that each computational transformation of a text, whether through the creation of an archive or tool or through data analysis or visualization, crystallizes each decision made by a scholar into a representational given. These givens are too often invisible, ignored or suppressed, all in the name of strengthening our arguments—yet it is only by including our disciplinary knowledges at 12 S. ROSS every level of digital analysis that we can control for, compensate for or minimize the distortions of texts and humans by machinic representation. Moholy-Nagy was therefore correct to assert that the task of the subjects of each machinic espisteme is “to fi ght for a new spirit to fi ll the forms stamped out by the monstrous machine” (300). Indeed, all the chapters comprising Reading Modernism with Machines engage in some way with Moholy-Nagy’s call, whether they leverage or critique the stamped-out forms of our own machinic episteme. As a result, in reading modernism with machines, our contributors not only illuminate the modernist episteme of the machine, but also, just as signifi cantly, add their own infl ections to the “new spirit” of our century’s new machines.

NOTES 1. László Moholy-Nagy, “Constructivism and the Proletariat,” in Modernism: An Anthology of Sources and Documents , ed. Vassiliki Kolocotroni, Jane Goldman and Olga Taxidou (Chicago: U of Chicago P, 1998): 299. 2. Johanna Drucker, “Humanities Approaches to Graphical Display,” DHQ 5.1 (2011); Alan Liu, “Where Is Cultural Criticism in the Digital Humanities?” Debates in the Digital Humanities , ed. Matt K. Gold (Minneapolis: U of Minnesota P, 2011): 490–509. 3. Jesse Stommel, “Critical Digital Pedagogy: A Defi nition,” Hybrid Pedagogy (18 Nov. 2014). 4. Richard Lanham, “The Electronic Word: Literary Study and the Digital Revolution,” New Literary History 20.2 (1989): 265–90. 5. Jessica Pressman, Digital Modernism: Making it New in New Media (Oxford: Oxford UP, 2014); Stephen Ross and Jentery Sayers, “Modernism Meets Digital Humanities,” Literature Compass 11.9 (2014): 625–33. 6. Ross’s collaborative effort, “Manifesto of Modernist Digital Humanities” (Alex Christie, Andrew Pilsch, Shawna Ross and Katie Tanigawa, http:// www.shawnaross.com/manifesto , 14 Nov. 2014) self-consciously adopts the style of Wyndham Lewis’s manifestoes into HTML format to argue that modernism can be seen as a particular methodology used by the Digital Humanities, not only as an object of analysis. This adoption was intended to reproduce Lewis’s tone of playful antagonism and to perform, in an exaggerated fashion, the tendency of modernist digital humanities to value modernist argumentation but enact realist methodology. Though readers are certainly free to interpret the Manifesto as they like, the intention was not necessarily to privilege modernist methods monolithically but to raise awareness that they could be developed alongside realist methods. INTRODUCTION 13

7. Shawna Ross, “Digital Modernism as Method: Recent Publications in Digital Humanities,” Journal of Modern Literature 39.3 (2016). 8. See Stephen Ramsay, “On Building,” Stephen Ramsay (11 Jan. 2011); Geoffrey Rockwell, “Is Humanities Computing an Academic Discipline?” Defi ning Digital Humanities , ed. Melissa Terras, Julianne Nyhan and Edward Vanhoutte (Burlington, VT: Ashgate, 2013): 11–34. 9. Cecire, Natalia. “Introduction: Theory and the Virtues of Digital Humanities,” Journal of Digital Humanities 1.1 (2011); Steven E. Jones, “Introduction,” The Emergence of the Digital Humanities (New York: Routledge, 2014): 1–11. 10. Jean Bauer, “Who You Calling Untheoretical?” Journal of Digital Humanities 1.1 (2011); Chris Forster, comment posted to Bauer (4 Nov. 2011). 11. Croxall, Brian, “Methods and More for ‘Beyond the Digital’ at MLA 2014,” ACH.org (30 Dec. 2013). 12. Hal Foster, “Prosthetic Gods,” Modernism/Modernity 4.2 (1997): 7. 13. N. Katherine Hayles and Jessica Pressman, eds., Comparative Textual Media: Transforming the Humanities in the Postprint Era (Minneapolis, MN: U of Minnesota P, 2013). 14. Douglas Eyman, ed, Digital Rhetoric: Theory, Method, Practice (Ann Arbor, MI: U of Michigan P, 2015). 15. Laura Estill, Diane Jakacki and Michael Ullyot, eds., Early Modern Studies after the Digital Turn (Toronto: Iter and Tempe, AZ: Arizona Center for Medieval and Renaissance Studies, 2016). 16. Janelle Jenstad and Jennifer Roberts-Smith, eds., Shakespeare’s Language and Digital Media: Old Worlds, New Tools (Burlington, VT: Ashgate, 2016). 17. Veronica Alfano and Andrew Stauffer, eds., Virtual Victorians: Networks, Connections, Technologies (Basingstoke: Palgrave Macmillan, 2015). 18. Andrew Stauffer, “Introduction,” Virtual Victorians : 8. 19. Dean Irvine, Vanessa Lent and Bart Vautour, eds., Making Canada New: Editing, Modernism, and New Media (Toronto: U of Toronto P, 2016). CHAPTER 2

ModLabs

Dean Irvine

In the spirit of Émile Zola’s Le roman expérimental (1880), in which the naturalist novel becomes a kind of laboratory subject to the laws of scientifi c method, the modernists set up their own laboratories for the production of literature, fi lm, radio, the performing arts and the visual and plastic arts. From Hugo Munsterburg’s psychology lab at Harvard and Alfred Stieglitz’s 291 gallery in New York at the turn of the twentieth century to the European and North American art and design labs of the 1920s and 1930s, the modernist period witnessed the emergence of institutional formations that brought together artists, writers, fi lm makers, architects, sociologists, psychologists, anthropologists, economists, scientists and engineers in a transatlantic cultural movement that tra- versed disciplinary boundaries and fostered new modes of collaboration. What these laboratories of art and design consistently demonstrate is the extent to which the institutional structures of the artistic avant garde were informed by their critique of scientifi c and corporate models of research; their critique of the very models they reference varied from ideological rejection of industrial capitalism and Western science to aestheticist reform of industrial design. With the creation of studio-laboratories in

D. Irvine ( ) Dalhousie University, Nova Scotia, Canada

© The Editor(s) (if applicable) and The Author(s) 2016 15 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_2 16 D. IRVINE the 1960s and 1970s, the late twentieth century welcomed a new generation of collaboration among artists, scientists, engineers and industry that modeled itself on avant-garde labs of the early twentieth century. The opening of MIT’s Media Laboratory in 1985 announced its experiments with digital media being “as much like the Bauhaus as a research lab,” 1 which at once moved toward the formation of digital-humanities and new-media laboratories of the late twentieth and early twentieth-fi rst centuries and, at the same time, returned to the avant-garde labs of the modernist period. This chapter extends from research on the genealogies of modernist laboratories (aesthetic, scientifi c, and corporate) and their relationship to the formation of digital-humanities and new-media laboratories. The trajectory of this project, broadly conceived, leads from the modular principles of early twentieth-century industrial design and avant-garde aesthetics, to the implementation of modular architecture in mid-century corporate science labs, to the modularization of markup languages, interoperable digital tools, and collaborative and cross-disciplinary lab environments in the late twentieth and early twenty-fi rst centuries. In doing so, I traverse the conjuncture of multiple disciplinarities—from architecture, art, and industrial design to markup languages, source code, and digital tools—in the historical contexts of cultural, industrial, and postindustrial modernities. I bring into focus the intersection between the production of modernist art as aesthetic experiments conducted under laboratory conditions— sometimes literally, other times fi guratively—and the positioning of digital humanists in lab environments experimenting with data derived from digitally remediated materials originally produced in analog formats by the historical avant garde. Of the contemporary practitioners and labs prob- ing novel approaches to textual studies, discussed here are the Speculative Computing Laboratory at the University of Virginia, founded by Johanna Drucker, Jerome McGann and Bethany Nowviskie; the Stanford Literary Lab, founded by Franco Moretti and Matthew Jockers; the Harvard Cultural Observatory, founded by Jean-Baptiste Michel and Erez Lieberman Aiden; and the Modernist Journals Project Lab, co-directed by Sean Latham, Robert Scholes and Jeff Drouin. 2 Over the past two decades, such labs have emerged as institutional locations at which critical theorists, textual scholars, librarians, archivists, programmers, scientists and engineers converge to implement collaborative methodologies that have transformed and tested the increasingly expansive scale and versatile disciplinary mix of modernist textual studies. MODLABS 17

NEW-MEDIA MODERNISM As much as tropes of innovation and experiment were commonplace for the ways in which modernists imagined themselves and their aesthetic practices, they have now become synonymous in the co-emergent formations of “new modernist” and “new media” studies. One of the common tropes in new-media studies, at least since Jay David Bolter and Richard Grusin’s Remediation (1999), constructs analogies between late twentieth and early twenty-fi rst century digital media and early twentieth- century modernism. Bolter and Grusin’s comparison of “the rhetoric of cyberspace” to futurist manifestos and the suggestion that “cyberspace enthusiasts have a similar relationship to technologies of representation that Marinetti and the futurists had to technologies of motive power” is one example of their collocation of modernism and digital media. 3 Their premise is that digital objects call attention to their remediated materiality in a manner analogous to the aesthetics of modernist self-referentiality, which foregrounds its own mediations. This premise is not theirs alone: it’s a recurrent claim in the work of new-media theorists. 4 In other words, just as the modernists were audacious enough to proclaim themselves the fi rst to be modern, so theorists and creative practitioners working with digital media claim for their discipline and their art the distinction of being the fi rst to be new. As infl uential as the work of new-media theorists has been in shaping my thinking about remediated modernism, I am uneasy with the analogi- cal correlation of either modernism or the avant garde and digital media. This seductive trope permits a fantasy of a transhistorical modernism that focuses on what Matthew Kirschenbaum calls the “formal materiality” of interface rather than the “forensic materiality” of storage. 5 New media’s always-new modernism is troubled by persistent reminders of the historicity and materiality of digital objects. One of the ways we experience this online is when we click on a broken link and get an “Error 404: Not Found” message. If, in the normal functioning of a website, we expect that the information we seek will be rendered in what Lisa Gitelman inge- niously calls (in a variation on Gertrude Stein) the “continual, continuous present tense” of the Internet, the error message makes us aware of the historicity of hyperlinks whose brokenness refers us to a temporality to which we no longer have access. 6 They interrupt our navigation of digital space, create ruptures and remind us of the persistent materiality of remediated objects. Errors also force us to acknowledge the labor necessary 18 D. IRVINE to the production of digital media, for when the link is broken, we must confront the fact that remediated materialities get old, require rebuilt code and reside on machines that need maintenance. What gets effaced in reading modernism in the “continual, continuous present tense” of new media’s always renewable interface is the sedimentation of labor time and economic mode of production in the storage and retrieval of data. The fantasy of an always-new modernism dissolves once we come to understand the ways in which digital objects are not perennially renewable resources, but rather material entities that age, degrade and break down and require continual reinvestments of labor and capital. Where I do see the productive correlation of modernism and its digital remediations is in the turn to the laboratory as an environment for research, experimentation and innovation. Lev Manovich’s correlation of the laboratory research of the 1920s avant garde with the software research lab seeks “to reactivate the concept of laboratory experimentation,” which he locates in “the kind of research undertaken by Russian and German avant-garde artists of the 1920s in places like Vkhutemas and Bauhaus, as they explored the new media of their time: photography, fi lm, new print technologies, telephony.” 7 While Manovich helpfully locates among the historical avant garde specifi c institutional formations that correspond to the research environment of digital humanities labs, his desire to correlate avant-garde software design and modernist new media tends toward an aestheticization of technology that abstracts it from the material conditions that fi gure so concretely in laboratory research. The actual laboratory environments in which members of the avant garde generated research have been the subject of case studies and histories of cultural movements, but these laboratory modernists extend far beyond the Russian and German examples to which Manovich alludes. What remains to be considered is not only the transatlantic reach of modernist laboratory practice from Europe to North America but also the experimental interpretive strategies that digital-humanities practices bring to the examination of laboratory modernisms using the tools and methods of lab research.

LABORATORY MODERNISMS Hugo Münsterberg’s psychology lab at Harvard in the 1890s marks a formative moment and location in laboratory modernisms. His experimental “study of the aesthetic feelings” 8 —launched with a catalogue of MODLABS 19 his lab’s instruments, charts and models for the 1893 World’s Columbian Exhibition 9 —culminated in a series of publications on visual art and fi lm, including The Principles of Art Education (1904), The Photoplay (1916) and the posthumous Twentieth Century Painting (1951). If, with his textual and photographic inventories of the Harvard lab, he transformed psychological mechanisms into aesthetic objects, his investigations of early fi lm transformed aesthetic objects into mechanisms of psychic life. The dual attention in his writings to scientifi c and psychological mechanisms and their aesthetic and psychological effects spurred him to coin “psy- chotechnology” 10 to defi ne a fi eld of research that views psychic life as technological apparatus and technologies as psychic instruments. 11 Where Giuliana Bruno reads Münsterberg’s representations of the Harvard lab as elements of a “precinematic design,” 12 I read his writings on fi lm and painting as extensions of his laboratory research. Münsterberg’s The Photoplay intimates as much, concluding that the “screen ought to offer a unique opportunity to interest wide circles in psychological experiments and mental tests and in this way to spread the knowledge of their importance for vocational guidance and the practical affairs of life.” 13 The “picture house,” as he calls it, is a laboratory for and of the masses; this is consistent with his research on industrial psychology and art education, which addresses the psychic life of the masses. With Henning Schmidgen’s 2008 digital exhibit, “Münsterberg’s Photoplays,” which is part of a larger initiative at the Max Planck Institute for the History of Science called The Virtual Laboratory , it is possible to reconstruct “the material culture of Münsterberg’s psychological laboratories.” Schmidgen’s curatorial work on Münsterberg’s labs attends to the material histories of knowledge production typically relegated to the background of dominant narratives of new media from the modernist period; that is, instead of focusing on his lab’s experiments or on his theorization of fi lm, it calls attention to the apparatus itself and its materiality as cultural phenomena whose histories have much to tell us about the institutional formations of modernism. Another German-American, the photographer Alfred Stieglitz, established in February 1908 the earliest of the aesthetic laboratories of the modernist period. Stieglitz was at once artist and scientist, trained in chemistry, laboratory methods and new technologies related to cameras, lenses, developing and printing processes. 14 Building on the scientifi c laboratories that he had designed and directed (which included the Photochrome Engraving Company and the Camera Club of New York), his curation of shows for the Photo-Secession group, and his editorship 20 D. IRVINE of the magazine Camera Work , 15 Steiglitz opened a new gallery on Fifth Avenue in New York. Cryptically named “291,” its new address was actually 293 Fifth Avenue, it was housed in premises across the hall from the Little Galleries of the Photo-Secession. 16 What distinguished 291 from its predecessor was that its shows broke down traditional boundaries between art and photography by exhibiting paintings, sculptures, drawings alongside photographs. The intention was to create a space for dialogue among painters, draftsmen, sculptors and photographers; it also aimed to generate dialogue between European and American artists (by introducing avant- garde works by Matisse, Rodin, Cézanne, Picabia and Duchamp) and between established and emergent creative practitioners. Given Stieglitz’s scientifi c training and his cultivation of a collaborative, intermedia and interarts environment, it is natural that he would consistently refer to 291 as a “laboratory.” More provocatively, he was fond of telling newspaper reporters that the gallery was really a “laboratory,” a “storm center” for revolt “against all authority in everything” and for “testing the taste of the public.” 17 While Stieglitz’s progressive rhetoric waxed revolutionary, his 291 laboratory mounted a rather timid revolt in comparison to postwar modernists. The laboratories of the Soviet constructivists were geared toward research on and production of art in the service of the post-October revolution Bolshevik government. Returning from the Soviet Union in 1927, Walter Benjamin offered a portrait of postrevolutionary Moscow in which every citizen is part of a mass experiment: “Each thought, each day, each life lies here as on a laboratory table. And as if it were a metal from which an unknown substance is by every means to be extracted, it must endure experimentation to the point of exhaustion.” 18 Even so, the laboratories of the constructivists were not virtual; they inhabited actual institutions and conducted hands-on research. Constructivism in Moscow found its home in the Vkhutemas, the school for art and design established in 1919, at Inkhuk, the Institute of Artistic Culture founded in 1920, and at Obmokhu, the Society of Young Artists, which fi rst exhibited in 1921. There were, in effect, two constructivisms: a laboratory phase of 1920–21, followed by a production phase. The two phases are commonly confl ated, and more often than not the constructivists are associated with a set of internationally recognized stylistic features—abstract, geometric, industrial, spatial construction—and categorized as “laboratory” art. Equally prevalent is the assumption that the constructivists embraced a function- alist aesthetic, which may be true enough of their production phase but MODLABS 21 not of their early laboratory experiments. The constructivist laboratory mandates a mode of artistic production that operates in concert with the ambitions of the Soviet revolutionary state by eradicating marks of individual expression from the work of art. At the same time, the constructivists’ appropriation of the laboratory as a space for artistic research moves away from the model of the artist’s studio toward the dynamic, collaborative, cross-disciplinary and rationalized experimental spaces inhabited by scientists and engineers. 19 Often overshadowed by the legacy of design workshops and laboratories at the Bauhaus, Roger Fry and other members of the Bloomsbury Group preceded their German counterparts when they conceived the London-based Omega Workshops in late 1912. 20 During its period of operations from 1913 to 1919, the Omega Workshops saw Fry’s aesthetic vision extend from the production of paintings and drawings to hand- painted furniture, murals, mosaics, ceramics, stained glass, carpets and tex- tiles; it even ventured into book design and dressmaking. Subsequent to the Omega Workshops, Fry continued to develop his ideas about the pro- vision of space and material support for experimentation in art and design. What he proposed, in 1924, was the establishment of a “Laboratory of Design rather than a School of Design … an institution where all the necessary apparatus … for making designs would be supplied, where also there would be a certain small staff fully equipped with the purely technical knowledge of the exact requirements of various industries.” 21 Drawing analogies to the economic model of scientifi c research and development, Fry posited that “it would be essential that the burden of executing trial pieces and getting them tested on the market should be borne by the Laboratory of Design itself.” 22 While Fry managed to sustain the Omega Workshops for seven years, its collapse after the First World War exposed the fact that his economic experiment was subject to the vicissitudes of the art market. Just as the Omega Workshops closed down, another model of a state- funded “Laboratory of Design” that Fry theorized came into existence with the founding of the Staatliches Bauhaus in 1919. During its decade or so of operation, the Bauhaus existed in three German cities: at Weimar (1919–25), at Dessau (1925–32), and at Berlin (1932–33), when it was shut down under pressure from the Nazi state. Ever since they opened, there have been numerous comparisons between the Vkhutemas and the Bauhaus, both of which were state-funded institutions that merged craft traditions with industrial modes of production, but not without dissent 22 D. IRVINE from the Soviets, who protested that their German counterparts “simply designed aesthetic versions of capitalist industrial objects in the tradition of applied or decorative art” and that the constructivists “aimed to produce an entirely new order of objects within socialist production.” 23 Not long after he was installed at Dessau, Walter Gropius published a manifesto in which these objections to false comparison seem reasonable enough. His plan outlined an extension of capitalist production, with

workshops [that] are essentially laboratories in which products suitable for mass production and typical of our time are developed and constantly improved. In these laboratories the Bauhaus wants to train a new kind of collaborator for industry and the crafts, who has an equal command of both technology and form. …Speculative experiments in laboratory workshops will yield models and prototypes for productive implementation in factories. 24

With the mass exodus of German intellectuals after the Nazis took power, the Bauhaus did not so much disappear as continue its itinerant narrative of relocation and transformation. Many of its directors and principal instructors eventually landed in the United States: Gropius taught at the Harvard School of Design, Ludwig Mies van der Rohe settled in Chicago, as did László Moholy-Nagy. In 1937, at the invitation of the Association of Art and Industry, Moholy- Nagy moved to Chicago, where he founded a design school, which he called the New Bauhaus. The infl uence of the Bauhaus laboratory model is evident elsewhere, so much so that some historians have suggested that the its design principles, pedagogical methods and signature “international style” were practiced as much—if not more—elsewhere as at the Bauhaus itself. With a curriculum lifted from Bauhaus, the New York-based Design Laboratory was founded in 1935; it started as one of the Works Progress Administration cultural programs, before undergoing a series of reorga- nizations until it was dismantled in 1940. 25 The Design Laboratory, far more than an art school, “furnished a vibrant point of contact between the business culture of America’s industrial design entrepreneurs, the experimental modernism of the depression-era avant-garde, the unprecedented public arts bureaucracy of the New Deal, the militant unionism of the Congress of Industrial Organizations, and the radical cultural politics of the Popular Front.” 26 The year before, he was appointed director of New York’s Museum of Modern Art (MoMA), Alfred H. Barr, Jr. toured Gropius’s Bauhaus. 27 MODLABS 23

Adopting Bauhaus principles of minimalist aesthetics and industrial design, Barr applied them to a style of display at MoMA—the “white cube”—that is now ubiquitous in modern museums. MoMA’s 1938 exhibition, curated by former Bauhaus instructor Herbert Bayer, produced a catalogue that not only served as a sourcebook of primary materials for American audiences but also implemented the asymmetrical typographic design characteristic of Bauhaus publications. On the occasion of MOMA’s tenth anniversary in 1939, Barr organized the “Art in Our Time” exhibition and contributed to its catalogue one of his most frequently quoted statements: “The Museum of Modern Art is a laboratory: in its experiments the public is invited to participate.” 28 When, in November 2009, MoMA opened its “Bauhaus 1919–33: Workshops for Modernity” exhibition, it foregrounded Barr’s vision, inviting visitors to participate in a program advertised as the “Bauhaus Lab”:

a new interactive space that reimagines the classrooms of the historic Bauhaus School in Germany. … Led by artists, educators, and art historians, an ongoing series of hands-on artmaking workshops offers participants the opportunity to engage in techniques and processes integral to the Bauhaus, such as drawing, collage, graphic design, color theory, and mechanical construction. 29

A total of seven workshops were offered, each based on the courses offered by Bauhaus instructors—Klee, Albers, Moholy-Nagy, Itten and Gropius— but recast for a broader demographic in the style of an art camp. 30 Contemporary with the lab-based art of Russia and Germany, France and England experimented with modes of laboratory surrealism. In Paris, Breton and other surrealists working out of the Bureau de recherches Surréalistes , which opened in October 1924 under the direction of Antonin Artaud, not only launched its fi rst journal, La Révolution surré- aliste , on the model of a scientifi c journal but also published the results of psychical laboratory experiments that conducted even as they parodied scientifi c research. 31 The fi rst cover of La Révolution surréaliste shows a photograph by Man Ray that stages a collective act of automatic writing; surrealists swarm around a female typist, who places her fi ngers on the typewriter in a gesture that is conspicuously not typing. Rather, her fi ngers rest on the keyboard in a manner reminiscent of a ouija board, her gaze fi xed on an open book tipped toward the camera that focalizes the collective gaze of some—but not all—of the men huddled around her. This pho- 24 D. IRVINE tograph obviously parodies the scientifi c method of detached observation and recording of experiments, their collective seance amid the apparatus of offi ce equipment with which they conduct an experiment of psychic automatism displacing the apparatus of laboratory instruments. The hidden recording apparatus that cannot be seen, however, is Man Ray’s camera: it captures a staged image of automatic writing in which nothing is written, and in doing so translates into a visual medium what Breton, in reference to écriture automatique , called a “veritable photography of thought.” 32 The emergence of the British Mass-Observation movement coincided with the 1936 International Surrealist Exhibition in London. This exhibition at the New Burlington Galleries, curated by a committee that included Humphrey Jennings and David Gascoyne, claimed to represent works by every prominent Paris surrealist alongside new works by British artists. 33 As in the research conducted at the Bureau de recherches Surréalistes , Mass-Observation, which started in 1937 and lasted through the 1950s, regularly employed tropes drawn from the sciences 34 and conducted cross- disciplinary research by collaborators, including anthropologists, photographers, fi lm makers, literary critics, poets and visual artists. The “sciences” that Mass-Observation applies are not the traditional laboratory-based disciplines of chemistry, physics and biology but the “new” sciences of psychology, anthropology and sociology. 35 The laboratories that Mass- Observation mobilizes among the masses are the institutions of everyday life. Rather than confi ning surrealism to creative practitioners working in an avant-garde laboratory, the surreal “poetry of everyday life” 36 would be found by those independent of surrealism as an aesthetic style through “the observation by everyone of everyone, including themselves” and the everyday world. 37 The movement’s initial manifesto (1937) embraces the aleatory qualities of surrealism, providing a random list of subjects, at once scientifi c and commonplace:

Behavior at war memorials. Shouts and gestures of motorists. The aspidistra cult. Anthropology of football pools. Bathroom behavior. Beards, armpits, eyebrows. Anti-semitism. Distribution, diffusion and signifi cance of the dirty joke. Funerals and undertakers. Female taboos about eating. The private lives of midwives. 38 MODLABS 25

Although most the listed items could be subjects of social-scientifi c investigation, the cumulative effect of the montage is decidedly unscientifi c: it is an archeology of behaviors and things pulled from everyday life, a “systematically unsystematic” 39 catalogue of oddities arranged without explicit method. As the Mass-Observation group reported in its 1943 publication on pub life, “So far scientists have not been very scientifi c about English life. They have confi ned science to limited fi elds of research, and the same man who is a scientist in his laboratory is generally noticeably ‘unscientifi c’ outside it.” 40 Perhaps, then, the Mass-Observation method could be called the unscientifi c method. Analogous to the virtual laboratory mounted by for Münsterburg’s laboratories, Mass-Observation has created its own online archive hosted by the University of Sussex. Mass Observation Online consists of a series of “modules” 41 ; its searchable archive allows users to sift through an immense repository of documents, including photographs, diaries, surveys, reports, research fi les and original Mass-Observation publications. The modular organization of the archive is decidedly systematic; but search results collapse the organizational logic of its archival modules, allowing a user to encounter the logarithmically generated montage of an archive’s contents displayed according to keywords or other search parameters. This is not the aleatory effect of the surrealist montage; it is, rather, the indexical ordering of the digital archives through the modular principles of information management and database architectures. While not all modernists were inclined to found institutional laboratories or practice their craft in such environments, their writing is suffused with the language of laboratory experiment. Among the Anglo-American modernists who drew analogies between aesthetic and laboratory experiment—including T.S. Eliot, Ezra Pound, Mina Loy and William Carlos Williams—the earliest example of which I am aware is Gertrude Stein. After all, Stein actually worked from 1893 to 1897 in Münsterberg’s psychology lab at Harvard, where she assisted on experiments, included herself among the research subjects, and published papers on her fi ndings and studied from 1897 to 1903 in the anatomy laboratories at Johns Hopkins Medical School. 42 In the Autobiography of Alice B. Toklas (1933), Stein makes explicit the connection between the Harvard lab’s psychological experiments in automatic writing and her early prose style:

She was one of a group of Harvard men and Radcliffe women and they all lived very closely and very interestingly together. One of them, a young 26 D. IRVINE

philosopher and mathematician who was doing research work in psychology left a defi nite mark on her life. She and he together worked out a series of experiments in automatic writing under the direction of Münsterberg. The result of her own experiments, which Gertrude Stein wrote down and which was printed in the Harvard Psychological Review was the fi rst writing of hers ever to be printed. It was very interesting to read because the method of writing to be afterwards developed in Three Lives and Making of Americans already shows itself. 43

To take Stein as an early case of modernist writing informed by the research lab, it is useful to recall that she returns to these early works in her autobiography and her Lectures in America (1935) as a literary critic interested in the ways in which scientifi c research infl uenced her own composition methods. In other words, Stein’s literary criticism returns to the laboratory in search of her own avant-garde institutional formation, one that corresponds to the laboratories founded by the interwar avant garde. This is instructive in several ways, not least of which is the model that it offers literary critics who research the interwar avant garde and digital humanists who have returned Stein to new laboratories, where her writing is the subject of research experiments in text mining and visualization, where the modules of her recombinant prose are plotted into graphs, where the rhythms of her reiterative syntax are computed by the functions of logarithms. 44

LABORATORIES, COLLABORATORIES, OBSERVATORIES Bruno Latour and Steve Woogar’s landmark 1979 study Laboratory Life records their ethnography of a neuroendocrinology lab in which they document the observations of a “fi ctional character” who posits, after an initial period of observation, that “the laboratory began to take on the appearance of a system of literary inscription.” 45 This hypothesis “incurred the considerable anger of members of the laboratory, who resented their representation in some literary activity.” 46 The corollary of this hypothesis is obvious enough: if the lab is a literary institution, then the institutional formation of the lab isn’t the exclusive property of the sciences. Or, more troubling, the sciences are themselves a literary institution. What if we were to reverse their hypothesis and examine the system of literary inscription in the humanities as one of laboratory experiments? Amy E. Earhart’s study of collaboration in the digital humanities tentatively takes up this premise, underscoring the need to differentiate MODLABS 27

collaborative institutional formations and practices in the humanities from concepts of “interdisciplinarity or intradisciplinarity.” 47 While Earhart isn’t prepared to advocate that the digital humanities uncritically adopt the “science laboratory model” as the basis for collaboration, she concedes that the laboratory is predicated on an institutional economic model different in both scale and kind from the centers and institutes at universities in which digital humanities research more typically fi nds infrastructural support. 48 “This model,” she concludes, “has not made headway into the digital humanities.” 49 While the recent establishment of digital-humanities laboratories across North America indicates that the nomenclature of the sciences has been transposed to the humanities, it is evident after even a cursory survey that some are comparable to science labs, while others are labs in name only. As a research unit, the digital-humanities lab is no less in need of material infrastructure—space, equipment, personnel, administration—than any science lab. This isn’t to say that centers and institutes can exist without such support but that the dedicated infrastructure of a lab is recognized for its function as a research unit capable of securing resources from government, funding agencies, private donors and industry partnerships to ensure its sustainability. What makes the digital-humanities lab possible under this institutional regime is the development of funding models designed to capitalize on digital economies. I do not want to suggest, however, that this necessitates the capitulation of the humanities to corporate laboratory models. Rather, as in the case of the historical avant garde, the digital-humanities lab may follow a different cultural logic, one that cannot exist without the digital economies of postindustrial capitalist modernity but may instead serve as a structure of institutional critique. The digital-humanities lab is not merely an interdisciplinary or intradis- ciplinary expression of postindustrial modernity; it is an institutional formation whose cultural logic and interests often subsist in tension with both the economic regime that drives technological innovations and the industry-oriented research agendas of colleges and universities that make such labs possible. Rather than intradisciplinarity or interdisciplinarity, the digital humanities and its institutional formations might be better described in terms of extradisciplinarity and paradisciplinarity. Matthew Weinroth applies extradisciplinarity to research that “may take place outside of the traditional confi nes of disciplines, but it remains within the context of epistemic cultures in that it refers to the structure of disciplines. …The disciplinary frameworks provide the grounds for the pursuit of solving a problem 28 D. IRVINE but take these further in the collaborative element.” 50 By paradisciplinarity, he points to “collaboration in-between disciplines through dialogue and exchange, drawing on the differences between collaborators. Thus, paradisciplinarity can be conceptualized as counter-positioning disciplines by comparing formal aspects of disciplines…in order to open up and/or widen relations between individual areas of knowledge production.” 51 In either case, collaboration is the cooperative keyword: where extradisciplinary research retains disciplinarity as an epistemological structuration that pursues problem-solving outside traditional disciplines via collaborative disciplinary processes, paradisciplinary research occupies an interstitial position between disciplines that foregrounds their divergence as a collaborative means of generating knowledge. Reframed by the modularity principle, extradisciplinarity provides a model in which the digital humanities works outside traditional disciplinary regimes to access new modes of modular interoperability, and paradisciplinarity locates the digital humanities at the juncture between disciplines so that otherwise contradictory modularities—for instance, corporate and avant-garde—engage in counterpositional and collaborative critique. Where extradisciplinary collaboration operates beyond inherited institutional formations, and may lead at times to the transformation of the laboratory itself, paradisciplinary collaboration situates the modernist lab at historical moments of inter- modular disjunction and uneven development. Although the nomenclature of the lab has become more common in humanities contexts, most of these institutional formations bear closer resemblance to the collaboratory. Coined by William A. Wulf in a 1989 white paper, the collaboratory is defi ned as “a center without walls, in which the nation’s researchers can perform their research without regard to geographical location.” 52 “In operational terms,” as the authors of a 1993 National Research Council (NRC) report elaborate, “a collaboratory is a distributed computer system with networked laboratory instruments and data-gathering platforms; tools that enable a variety of collaborative activities; fi nancial and human resources for maintaining, evolving, and assisting in the use of computer-based facilities; and digital libraries that include tools for organizing, describing, and managing data.” 53 Given these operational defi nitions, it is not unreasonable to extrapolate from Latour and Woogar’s analysis to describe the collaboratory as “a system of literary inscription.” Current research initiatives have already claimed the concept for the digital humanities, including the Humanities, Arts, Science, and Technology Advanced Collaboratory (HASTAC), co-founded in 2008 by MODLABS 29

Cathy N. Davidson and David Theo Goldberg, and the Canadian Writing Research Collaboratory/Le Collaboratoire scientifi que des écrits du Canada (CWRC/CSEC), founded in 2010 and directed by Susan Brown. If at its inception the collaboratory excluded the humanities, these recent institutional formations instead foreground the collaborative and complex disciplinary potential of the concept itself. Most accounts of modernist laboratories and their infl uence on contemporary institutional formations located at the intersection of information technologies and the arts and design begin in the 1960s with the emergence of “studio-laboratories.” Among the earliest of these was Experiments in Art and Technology (EAT), founded by artist Robert Rauschenberg and Bell Labs physicist Billy Klüver in New York in 1966. As outlined in the proceedings from its 1969 conference, EAT created “an international network of experimental services and activities designed to catalyze the physical, economic and social conditions necessary for coop- eration between [sic ] artists, engineers, and scientists.” 54 EAT rejected the idea of a “single laboratory or information center” and positioned itself as a “matching agency” between artists and engineers to facilitate collaboration. 55 As an institutional formation, EAT functioned as a pre- digital collaboratory; rather than digital networking, it implemented at its earliest stages the mimeograph technology of its modernist periodical predecessors, publishing EAT News (1967–68) with detailed reports on its artist-engineer experiments, installations and collaborations. With Klüver’s affi liation with Bell Labs, EAT served as an extension of the mid- century corporate laboratory into the New York art world. 56 Also in the late 1960s, the Hungarian-American artist Gyorgy Kepes founded in 1967 the Center for Advanced Visual Studies (CAVS) at the Massachusetts Institute of Technology (MIT). A former instructor at Moholy-Nagy’s New Bauhaus in Chicago from 1937–43, Kepes began in 1946 teaching visual design at MIT, where he devoted much of his research to the interconnections between technology and art. 57 At CAVS, he revived a New Bauhaus ethos to bring about the “absorption of the new technology as an artistic medium” and “the interaction of artists, scientists, engineers, and industry.” 58 A one-time artist-fellow at CAVS, Muriel Cooper carried on its Bauhaus legacy at MIT as art director at the MIT Press and as founder of the Visible Language Workshop (VLW). David Reinfurt’s concise history of Cooper’s work at MIT places considerable emphasis on her engagement with the Bauhaus, calling attention to her innovative engagement with graphic design and mass production. 30 D. IRVINE

Her most durable and visible mark on the MIT Press was her 1964 design of its logo, a row of seven books, with the fourth pulled up and the fi fth pulled down, which not only produces “an abstracted form of the abbreviation ‘MITP’” but also “a clear barcode—as the products of mass production sit together in an orderly row, dematerialized into the pure information of a machine-readable graphic.” 59 Further, Cooper’s most famous book design is the MIT Press English translation of Hans Wingler’s Bauhaus: Weimar, Dessau, Berlin, Chicago (1969); as she explains, “My design approach always emphasized process over product, and what better place to explore this than in a tome on the Bauhaus, the seminal exploration of art and design in an industrial revolution.” 60 Combined with its expressed purpose to serve as an extension of the Bauhaus-Archiv in Darmstadt, the monumental proportions of the volume speak to Cooper’s work as the architect and builder of a mobile institutional structure bearing the signature markings of Bauhaus modular design. Beyond her work at MIT Press, Cooper went on to implement elements of Bauhaus pedagogy with her founding of the Visible Language Workshop in 1974. Vividly recalling Gropius’s “laboratory workshops,” she characterized the VLW as “a unique interdisciplinary graphics laboratory,” 61 one that brought a Bauhaus design ethos and production environment into the age of digital media. With the 1985 opening of the MIT Media Laboratory, under the direc- torship of Nicholas Negroponte, Cooper merged the VLW with the new lab. Reiterating the VLW’s homage to the Bauhaus—in a 1989 special issue of Design Quarterly on “Computers and Design”—she positioned the Media Lab in the same genealogy: “The Media Laboratory is a pioneering interdisciplinary center that is a response to the information revolution, much as the Bauhaus was a response to the industrial revolution.” 62 In her representation of collaborations between artists and scientists at the Media Lab, Cooper returns to the Bauhaus analogy: “The Media Lab’s greatest strength may prove to be the collision of the disparate disciplines and values represented there. …In much the same way, the meaning of the Bauhaus was in the confl ict between painters like Klee and Feininger, and technocrats like Moholy-Nagy.” 63 For a 1993 interview with Wired magazine, Negroponte confi rmed Cooper’s suggested lineage, saying, “You also have to realize how different we are from Bell Labs. …We are as much like the Bauhaus as a research lab. No photographers, fi lmmakers, or typographers go to work at Bell Labs in the same way that they do at the Media Lab.” 64 In his 1995 book Being Digital , a reworking of material originally published in Wired , Negroponte offers a different historical MODLABS 31 precedent: “As in 1863, when the Paris art establishment declined to let the Impressionists into its offi cial show, the founding faculty members of the Media Lab became a Salon des Refusés and had one of their own, in some cases too radical for their academic department, in some cases too extraneous to their department, and in one case with no department at all.” 65 From its inception, when Negroponte convinced MIT to build the Media Lab in 1979, it aimed to be an extradisciplinary formation, a principle endorsed by his claim that “the common bond” among its founding faculty “was not a discipline, but a belief that computers would dramatically alter and affect the quality of life through their ubiquity” (225). Here, he anticipates how the mass production of computers would lead to social and institutional transformation during the 1990s in which the avant-garde was no longer Media Lab but a globally dispersed network of hackers. 66 Already apparent in Cooper’s 1989 portrait of the Media Lab, Negroponte’s originary vision of its founders’ avant-garde extradisciplinarity migrates toward institutional paradisciplinarity. As early as the 1985 opening of the fi fty-million-dollar Weisner Building designed by architect I. M. Pei (fresh from his work on the Louvre), the Media Lab could no longer lay claim to its avant-garde status as a salon des refusés . Rather, appropriate to the aesthetic heritage of MIT’s studio-laboratories, Pei’s style paid tribute to the late-twentieth century’s institutionalization of visual and architectural modernism. 67 While the studio-laboratories of the 1960s, 1970s and 1980s were primarily oriented toward collaborations among engineers, scientists and practitioners in the visual arts and architectural design, the digital-humanities laboratories from the decades at the millennial turn brought experimentation with literary texts and critical theory into lab environments. Under the auspices of the Speculative Computing Laboratory (SpecLab) at the University of Virginia, the earliest literature-based digital laboratory was founded in the 1990s by Johanna Drucker, Jerome McGann, Bethany Nowviskie and others. SpecLab takes its inspiration directly from the methods employed by the modernist avant garde. SpecLab’s “working unit,” which they call Applied Research in ’Patacriticism, pays homage to the work of Alfred Jarry, who in 1893 coined the term ’pataphysics— itself a paronym that plays on metaphysics—to refer to “the science of imaginary solutions.” 68 While SpecLab has been ostensibly responsible for developing tools under the institutional rubric of NINES, Drucker differentiates between making “digital tools in humanities contexts,” which she attributes to the digital humanities, and making “humanities tools in digital 32 D. IRVINE contexts,” which she aligns with the ’patacritical project of “speculative computing.” 69 Drucker, principally concerned with preserving a space for experimental digital research in the humanities that aligns itself with the historical avant garde, positions speculative computing squarely beside the avant-gardism of other new-media modernisms. There are legitimate reasons why she correlates speculative computing with the cultural logic of the avant garde, not least of which is to formulate “a set of principles through which to push back on the cultural authority by which computational methods instrumentalize their effects across many disciplines.” 70 For Drucker, the digital humanities represent the instrumentalization of the humanities. Although she does not cite any of the prominent and infl uential twentieth-century critiques of instrumental reason by Horkheimer and Adorno, Heidegger, Gadamer and Habermas, these thinkers stand some- where behind her conception of speculative computing as a ’patacritical project. That is, she positions speculative computing in relation to the historical avant garde’s self-distantiation from an instrumentalized bour- geois society by disarticulating SpecLab from an instrumentalized digital humanities. What Drucker and her SpecLab colleagues neglect to interro- gate, however, is the institution of the modernist laboratory itself—a space of scientifi c and aesthetic experiment inhabited by practitioners of physics and ’pataphysics alike. A collaboratory by any other name, one of the most news-making digital-humanities institutions of the past decade has been the Stanford Literary Lab, founded in 2010 and directed by Franco Moretti (together with Matthew Jockers from 2010 to 2012). Anyone familiar with contemporary scholarship in the digital humanities will be aware of Moretti’s concept of “distant reading,” coined in his 2000 essay “Conjectures on World Literature.” When Moretti posited his original defi nition of distant reading, he placed it in opposition to “close reading (in all of its incarna- tions, from new criticism to deconstruction).” 71 In doing so, he aligned his method with “scientifi c work.” 72 Moretti claimed for distant reading the status of a “new ‘science’” in which “a new problem is pursued by a new method.” 73 Yet this critique of close reading is premised on a miscon- ception of its object, since I. A. Richards not only conceived of his “practical criticism” as an “experiment” analogous to scientifi c method 74 but also recognized the classroom as a “philosophic laboratory.” 75 Richards opens Principles of Literary Criticism , “A book is a machine to think with,” 76 yet few today would consider him a precursor to the algorithmic “reading machines” imagined by digital humanists. 77 Rather than a methodological MODLABS 33 break, then, distant reading is better conceptualized as a radical return to an experimental science of literary investigation. Even so, it is symptom- atic that digital humanists took up Moretti’s claim for a new science so readily: this is the recurrent trope of modernist innovation that pervades the discipline. Even as digital humanists adopted distant reading, Moretti separated himself from the term as he moved into his own phase of digital research at the Stanford Lab. Notably, there is no mention of digital technologies in his original formulation of distant reading, nor does it appear in the “Graphs, Maps, Trees” essay series (2003–04) or its revision collected under the same title in his infl uential 2005 book. Not until 2009 does Moretti raise the possibility that “in a few years, we will have a digital archive with the full texts of (almost) all novels ever published.” 78 More recently, in his 2011 pamphlet Network Theory, Plot Analysis , he speculates that “the rise of quantitative evidence” in literary studies now promises new results “because this time we have digital databases, and automated data retrieval.” 79 At once utopian and determinist, Moretti’s digital turn accentuates its investment in the technological progress of modernity. In particular, he points to the collaboration between the Cultural Observatory at Harvard and Google Labs on the culturomics project and their December 2010 launch of the Ngram Viewer as an exhibition of the ways in which “the width of the corpus and the speed of the search have increased beyond all expectations” 80 Echoing Moretti’s assertions about distant reading a decade earlier, the lead culturomics researchers Jean- Baptiste Michel and Erez Lieberman Aiden lay claim in their report to “a new type of evidence in the humanities” comparable to a “great cache” of fossilized bones “from which to construct the skeleton of a new science.” 81 That their culturomic analysis of “linguistic changes” as evidence of “cultural change” produces model “trajectories” 82 of linear progress is fi tting, since their methodology is specifi cally intended to redesign humanities research using the tools of a new science. What differentiates the Stanford Literary Lab from the Harvard Cultural Observatory and Google Labs? Both are models of institutional and paradisciplinary collaboration engaged in digital text analysis and visualization—one based in the humanities, 83 the other in the sciences—but the modes of interpretation are premised on different analytic concepts. The most prominent of these concepts is culture. For a project that purports to examine the textual records of entire cultures, it’s unclear why the Harvard Observatory makes no attempt to defi ne “culture,” let alone unpack its 34 D. IRVINE neologism “culturomics.” 84 The Stanford Lab’s immediate response after the Ngram Viewer’s launch was decidedly skeptical, as evidenced by codirector Matthew Jockers’s blog post, which calls into question both the conceptualization of culture employed by the Harvard Observatory and its reliance on data derived from Google Books: “To call these charts representations of ‘culture’ is, I think, a dangerous move. Even at this scale, the corpus is not representative of culture. … It probably represents the collection practices of major research libraries.” 85 While Jockers is doubt- less correct in his methodological critique that a corpus of books doesn’t constitute a culture—especially one that systematically excludes magazines, newspapers, printed ephemera, manuscripts, maps, artwork and so on—he sidesteps the concept of culturomics itself. According to Michel and Aiden, culturomics refers at once to the object of analysis (“linguistic and cultural phenomena” 86 ) and its methodology (“the application of high-throughput data collection to the study of human culture” 87 ), but the keyword itself is left undefi ned. Linguists have blogged about multiple interpretations, one concluding that the model is most likely the fi eld of biology and its allied sciences in which ‘-omics’ coinages proliferate (for example, genomics) but conceding that the “temptation to read—omics as connected with economics is a strong one” in the contemporary vernacular. 88 Whether deliberate or not, the portmanteau culturomics is itself a sign of linguistic modernity, an experimental coinage of a new science in a new language. Commonalities between the Harvard and Stanford labs reside in their mutual interests in genomics and economics. Moretti’s adaptation of models from evolutionary biology and genomics is examined at length in Alberto Piazza’s afterword to Graphs, Maps, Trees. 89 Jockers, too, is the progenitor of a neologistic methodology that he calls “macroanalysis,” which “is in general ways akin to the social-science of economics or, more specifi cally, macroeconomics.” 90 There are suggestive linkages between macroanalysis and the economic interpretation of culturomics. Similar to macroeconomic principles that focus on the performance and trends of whole economies rather than the microeconomics of individual agents and specifi c markets, culturomics constructs abstract models that represent short-term fl uctuations in cultures and long-term cultural trajectories refl ected in changes to linguistic phenomena across decades and centuries. For instance, Michel and Aiden bring this approach to an analysis of the “history of economics,” tracing the linguistic shift from older concepts such as “banking” during the Great Depression to a MODLABS 35

“new economic vocabulary” (“recession,” “GDP,” “the economy”) that “entered everyday discourse” at mid-century. 91 Although they discuss the emergence into the vernacular of concepts derived from modern macroeconomics, the computational method that they practice is consistent with the principles of “modernist” economics. 92 That is, their text analysis and visualization constitute a model of the cultural trajectory that leads to the formation of culturomics itself.

MODULARITY To take the longer historical view of these modernist and new-media modernist institutions—one that spans some version of the long twentieth century—we may begin to see how their recursive formations exhibit what Manovich calls the “modularity principle,” whose para- and extradisciplinary dispersion has informed more than a century of laboratory design and research in the arts, sciences and industry. The laboratory’s progress through this long century has been one loaded with disciplinary obstacles, not least of which are the contradictions between industrial and cultural conceptions of modularity. Manovich attempts to address these contradictions by theorizing it in terms of the uneven development of industrial and cultural modes of production under the economic regime of modernity. This contradiction between corporate and avant-garde modularities speaks to the ways in which the laboratory typifi es the asynchrony of industrial and cultural modernities. While the factory division of labor can be seen in the studio production of animated fi lms and video games, this kind of cultural labor is qualitatively and quantitatively different from the “systematic character of industrial standardization” achieved by Henry Ford. 93 Rather than anticipate the time when cultural modularity catches up to industrial modularity, the uneven development of modularities is such that their respective institutional formations are perpetually out of sync. Cultural modularity required the development of industrial modularity in order to constitute itself (as Adorno argued for modernist art), 94 but as a mode of negation and critique; both owe their existence to the historical conjuncture of industrial capitalism, and neither can be understood without reference to the other. Consistent with Manovich’s analysis of new-media objects, both the Harvard Observatory’s and the Stanford Lab’s experiments are predicated upon the modularity principle. Where Manovich extends the “fractal structure of new media” to the modular organization of the Internet as a 36 D. IRVINE whole, 95 culturomics and macroanalysis investigate a subset of that modularized structure: the online database. Both approaches redistribute the text’s linguistic code into a corpus consisting of statistical tables—which can in turn be ported into different types of databases—that effectively transform the bibliographic object into standardized modular units made interchangeable with other texts. With its transformation into statistical tables, modularization strips the text of its bibliographic code and reifi es its linguistic code. The reifi cation of linguistic code renders the bibliographic object as a raw resource, a machine-readable textbase abstracted from its human-readable commodity form and its material history in the social life of exchange and consumption. This creates the conditions necessary for the mass production of textuality: data visualization produced on demand, customizations “on the fl y,” “just in time” deliveries made possible by the postindustrial logic of digital modularity. 96 Aggregated as statistical abstractions, linguistic code is remediated by the culturomic or macroanalytic interface as the graphical inscription of a new modular artefact. Rather than reinscriptions of print media, these are new inscriptions in a new medium. This is the realization of what Fredric Jameson calls the “media paradox” of modularity, “where intensifi ed change is enabled by standardization itself.” 97 This, too, is an instantiation of new-media modernism: freed from medium specifi city, modularized data is released from the bibliographic codes that anchor the text to its prior materiality and cultural modality, not to be recombined in the same repeated form (as under industrial modernity) but reconfi gured in endlessly variable patterns. 98 One of the more recent laboratory environments to emerge comes out of a pioneering institution in digital modernist studies. The Modernist Journals Project (MJP), which is based and jointly operated at Brown University and the University of Tulsa, was founded in 1995 as “a website of digital editions of periodicals connected to the rise of modernism in the English-speaking world.” 99 Where the main MJP site is oriented toward the production of reading and search interfaces for its digital editions, the MJP Lab is organized around the visualization of the data “generated over the course of digitizing magazines from the early 20th century.” 100 If the main site is directed toward the creation of a resource that backs the MJP’s claims about the institutionalization of an Anglo-American modernist literary-historical tradition, the lab site looks toward the expansion of that institutional formation through the expressive modes of visual abstraction (timelines, tables, charts, trees, sunbursts) and new inscriptions of modularized data in graphical interfaces. The trope of innovation that surfaces MODLABS 37 on the lab site foregrounds the ways in which the MJP Lab is itself a new institutional formation and renders the textual data of modernism as new digital objects. Upon entering the MJP Lab, for instance, we are welcomed into a space that not only announces itself as innovative but also professes to facilitate cutting-edge research: “The site is experimental,” we are cautioned, “but it’s also dedicated to experimentation―playing with the MJP data, and drawing new patterns and knowledge.” 101 This is new media’s modernism, one that aligns contemporary digital experimentation with the aesthetic innovations of the modernists themselves. The MJP Lab also emphasizes digital humanities pedagogy. This is the digital companion to the MJP’s teaching pages and instructional wiki, which are mainly oriented toward reading magazines principally as print objects and only supplementally in the remediated format of the digital repository. Informed by instructions from MJP director Robert Scholes and technical advisor Clifford Wulfman in their chapter “How to Study a Modern Magazine,” 102 the website includes its own pedagogical guide, “How to Read a Magazine.” 103 This online manual advises how to read a magazine in the context of print culture, how to parse a magazine into its component parts, how to devise research projects and how to read a magazine as a material, printed object. Readers are briefl y instructed to familiar- ize themselves with the MJP search engine, but there are no instructions on how to perform customized searches or how to interpret search results. How-to readers are left with the impression that one doesn’t actually read magazines by using a search engine, or that database searching isn’t itself a mode of reading, but a supplement to reading practices oriented around the magazine’s materiality. With the MJP Lab, which “may be viewed as a supplement to the MJP’s search pages,” 104 the site reorients its pedagogy toward digital reading practices. In order to facilitate its pedagogy, the lab provides open access to a portion of its data and metadata, produces sample visualizations and proposes ideas for further visualizations. Available for download from the MJP Lab’s Sourceforge page, 105 its datasets include Metadata Object Description Schema (MODS) catalogue records, tran- scriptions with Text Encoding Initiative (TEI) markup, and plain-text fi les (derived from the MODS records) for six journals. Complete with instructions on how to use digital tools, perform quantitative analysis and generate visualizations, these datasets are not merely supplemental to the digital editions of the magazines; they are integral to the development of modes of reading remediated modernist magazines as digital objects and to the inauguration of an experimentally practiced pedagogy that foregrounds 38 D. IRVINE the materiality of digital media. Even prior to the creation of its lab and its experimentation with modularized datasets, the MJP’s digitization and markup of modernist magazines remediated their modularity. After all, the modernist magazine is already modular in design: its structural elements were conventionalized and standardized in such a way that allows the MJP to compile “an anatomical study of a magazine.” 106 As repositories of aesthetic experimentation, these magazines were at once extensions of the avant-garde’s laboratories and, as such, inscriptions of modular design. Infl uenced by principles of modular design incubated in modernist laboratory environments, avant-garde periodicals of the 1920s and 1930s were paper prototypes of modular labs, collapsible and portable storehouses of experimental research and art. Remediation of the modernist magazine is an act of “media translation” 107 in which the MJP’s digitization, transcription and markup workfl ow transcodes the modularity of the analog source into modular data and metadata of the digital object. With the encoding of its magazines according to the TEI Guidelines and MODS, the MJP repository implements both modular design through its markup language and metadata. As a modular and extensible markup language, the TEI in the most recent iteration (P5) consists of “a core module with essential common elements, and considers all further tagsets as additional modules which can be combined, modifi ed, and trimmed to suit the user’s needs.” 108 As the Library of Congress cataloging standard, MODS is predicated on the assumption that “[m]etadata modularity is a key organizing principle. … In a modular metadata world, data elements from different schemas as well as vocabularies and other building blocks can be combined in a syntactically and semantically interoperable way.” 109 At once highly modular and strongly hierarchical, XML-based architectures such as the TEI and MODS pose considerable obstacles to reading the non-hierarchical and overlapping organization of the heterogenous parts of a magazine. According to MJP director Sean Latham, the magazine is an “ergodic” structure, which is to say that every constitutive part is equally representative of the whole. 110 This poses a problem for magazines marked up in TEI, in so far as any non-sequential reading practice typical of magazines (which, according to the MJP, includes skimming, browsing, surveying, grazing, sampling, rooting around, drifting or circling back) bypasses the hierarchical routes of XML-based encoding schemas. 111 Dating from the late 1920s, Tschichold’s new typography provides a characteristically modernist model of ergodic reading in which he posits that the “principles MODLABS 39 of asymmetry” release the reader from the linear strictures of “central-axis typography” as “an expression of our own movement and that of modern life;” as Tschichold explains, “Every part of a text relates to every other part by a defi nite, logical relationship of emphasis and value, predetermined by content. … It is true that we usually read from top left to bottom right— but this is not law.” 112 The correlation of the new typography’s principles of modular design with his mode of ergodic, nonlinear and spatialized textual navigation produces an historical model of constructivist reading in which the reader becomes a producer of meaning. Ergodics exposes an ideological incompatibility between analog and digital modularities, or what Richards called in the execution of his own reading experiment “the record of a piece of fi eld-work in comparative ideology” 113 —that is, if ergodic reading is a constructivist mode of cultural production, nonergodic reading is a consumerist mode of cultural consumption. Historically speaking, the modular XML architecture of the TEI was designed to address “a real concern that the entrepreneurial forces which (then as now) drive information technology forward would impede such integration by the proliferation of mutually incompatible standards.” 114 If the analog magazine is at once an ergodic structure and expression of modular design, its digital transcoding produces a nonergodic translation, one that reproduces a modular structure that prescribes reading as consumption of a linear and hierarchical program language rather than traversing the multidirectionality of a modular affordance. In other words, TEI markup implements its standardized language and modular architecture in the service of an instrumental logic originally intended as a business solution aimed at economic effi ciencies, not at modernism’s cultural modularities. Ergodic reading may therefore provide a mode of critique of the magazine’s digitally transcoded modularity, a constructivist reading practice that requires the existence of a programmatic structure against which it enacts its “nontrivial” and “nonlinear” traversal of the text. 115 To read ergodically may further counteract what Suzanne W. Churchill and Adam McKible call “critical practices of strip-mining” 116 magazines for textual content—or, translated to digital practice, text-mining magazines whose datasets are scrubbed of typographic codes, graphic design, layout grids and advertising spreads and either rendered as modularized plain text or abstracted as metadata. Given the MJP directors’ documented critique of periodical studies in which texts and other paratextual materials are stripped away from their original contexts, 117 the MJP Lab’s practice of “strip-mining” remediated 40 D. IRVINE magazines for their data and metadata raises questions about the methods used for its sample visualizations. This is not to say that the MJP directors are by any means at odds with the ideological operations of their own lab, but rather that the lab foregrounds the articulation of avant-garde, scientifi c and corporate modernities in a way coincident with the histories of the early twentieth-century magazine and cultural modernism. To prepare the way toward these kinds of digital laboratory experiments, Scholes and Wulfman’s rethink the medium in which we study them as cultural objects, a shift from genre to database: “we need to develop a language of magazines ,” 118 they implore, showing how media translation necessitates not only understanding how to catalogue magazines as bibliographic objects in a database, but also knowing how to encode magazines in various markup languages that enable database queries that abstract from their modular design to render new modularities and reading interfaces. To seed the databases required to produce this kind of innovative research cannot be the solitary task of the individual scholar; if we are to develop suffi cient competencies in reading and encoding the “language of magazines,” they must issue from the collaborative research environment of the laboratory. In their prescient 2005 article “The Rise of Periodical Studies,” Latham and his codirector Robert Scholes not only anticipate the emergence of the MJP Lab but articulate the para- and extra-disciplinary institutional formation of the modernist laboratory:

To develop such research models, we might look profi tably to the sciences, where laboratories are often structured around precisely this kind of intellectual challenge. In such settings, large experiments are broken down into component parts, and particular sets of skills and expertise are brought to bear on them. The fi nal product is then eventually integrated and published either in whole or in logical parts. … In applying this model to periodical studies, we might therefore consider the creation of humanities labs: similarly collaborative networks of researchers and institutions that lend their collective expertise to textual objects that would otherwise overwhelm single scholars. 119

To look at the large number of contributors to the experiments already conducted at the MJP Lab, it is apparent that the networks that Scholes and Latham originally proposed in 2005 have found their realization in their inter-institutional collaboration. The humanities lab they propose here is highly modularized, and their modes of collaborative research and publication closely resemble the science labs that Latour and Woolgar MODLABS 41 document. All the same, I wonder why they might have overlooked models even closer to hand, specifi cally the laboratories of the historical avant garde and their mobile extensions—that is, the collapsible labs that the MJP takes as its primary objects of examination and experimentation. Were they to take these institutional and periodical models into consideration, Scholes and Latham might have envisioned the laboratory itself as a site of experimental reading and inscription, an extension of the models of visual and architectural modularity expressed at once in avant-garde periodicals and in data visualizations generated by digital tools.

NOTES 1. Nicholas Negroponte, quoted in Thomas A. Bass, “Being Nicholas,” Wired 3.11 (1993). Reprinted in HotWired , http://archives.obs-us. com/obs/english/books/nn/bd1101bn.htm. 2. For a history of digital modernist studies in Canada and the development of a digital laboratory environment by the Editing Modernism in Canada project, see Dean Irvine, Bart Vautour and Vanessa Lent, “Introduction,” in Making Canada New: Editing, Modernism, and New Media , eds. Dean Irvine, Bart Vautour and Vanessa Lent (Toronto: University of Toronto Press, 2016). See also Editing Modernism in Canada, Modernist Commons , http://modernistcommons.ca . For interim reports on experiments in progress involving large modernist datasets, see posts by members of Jentery Sayers’ Maker Lab in the Humanities at the University of Victoria, http://maker.uvic.ca /. 3. Jay David Bolter and Richard Grusin, Remediation: Understanding New Media (Cambridge, MA: MIT Press, 1999), 54. 4. See Lev Manovich, “Avant-Garde as Software,” in Ostranenie , edited by Stephen Kovats (Frankfurt and New York: Campus Verlag, 1999), “Generation Flash,” RHIZOME http://archive.rhizome.org/ digest/?msg=00014 2002, The Language of New Media (Cambridge, MA: MIT Press, 2001), “New Media from Borges to HTML,” in The New Media Reader , eds. Noah Wardrip-Fruin and Nick Montfort (Cambridge, MA: MIT Press, 2003), 13–25; Mark Goble, Beautiful Circuits: Modernism and the Mediated Life (New York: Columbia University Press, 2010); Michael North, Camera Works: Photography and the Twentieth-Century Word (New York: Oxford University Press, 2005); Jessica Pressman, Digital Modernism: Making It New in New Media (New York: Oxford University Press, 2013). 5. Matthew G. Kirschenbaum, Mechanisms: New Media and the Forensic Imagination . Cambridge: MIT Press, 2008), 10–15. 42 D. IRVINE

6. Lisa Gitelman, Always Already New: Media, History, and the Data of Culture (Cambridge, MA: MIT Press, 2006), 145. 7. Lev Manovich, The Language of New Media (Cambridge, MA: MIT Press, 2001), 15. 8. Hugo Münsterberg, “The New Psychology and Harvard’s Equipment for Teaching It,” Harvard Graduate Magazine 1.2 (1893), 206. 9. Hugo Münsterberg, Laboratory of Harvard University (Cambridge, MA: Harvard, 1893). 10. Friedrich A. Kittler, Gramophone, Film, Typewriter , trans. Geoffrey Winthrop-Young and Michael Wutz (Stanford: Stanford University Press, 1999), 159. 11. Giuliana Bruno, “Film, Aesthetics, Science: Hugo Münsterberg’s Laboratory of Moving Images,” Grey Room 36 (2009), 90. 12. Bruno, “Film, Aesthetics, Science,” 109. 13. Hugo Münsterberg, The Photoplay: A Psychological Study (New York: D. Appleton, 1916), 28. 14. Katherine Hoffman, Stieglitz: A Beginning of Light (New Haven: Yale University Press, 2004), 303. 15. Hoffman, Stieglitz , 230, 44, 202. 16. Hoffman, Stieglitz , 245. 17. Alan Trachtenberg, Reading American Photographs: Images as History Mathew Brady to Walker Evans (New York: Hill and Wang, 1989), 167, 168. 18. Walter Benjamin quoted in Marian Gough, The Artist as Producer: Russian Constructivism in Revolution (Berkeley and Los Angeles: University of California Press, 2005), 1. 19. Gough, The Artist as Producer , 160. 20. Judith Collins, The Omega Workshops (University of Chicago Press, 1984), 31, 47, 53. 21. Roger Fry, “Art and the State” in Art and the Market: Roger Fry and Commerce in Art , ed. Craufurd D. Goodwin (Ann Arbor: University of Michigan Press), 198. 22. Fry, “Art and the State,” 198. Fry elaborated on his laboratory proposals in his 1932 memorandum, “Art and Industry,” in Art and the Market: Roger Fry and Commerce in Art , ed. Craufurd D. Goodwin (Ann Arbor: University of Michigan Press), 209–12, 214–16. 23. Christina Kiaer, Imagine No Possessions: The Socialist Objects of Russian Constructivism (Cambridge, MA: MIT Press, 2005), 37. 24. Walter Gropius, “Principles of Bauhaus Production [Dessau],” in Programs and Manifestoes on 20th-century Architecture , ed. Ulrich Conrads, trans. Michael Bullock (Cambridge, MA: MIT Press, 1970), 96. 25. Shannan Clark, “When Modernism Was Still Radical: The Design Laboratory and the Cultural Politics of Depression-Era America,” American Studies 50. 3–4 (Fall-Winter 2009), 35–6. MODLABS 43

26. Clark, “When Modernism Was Still Radical,” 35–6. 27. Sybil Gordon Kantor, Alfred H. Barr, Jr. and the Intellectual Origins of the Museum of Modern Art (Cambridge, MA: MIT Press, 2002), 152. 28. Alfred H. Barr, Jr., “‘Art in Our Time’ The Plan of the Exhibition,” in Art in Our Time (New York: Museum of Modern Art, 1939), 15. 29. “Bauhaus 1919–1933: Workshops for Modernity,” Museum of Modern Art , http://www.moma.org/calendar/exhibitions/303?locale=en 30. For an account of Scandinavian adaptations of Bauhaus principles to digital technologies, see Thomas Binder, Jonas Löwgren, Lone Malmborg, eds. (Re)Searching the Digital Bauhaus (London: Springer-Verlag, 2009). 31. David Bate, Photography and Surrealism: Sexuality, Colonialism and Social Dissent (London: I. B. Tauris, 2004), 82. 32. André Breton, “Max Ernst,” in The Lost Steps , trans. Mark Polizzotti (Lincoln and London: University of Nebraska Press, 1996), 60. 33. Ben Highmore, Everyday Life and Cultural Theory: An Introduction (London and New York: Routledge, 2002), 82. 34. Highmore, Everyday Life , 83. 35. Highmore, Everyday Life , 83. 36. Highmore, Everyday Life , 83. 37. Mass-Observation, Mass-Observation (London: Frederick Muller, 1937), 10. 38. Tom Harrisson, Humphrey Jennings and Charles Madge, “Anthropology at Home,” The New Statesman and Nation , 30 January 1937, 155. 39. Highmore, Everyday Life , 84. 40. Mass-Observation, The Pub and the People: A Worktown Study (London: Gollancz), 328. 41. Mass Observation Online, http://www.massobservation.amdigital.co.uk 42. Michael J. Hoffman, “Gertrude Stein in the Psychology Laboratory,” American Quarterly 17.1 (Spring 1965), 127. See also Steven Meyer, Irresistible Dictation: Gertrude Stein and the Correlations of Writing and Science (Stanford: Stanford University Press, 2003). 43. Gertrude Stein, Selected Writings of Gertrude Stein , ed. Carl Van Vechten (New York: Vintage Books, 1990), 73. 44. See Tanya E. Clement, “‘A thing not beginning and not ending’: Using Digital Tools to Distant-Read Gertrude Stein’s The Making of Americans ,” Literary and Linguistic Computing 23.3 (2008): 361–81 and Tanya Clement, et al., “Distant Listening to Gertrude Stein’s ‘Melanctha’: Using Similarity Analysis in a Discovery Paradigm to Analyze Prosody and Author Infl uence,” Literary and Linguistic Computing 28.4 (2013): 582–602. 44 D. IRVINE

45. Bruno Latour and Steve Woogar, Laboratory Life: The Construction of Scientifi c Facts , 2nd edn (Princeton: Princeton University Press, 1986), 52. 46. Latour and Woogar, Laboratory Life , 53. 47. Amy Earhart, “Challenging Gaps: Redesigning Collaboration in the Digital Humanities” in The American Literature Scholar in the Digital Age, eds. Amy Earhart and Andrew Jewell (Ann Arbor: University of Michigan Press, 2011), 28. 48. Earhart, “Challenging Gaps,” 31. Specifi cally, Earhart refers to the model in which “the science lab is often institutionally fi nanced at start-up” and that this “base funding allows scientists to purchase equipment and fund personnel necessary to the development of a project prototype that can then be used to secure external funds.” An alternative for digital humanists without any dedicated infrastructural support is the example of “science and engineering faculty [who] come to institutions that do not have a center of expertise in their area” and who “combine start-up funds with collaborative work” and “cabl[e] together equipment and expertise” that allow them to secure external funding (33). There are still other models she leaves unmentioned, including inter-institutional collaborations among digital humanists, not all of whom have local access to digital infrastructure but who participate in distributed networks of intellectual and economic exchange. 49. Earhart, “Challenging Gaps,” 33. 50. Matthias Wienroth, “Disciplinarity and Research Identity in Nanoscale Science and Technologies,” in Size Matters: Ethical, Legal and Social Aspects of Nanobiotechnology and Nanomedicine , eds.Johann S. Ach and Christian Weidemann (LIT Verlag Münster, 2009), 165–6. 51. Weinroth, “Disciplinarity and Research Identity,” 166. 52. Quoted in National Research Council, National Collaboratories: Applying Information Technology for Scientifi c Research (Washington: National Academy Press, 1993), vii. 53. National Research Council, 7. 54. Quoted in National Research Council, Beyond Productivity: Information Technology, Innovation, and Creativity (Washington: National Academies Press, 2003), 122. 55. Billy Klüver and Robert Rauschenberg, Editorial, EAT News 1.2 (June 1967), 2. 56. For an account of EAT collaborations with Bell Labs and, specifi cally, computer art projects facilitated by Bell Labs between 1961 and 1972, see Zabet Patterson, Peripheral Vision: Bell Labs, the S-C 4020, and the Origins of Computer Art (Cambridge, MA: MIT Press, 2015). 57. On Kepes’s affi liations with the Bauhaus and New Bauhaus and his early engagement with cybernetics, most notably in The New Landscape of Art MODLABS 45

and Science (1956), see Reinhold Martin, The Organizational Complex: Architecture, Media, and Corporate Space (Cambridge, MA: MIT Press, 2003), 42–79. 58. Quoted in Elizabeth Finch, “A Brief History of the Center for Advanced Visual Studies, Massachusetts Institute of Technology,” Center for Advanced Visual Studies , http://act.mit.edu/cavs/history 59. David Reinfurt, “An Accidental Archive at the Center for Advanced Visual Studies,” http://www.dextersinister.org/MEDIA/PDF/ Thisstandsasasketchforthefuture.pdf . 60. Quoted in Reinfurt, “An Accidental Archive,” 10. 61. Quoted in Reinfurt, “An Accidental Archive,” 10. 62. Muriel Cooper, “Computers and Design,” Design Quarterly 142 (January 1989), 18. 63. Cooper, “Computers and Design,” 20. 64. Nicholas Negroponte, quoted in Thomas A. Bass, “Being Nicholas,” Wired 3.11 (1993). Reprinted in HotWired , http://archives.obs-us. com/obs/english/books/nn/bd1101bn.htm. 65. Nicholas Negroponte, Being Digital (New York: Knopf, 1995), 225. 66. Negroponte, Being Digital , 226. 67. Michael T. Cannell, I. M. Pei: Mandarin of Modernism (New York: Carol Southern Books, 1995); Carter Wiseman, I. M. Pei: A Profi le in American Architecture , rev. edn (New York: H. N. Abrams, 2001). 68. Alfred Jarry, Gestes et opinions du Docteur Faustroll, ’pataphysicien in Oeuvres Complètes , vol. 2, edited by Michel Arrivé (Paris: Gallimard, 1972), 7. 69. Johanna Drucker, SpecLab: Digital Aesthetics and Projects in Speculative Computing (Chicago: University of Chicago Press, 2009), 25, original italics. 70. Drucker, SpecLab , 5. 71. Franco Moretti, “Conjectures on World Literature,” New Left Review 1 (Jan.–Feb. 2000), 57. 72. Moretti, “Conjectures on World Literature,” 54. 73. Max Weber, quoted in Moretti, “Conjectures on World Literature,” 55. 74. I. A. Richards, Practical Criticism : A Study of Literary Judgement , rev. edn (London: Routledge and Kegan Paul, 1952), vii. 75. I. A. Richards, Speculative Instruments (London: Routledge and Kegan Paul, 1955), 104. 76. I. A. Richards, Principles of Literary Criticism (New York: Routledge, 2001), vii. 77. Stephen Ramsay, Reading Machines: Towards and Algorithmic Criticism (Urbana: University of Illinois Press, 2011). 46 D. IRVINE

78. Franco Moretti, “Style, Inc. Refl ections on Seven Thousand Titles (British Novels, 1740–1850),” Critical Inquiry 36.1 (Autumn 2009), 134. 79. Franco Moretti, Network Theory, Plot Analysis , Stanford Literary Lab, Pamphlet 2 (1 May 2011), 2, http://litlab.stanford.edu/LiteraryLab Pamphlet2.pdf . 80. Moretti, Network Theory , 2. 81. Jean-Baptiste Michel and Erez Lieberman Aiden, et al., “Quantitative Analysis of Culture Using Millions of Digitized Books,” Science 331.6014 (December 2010), 182. 82. Michel and Aiden, “Quantitative Analysis of Culture,” 182. 83. In characterizing its own projects, the Stanford Lab states that it “discusses, designs, and pursues literary research of a digital and quantitative nature. … Ideally, research will take the form of a genuine ‘experiment’. … At the Lab, all research is collaborative (even though some outcomes may end up having a single author). We hold regular group meetings to evaluate the progress of a specifi c experiment, the status of existing hypotheses, and future research developments.” Stanford Literary Lab, “About,” http://litlab.stanford.edu /. 84. Michel and Aiden have recently expanded on the initial fi ndings of their 2010 article in their recent book, Uncharted: Big Data as a Lens on Human Culture (New York: Riverhead Books, 2013). 85. Matthew L. Jockers, “Unigrams, and Bigrams, and Trigrams, Oh My,” http://www.matthewjockers.net/2010/12/22/ unigrams-and-bigrams-and-trigrams-oh-my/ . 86. Michel and Aiden, “Quantitative Analysis of Culture,” 1. 87. Michel and Aiden, “Quantitative Analysis of Culture,” 5. 88. Ben Zimmer, “Buzzword Watch: ‘Culturomics’ and ‘Ngram.’” http:// www.visualthesaurus.com/cm/wordroutes/2551/ . 89. Alberto Piazza, “Evolution at Close Range,” in Franco Moretti, Graphs, Maps, Trees: Abstract Models for Literary History (London: Verso, 2005), 95–133. 90. Matthew L. Jockers, “On Distant Reading and Macroanalysis.” http:// www.matthewjockers.net/2011/07/01/on-distant-reading-and- macroanalysis/ . For an expanded defi nition and implementation of macroanalytic methods, see Matthew L. Jockers, Macroanalysis: Digital Methods and Literary History (Chicago: University of Illinois Press, 2013). 91. Jean-Baptiste Michel and Erez Lieberman Aiden, et al., “Supporting Online Material for Quantitative Analysis of Culture Using Millions of Digitized Books,” Science 331.6014 (December 2010), 73, http://www. sciencemag.org/content/early/2010/12/15/science.1199644/suppl/ DC1 . MODLABS 47

92. For a history and critique of modernist economics, see Deirdre McClosky, The Rhetoric of Economics (Madison: University of Madison Press, 1985). 93. Manovich, “Remixability,” 50. 94. Theodor W. Adorno, Aesthetic Theory , ed.and trans.Robert Hullot- Kentor (Minneapolis: University of Minnesota Press, 1997), 34. 95. Manovich, The Language of New Media, 31. 96. Manovich, The Language of New Media, 36. 97. Fredric Jameson, Postmodernism, or the Cultural Logic of Late Capitalism (Durham: Duke University Press, 1990), 58. 98. Manovich, The Language of New Media, 36; Lev Manovich, “Remixability” in After the Digital Divide? German Aesthetic Theory in the Age of New Media , edited by Lutz Peter Koepnick (Rochester, NY: Camden House, 2009), 46. 99. Modernist Journals Project, “About the MJP,” http://modjourn.org/ about.html . 100. Modernist Journals Project, “The MJP Lab,” http://cds.library.brown. edu/projects/mjplab/ . 101. Modernist Journals Project, “The MJP Lab” 102. Robert Scholes and Clifford Wulfman, “How to Study a Modern Magazine” in Modernism in the Magazines: An Introduction (New Haven: Yale University Press, 2010), 143–67. 103. Modernist Journals Project, “How to Read a Magazine,” http://www. modjourn.org/teaching/introduction/intro3_howto.html . 104. Modernist Journals Project, “The MJP Lab”. 105. Modernist Journals Project, “MJP Lab,” http://sourceforge.net/projects/mjplab/ . 106. Modernist Journals Project, “How to Read a Magazine”. 107. N. Katherine Hayles, My Mother Was a Computer: Digital Subjects and Literary Texts (Chicago: University of Chicago Press, 2005), 89. 108. The Text Encoding Initiative, “Introducing the Guidelines,” http:// www.tei-c.org/Support/Learn/intro.xml 109. Erik Duval, Wayne Hodgins, Stuart Sutton, and Stuart L. Weibel, “Metadata Principles and Practicalities,” D-Lib Magazine (April 2002), http://www.dlib.org/dlib/april02/weibel/04weibel.html . 110. Sean Latham, “Unpacking My Digital Library,” in Making Canada New: Editing, Modernism, and New Media , eds. Dean Irvine, Bart Vautour and Vanessa Lent (Toronto: University of Toronto Press (forthcoming 2016). Latham takes his defi nition of ergodic from Espen J. Aarseth’s Cybertext: Perspectives on Ergodic Literature (Baltimore: Johns Hopkins University Press, 1997), “a term appropriated from physics that derives from the Greek words ergon and hodos , meaning ‘work’ and ‘path.’” Aarseth’s coinage of the term “ergodic literature” refers to works that require “a 48 D. IRVINE

nontrivial effort … to traverse the text” (1). He further elaborates that the “ergodic work of art is one that in a material sense includes the rules for its own use, a work that has certain requirements built in that automatically distinguishes between successful and unsuccessful users” (179). Ergodic texts may exist in either print or electronic media. 111. Latham, “Unpacking My Digital Library”. 112. Jan. Tschichold, The New Typography: A Handbook for Modern Designers , trans. Ruari McLean (Berkeley: University of California Press, 1995), 67. 113. I. A. Richards, Practical Criticism , 6. 114. The Text Encoding Initiative, “TEI: History,” http://www.tei-c.org/ About/history.xml . 115. Aarseth, Cybertext , 1–2. 116. Suzanne W. Churchill and Adam McKible, “Introduction,” Little Magazines and Modernism: New Approaches (London: Ashgate, 2008), 4. 117. Sean Latham and Robert Scholes, “The Rise of Periodical Studies,” PMLA 121.2 (2006), 520–21. 118. Scholes and Wulfman, “How to Study a Modern Magazine,” 70; original italics. 119. Latham and Scholes, “The Rise of Periodical Studies,” 530. CHAPTER 3

Modeling Modernist Dialogism: Close Reading with Big Data

Adam Hammond , Julian Brooke , and Graeme Hirst

Of the many bold pronouncements in Matthew Jockers’s groundbreaking Macroanalysis (2013), perhaps the boldest is his claim that the advent of computational “distant reading” will make close reading obsolete as a method for investigating literary history. Jockers argues that the development of massive digital literary corpora has placed literary historians in a position in which they no longer need to rely on “partial sample” close readings, but can instead perform “investigations at a scale that reaches […] a point of being comprehensive.” 1 Jockers writes:

This work was fi nancially supported in part by the Natural Sciences and Engineering Research Council of Canada and the Social Sciences and Humanities Research Council of Canada.

A. Hammond ( ) San Diego State University, San Diego, CA , USA email: [email protected] J. Brooke University of Melbourne, Melbourne, VIC , Australia G. Hirst University of Toronto, Toronto, ON , Canada

Science has welcomed big data and scaled its methods accordingly. With a huge amount of digital-textual data, we must do the same. Close reading is not only impractical as a means of evidence gathering in the digital library, but big data render it totally inappropriate as a method of studying literary history. 2

Before Jockers, even the most passionate defenders of computational literary analysis tended to stop well short of the language of the “totally inappropriate.” Though Susan Hockey champions the “rigor and systematic unambiguous procedural methodologies” of computational analysis against the “serendipitous” procedure of close reading 3 —and although Julia Flanders argues that the computer is not just a “substantiator” of close readings but indeed “a device that extends the range of our perceptions to phenomena too minutely disseminated for our ordinary reading” 4 —neither goes so far as to claim that computational analysis could, or should, replace close reading. Stephen Ramsay and Tanya Clement, digital humanists responsible for perhaps the most illuminating digital work to date on modernist texts, are more careful still in insisting on a symbiotic relationship between close reading and computational analysis. In his work on Virginia Woolf’s The Waves , Ramsay presents quantitative analysis as a digital-age method of achieving the Russian Formalists’ goal of ostranenie —a making-strange that clears the path for renewed close reading. 5 Clement likewise presents her computational work on Gertrude Stein’s The Making of Americans as a way to “defamiliarize texts, making them unrecognizable in a way (putting them at a distance) that helps scholars identify features they might not otherwise have seen.” 6 For Clement and Ramsay, computational analysis can thrive only in an ecosystem of close reading, since its proper role is to enrich existing close readings and to prompt further ones. Despite Clement and Ramsay’s shared vision of a mutual coopera- tion between human and computational reading—and although Clement uses the term “distance”—what they advocate is not a combination of close and distant reading per se. While both seek literary insight through quantitative analysis, neither employs big data approaches or works at Jockers’s “comprehensive” scale; instead, they limit their analyses to single texts (a scale at which close reading remains putatively reasonable and uncontroversially “appropriate”) and to relatively simple statistical calculations of surface features, using metrics that make no use of features derived from large-scale text collections. Their reluctance to wade into this variety of “distant reading” is perhaps explained by the disappointing results of much scholarship that adopts this perspective. Jockers’s work in MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 51

Macroanalysis frequently fails to provide genuine critical insights; most often, he merely shows that his tools are working, employing them to support long-held critical commonplaces. The method Jockers describes in Macroanalysis for reliably predicting an author’s gender based on thematic topic modeling data—female authorship, he fi nds, is predicted by engagement with stereotypically female themes such as “fashion” and “children”—does little, on its own, to enrich our understanding of literary history. 7 More disquieting still are the conclusions of a paper by Ryan Heuser and Long Le-Khac, former colleagues of Jockers in the Stanford Literary Lab. In their “Quantitative History of 2,958 Nineteenth-Century British Novels,” they describe their discovery of two groups of words that exhibit exactly opposite frequency trends across their corpus: “abstract value” words such as “conduct” and “envy,” which decrease steadily in the period; and “hard seed” words including action verbs, body parts, colors and numbers, which collectively increase in precisely inverse relation. The authors explain these shifts in terms of a turn in narrative style from telling to showing—a movement toward literary realism predicated on empirical description. 8 Yet research by Ted Underwood and Jordan Sellars offers a rather more mundane explanation 9 : they argue that all literary language experienced a shift from Latinate to Anglo-Saxon diction in the nineteenth century, and, noting that “abstract value” words tend to be Latinate and “hard seed” words Anglo-Saxon, suggest that the trend observed by Heuser and Le-Khac is indicative not of a shift from telling to showing, but simply evidence of a much broader change in literary fashion. Underwood and Sellars also note that the eighteenth century exhibits a reverse trend from Anglo-Saxon to Latinate diction. Since it would diffi - cult to argue that this was the result of a shift in eighteenth-century narrative from showing to telling, Heuser and Le-Khac’s explanation falls apart. Their paper thus stands as a manifest example of the dangers of working at the scale of big data, where, without close readings to ground interpretation, it is all too easy to impose grand theories on ambiguous results. In Macroanalysis, Jockers, perhaps sensing that he has gone a step too far in calling close reading “totally inappropriate as a method of studying literary history,” appends a conciliatory footnote. Citing Erich Auerbach’s Mimesis (1948), he clarifi es that he hasn’t meant to “imply that scholars have been wholly unsuccessful in employing close reading to the study of literary history.” 10 His mention of Auerbach at this point is signifi cant, since Auerbach himself was extremely explicit about what he felt was the proper relationship between close and distant reading, big and small data, 52 A. HAMMOND ET AL. in literary scholarship. “Philology and Weltlitatur” (1952), for example, reads as preemptive pre-digital rebuttal of distant reading techniques of the kind employed by Jockers, Heuser and Le-Khac. Beginning from “a great mass of material,” Auerbach warns, will inevitably lead to “the introduction of hypostatized, abstract concepts of order.” 11 At a scale where data is ambiguous and concrete close readings are in short supply, he argues, “ready-made, though rarely suitable, concepts whose appeal is deceptive because it is based on their attractive sound and their modishness, lie in wait, ready to spring in on the work of a scholar who has lost contact with the energy of the object of study.” 12 For Auerbach, the solution lies in what we might now call a combination of close and distant reading. Though Auerbach—the author of works with grandiose titles like Mimesis: The Representation of Reality in Western Literature —is very much interested in the “big,” he argues that large-scale analysis must begin from concrete phenomena perceived in close reading. “In order to accomplish a major work of synthesis,” he writes,

it is imperative to locate a point of departure [Ansatzpunkt ], a handle, as it were, by which the subject can be seized. The point of departure must be the election of a fi rmly circumscribed, easily comprehensible set of phenomena whose interpretation is a radiation out from them and which orders and interprets a greater region than they themselves occupy. 13

In this chapter, we describe two projects that attempt an Auerbachian resolution of the close-versus-distant dilemma. Our “handle,” our Ansatzpunkt , is modernist dialogism: the ethically charged, politically infl ected tendency of modernist writers to include mutually differentially and often ideologically opposed voices in their works. Using cutting-edge techniques in computational stylistics, our work leverages the insights available at the scale of big data to model and explore dialogism as a concrete phenomenon in modernist texts. By developing new quantitative metrics that are trained on large datasets yet easily interpretable by humans, we build an important bridge between the scales of big and small data, and also between the disciplines of computer science and literary studies. Our approach is specifi cally tailored, moreover, to modernist literary studies, developing its computational style-based methodology in response to modernist-era accounts of the politics and ethics of genre (Mikhail Bakhtin’s “dialogism” and Auerbach’s “multipersonal representation of consciousness”). In our project He Do the Police in Different Voices , MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 53 which draws primarily on Bakhtin’s account of dialogism, we use extrinsic features based on information from massive corpora to identify possible points of stylistic transition in T. S. Eliot’s The Waste Land , and we employ our novel “stylistic profi le” method to produce human-interpretable analyses of individual “voices” in the poem. In our project The Brown Stocking , which takes its theoretical impetus and its name from Auerbach’s account of modernist polyvocality in the fi nal chapter of Mimesis , we use stylistic profi les to analyze free indirect discourse (FID) and character speech in Virginia Woolf’s To the Lighthouse and James Joyce’s “The Dead.” Our goal in these projects is not to produce defi nitive, computationally guar- anteed readings, but rather to use computational analysis to test, probe and enliven human close readings. Rather than using distant reading to confi rm broad critical metanarratives, we seek to establish a feedback loop in which the insights available at the scale of big data are employed to con- tinuously challenge particular close readings. Ours is a hybrid approach that places distant and close reading in a reciprocal dialogue, based on the conviction that each stands to benefi t from the perspective that the other has to offer.

BAKHTIN, AUERBACH AND THE POLITICS OF MODERNIST DIALOGISM The modernist period in Europe (roughly 1880–1950) was one of intense debate about the politics and ethics of genre, and the narratological theory of Mikhail Bakhtin and Erich Auerbach is representative of the modernist tendency to approach such questions through the lens of voice. 14 In “Discourse in the Novel,” written in exile in Kazakhstan in the 1930s, Bakhtin championed the novel on the grounds that its multi-voiced and open-ended form presented a model of a pluralist, democratic society at a time of brutal totalitarian repression in the USSR. 15 Bakhtin’s argument in favor of the novel is supported by an argument against poetry—particularly lyric poetry, which he positions as the novel’s single-voiced other. Bakhtin’s ideal novelist is one who renders the differentiated dialects of everyday life without seeking to order or purify them. In his account:

The prose writer does not purge words of intentions that are alien to him, he does not destroy the seeds of heteroglossia embedded in words, he does not eliminate those language characteristics and mannerisms glimmering behind the words and forms. 16 54 A. HAMMOND ET AL.

The poet, by contrast, purges, destroys, and eliminates in order to fi t the linguistic universe into a single, standardized pattern. “The language of the poet is his language,” Bakhtin writes: “he is utterly immersed in it, inseparable from it.” 17 His primary formal example of what he calls the “single-personed hegemony” 18 of the poet’s language comes in his analysis of rhythm, which, he argues, “destroys in embryo the social worlds of speech and of persons that are potentially embedded in word […] strip- ping all aspects of language of the accents and intentions of other people, destroying all traces of social heteroglossia and diversity of language.” 19 For Bakhtin, formal categories of genre such as rhythm surpass the bounds of the merely aesthetic by modeling politically infl ected modes of thought. The stylistically uniform lyric modeled acquiescence to totalitarianism, whereas the dialogic novel modeled open-ended democratic debate. Just as Bakhtin wrote “Discourse in the Novel” in exile from an authoritarian regime, Auerbach, a German Jew, wrote Mimesis in exile from Nazi Germany. Like Bakhtin, Auerbach pursues in Mimesis a political reading of multi-voicedness. For Auerbach, what is most signifi cant in twentieth- century fi ction is its development of a technique he calls the “multipersonal representation of consciousness” 20 —a close analogue of Bakhtin’s “dialogism.” At a time when totalitarian regimes in Europe were violently imposing their single-voiced interpretations, Auerbach perceived a form of artistic resistance in modernist texts that offered:

not one order and one interpretation, but many, which may either be those of different persons or of the same person at different times; so that overlapping, complementing, and contradiction yield something we might call a synthesized cosmic view. 21

Auerbach’s chief example of this bottom-up, multi-perspectival, multi- voiced conception of reality is Virginia Woolf’s novel To the Lighthouse , in which, he argues, “the writer as narrator of objective facts has almost completely vanished” and “almost everything stated appears by way of refl ection in the consciousnesses of the dramatis personae.” 22 In Auerbach’s reading of Woolf, one of the principal technical devices by which she achieves the “multipersonal representation of consciousness” is free indirect discourse (FID)—a narratological device for introducing character speech in such a way as to blur the boundaries between the voice of the narrator and that of the character, and so further to diminish narrator’s role as a dispenser of authoritative truth. Together, Auerbach and Bakhtin present a powerful MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 55 case for dialogism as a crucial feature of modernist literature: a stylistic device, practiced and theorized by modernists themselves, seen to have social and ethical reverberations well beyond the sphere of the literary. Dialogism also provides an excellent starting point for reading modernism with machines: since style has proven historically to be the most tractable literary element for computational analysis, dialogic style presents a practical “handle” with which to grasp modernism digitally. Whereas many stylistic categories draw critics away from political or social contexts, dialogism draws us into a confrontation with the politics of form.

MODELING DIALOGISM IN T HE WASTE LAND: IDENTIFYING VOICE SWITCHES For reasons of space, time and language, T. S. Eliot had no access to the writings of either Bakhtin of Auerbach, yet he shared many of their concerns. Perhaps the most prominent lyric poet of the modernist period, he was sometimes attacked by his contemporaries for his inability or unwill- ingness to admit mutually differentiated, competing voices into his work; Virginia Woolf, for one, called him “a monologist.” 23 In this respect, however, Eliot was perhaps his own harshest critic. Throughout the 1920s, Eliot repeatedly expressed his desire to abandon lyric poetry for a form even more multi-voiced than the novel: the narrator-less drama. 24 Eliot theorized the potential of dramatic form in essays of the period, such as “The Possibility of a Poetic Drama” (1920) and “Marie Lloyd” (1922), and experimented with it creatively in the hybrid poetic jazz drama pro- visionally titled Wanna Go Home, Baby , later published in fragmentary form as Sweeney Agonistes (1926–27). His best-known work, The Waste Land (1922), also bears the traces of Eliot’s experiments with dramatic form, standing as a hybrid of his earlier lyric forms and the multi-voiced verse drama he would adopt from the 1930s onward. The Waste Land teems with voices—voices young and old, rich and poor, mundane and eternal, speaking all manner of languages and class dialects. Though Eliot does not provide a dramatis personae or mark the points of transition between the poem’s voices, they emerge clearly in any good reading of the poem—for instance, in those by Alex Guinness and Fiona Shaw included on Faber’s 2011 Waste Land iPad app. Though Eliot’s own readings—two of which are included on the same app—are not nearly so vocally diverse, he clearly intended the poem to be understood as polyvocal, referring to 56 A. HAMMOND ET AL. the “personage[s] in the poem” in his famous endnotes, 25 and employing the working title “He Do the Police in Different Voices.” 26 The latter is the name we have taken for our long-term project to explore and highlight the dialogism of this most famous of modernist poems. 27 Our work began in 2011 with the creation of a digital edition designed to emphasize the poem’s uncertain generic status between single-voiced lyric and impersonal drama. The fi rst stage involved aggregating 140 student interpretations of The Waste Land into a “class-sourced” reading of voices in the poem. Having asked students in “The Digital Text,” a second-year English course at the University of Toronto, to indicate every instance in The Waste Land where they perceived a “voice switch,” we used this data to devise a reading of the poem in which we identifi ed sixty-eight voice switches and twelve characters. 28 On our project website, hedothepolice. org , we present this interpretation in the form of a digital edition (“What the Class Said” [WTCS]) that renders each unit of character speech in a unique typeface. The goal of this stage of the project was to teach students about modernist dialogism by having them act it out : to suggest that literary interpretation, particularly of dialogic modernist literary texts, is a communal, participatory act involving multiple competing perspectives. Crowdsourcing was thus employed not merely as a means to an end but also, to some extent, as an end in itself. Taken on its own and isolated from the polyvocal process of its creation, however, the WTCS edition runs the risk of suggesting that its interpretation is defi nitive or “fi nal.” To mitigate against this suggestion, our project website includes an interactive page on which users can indicate their own set of voice switches and assign these to particular characters (“Have Your Own Say”). To further unsettle the particular interpretation of the WTCS edition, and to encourage further exploration of voices in The Waste Land generally, we have also sought to insert a computational “voice” into the discussion (“What the Computer Said”). Our work has pursued quantitative methods for performing the two basic interpretive tasks described to this point: fi rst, segmenting the poem by identifying points where “voice switches” occur; second, clustering these discrete chunks into individual speaking voices. Our fi rst task was to develop a computational means of identifying the points in The Waste Land where one voice gives way to another. Our approach uses unsupervised techniques (that is, techniques that do not require human intervention at each step) in computational stylistics to locate instances of maximum “stylistic variation,” using a procedure— described in further technical detail elsewhere 29 —that functions roughly MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 57 as follows. For every word in The Waste Land , we calculate a measure of stylistic change that takes into account a number of features in the spans of text immediately preceding and following that word. The features we consider fall into two categories: surface and extrinsic. Surface features, which are by far the more common for conventional computational analysis of literature (for instance, the work of Clement and Ramsay described above), can be calculated entirely from the text itself, requiring no external resources. These features include word length, syllable count, punctuation frequency, parts of speech, verb tense and type-token ratio (a measure of lexical density). Extrinsic features, which are more novel in analysis of literature, rely on lexical information derived from large external corpora. Such features include readability, sentiment polarity (the positive or negative affective stance of a given span), formality, and less human- interpretable (but extremely useful) features from latent semantic analysis (LSA). Our method works by investigating the features in a “sliding window” of text on either side of each word; for instance, it might calculate the sentiment polarity of the fi fty words immediately preceding a given point in the text and compare it with the sentiment polarity of the fi fty words following that point. 30 Our metric is built from the sum of the changes of all the features, and identifi es voice switches at local maxima of the calculated change curve, such as the peaks represented on the curve in Fig. 3.1 . To test our method, we created artifi cial poems composed of randomly assembled sections of twelve poems of diverse style and authorship. 31 These artifi cial poems, with their unmarked transitions between styles and voices, mimic the stylistic diversity of The Waste Land . Our evaluation revealed that extrinsic features (particularly formality and LSA) slightly outper- formed surface features in identifying transitions in our set of artifi cial poems, though the best results of all came from combinations of surface and extrinsic features. Next, we applied our method to two versions of The Waste Land : a “full” version containing all text in the poem except for headers and the dedication; and an “abridged” version omitting stanzas which are less than twenty words in length or in a language other than English—both conditions that make it diffi cult for our method to suc- ceed. 32 Fig. 3.1 shows the change curve generated by our method for the abridged version of the poem, overlaying switches from the WTCS edition. In many instances, the switches identifi ed by the algorithm coincide almost perfectly with those identifi ed by human readers. Further, the model tends to predict more switches in sections where humans perceive numerous 58 A. HAMMOND ET AL.

Fig. 3.1 Stylistic change curve over the abridged version of The Waste Land switches, mostly notably in the last third of the poem. Our results thus bring us to the point at which computational analysis trained on large datasets can meaningfully begin to contribute to close reading: the results conform suffi ciently to human interpretation to convince us that they are not merely random, yet diverge suffi ciently to allow us to evaluate whether particular machine interpretations can offer something new. On closer inspection, we found that several of the points at which the computational model departs from the human interpretation suggests new and insightful interpretations. Consider the famous opening lines of the poem:

April is the cruellest month, breeding Lilacs out of the dead land, mixing Memory and desire, stirring Dull roots with spring rain. Winter kept us warm, covering Earth in forgetful snow, feeding A little life with dried tubers. Summer surprised us, coming over the Starnbergersee With a shower of rain; we stopped in the colonnade, And went on in sunlight, into the Hofgarten, And drank coffee, and talked for an hour. (1–11)

Here, our human reading placed a switch between lines 4 and 5: we attributed the fi rst lines to the narrator fi gure we named Tiresias, and MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 59 the next lines to the aristocratic character we named Marie, largely on the assumption “us” of line 5 refers to the latter’s family on vacation in Switzerland. The machine model, however, places the switch between lines 7 and 8. On refl ection, it does so with good reason. While the passage from lines 5–7 transitions gradually between the dreary, remote tone of the opening lines (“cruellest,” “dead land”) and the more neutral, slightly hopeful tones of lines 8–11 (“shower of rain,” “went on in sunlight”), the negative tone of the opening lines remains palpable in phrases like “forgetful snow” and “dried tubers.” As Michael Levenson argues in A Genealogy of Modernism , “the stylistic patterns shifts” between lines 7 and 8: in a human close reading that relies heavily on computationally tractable “surface features” such as syntax and verb tense, Levenson notes, “The series of participles disappears, replaced by a series of verbs in conjunction” and “The adjective-noun pattern is broken.” 33 Reconsidering the passage, we agree with the close readings of Levenson and of the algorithm that we ought to have placed the switch between lines 7 and 8. The computer model suggests another insightful interpretation in the following stanza:

Unreal City Under the brown fog of a winter noon Mr. Eugenides, the Smyrna merchant Unshaven, with a pocket full of currants C.i.f. London: documents at sight, Asked me in demotic French To luncheon at the Cannon Street Hotel Followed by a weekend at the Metropole. (207–14)

Our interpretation attributed the entirety of this stanza to the character we (the fi rst author and the 140 students) named “Crazy Prufrock,” the educated but increasingly unhinged character who earlier in the poem speaks of planting a corpse in a garden (60–76). We attributed this stanza to him largely because it begins with the words “Unreal City,” the same phrase that opens the account of the corpse. While we allowed this opening phrase to color our interpretation of the rest of the stanza, the computer model inserts a break early in the paragraph, between lines 208 and 209. On refl ection, this seems to us a preferable interpretation, since while the fi rst two lines have a Prufrockian air, the remainder of the stanza is deliv- ered in a balanced, detached tone more reminiscent of the poem’s narrator 60 A. HAMMOND ET AL. fi gure, Tiresias. Notably, Jewel Spears Brooker and Joseph Bentley also attribute this passage to Tiresias, following an exhaustive and ingenious close reading, in which they determine him to be “the only fi gure in the vicinity of the poem who can be trusted to see all about the fi gure [of Mr. Eugenides].” 34 In instances like these, our model shows the power of computational methods trained on massive corpora to contribute productively to the minutest of human close readings.

MODELING DIALOGISM IN THE WASTE LAND : STYLISTIC PROFILES Having segmented the poem into stylistically distinct chunks, our second computational task for The Waste Land involved clustering: determining whether we could group passages belonging to distinct characters. Pursuing a similar LSA-centered feature set to that we used for segmenting the poem, our work was moderately successful. 35 Since LSA produces results that are not particularly interpretable by humans, however, this work did little to further our goal of provoking new close readings. Realizing that a new metric was required in order to produce the sorts of interpretations that might prompt new close readings, we turned our focus toward developing our signature six-dimensional approach to linguistic style, which employs an automatically created lexical resource to produce human-interpretable “stylistic profi les.” 36 As our work proceeded in The Waste Land , we came to realize that we could use these stylistic profi les as a means of testing our intuitions in assigning particular spans of text to particular characters. The advantage of this approach—the reason we have found it so useful for literary analysis and the reason we believe it represents a large step forward from techniques like PCA and LSA—is its accessibility and transparency even to readers entirely unversed in computational stylistics. Our profi ling method is based on six discrete aspects of style: objectivity (use of words that project a sense of disinterested authority, such as invariable and ancillary ); abstractness (words denoting concepts that cannot be described in purely physical terms, and which require signifi - cant cultural knowledge to understand, such as solipsism and alienation ); literariness (words found in traditionally literary texts such as wanton or yonder ); colloquialness (words used in informal contexts such as booze and crap ); concreteness (words referring to events, objects or properties in the physical word, such as radish and freeze ); and subjectivity (words that are MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 61 strongly personal or refl ect a personal opinion, such as ugly and bastard ). Our process for building stylistic lexicons, described in detail elsewhere, 37 functioned as follows. First, we produced a list of approximately 900 words carefully selected for their stylistic diversity, which human annotators (fi ve university-educated native English speakers trained for the task with simple written guidelines) evaluated in terms of the six stylistic aspects listed above. (Annotators noted, for instance, that the word “brazen” projected subjectivity and literariness but none of the other aspects.) Once suffi cient inter-annotator agreement was reached, we used an automated procedure to collect information on how these 900 words are employed in a large corpus composed of all English texts in the 2010 image of Project Gutenberg . 38 Using this information, we were able to derive stylistic information automatically for any word; in this case, we investigated every word in The Waste Land (as well as signifi cant multi-word expressions, such as from time to time and ought to be ashamed 39 ) and, based on their employment in the same Project Gutenberg corpus, assigned a value between −1 and +1 (to twelve decimal places) for each of the six stylistic aspects. Using this information, we are able to produce stylistic profi les for particular segments and particular characters in the poem by aggregating results for individual words or multi-word expressions. This method proved extremely successful in capturing individual characters’ manners of speech in the WTCS reading of the poem (Table 3.1 ). Human readers often identify Woman in Bar, the Cockney woman whose speech dominates the end of “A Game of Chess,” as the most distinctive voice in the poem. Our computational approach likewise found her voice to be the most distinct. Her stylistic profi le—marked by extremely high colloquial and subjective values, and extremely low values for the objective and literary dimensions, all of which corresponds to our intuitions—is distinguishable from all other voices in the poem, in most of the six aspects, at statistical signifi cance of p < 0.001 (where p < 0.05 is considered a reliable threshold of statistical signifi cance). The stylistic profi les of other characters likewise conformed to our qualitative expectations. Marie, emotional and nostalgic with highly oral language, is marked by high subjectivity and high colloquialness. Crazy Prufrock, educated but unbalanced, is marked in our analysis by high abstraction, high colloquialness and high objectivity, indicating not mental stability but high cultural knowledge and education. The narrator fi gure we call Tiresias is marked by relatively low values for colloquialness and correspondingly high values for objectivity and literariness. 62 A. HAMMONDETAL.

Table 3.1 Stylistic profi les for various characters in The Waste Land Character Unique words Stylistic dimensions Objective Abstract Literary Colloquial Concrete Subjective Tiresias 460 0.09 − 0.04 0.03 − 0.21 0.02 − 0.03 Marie 132 −0.07 − 0.13 − 0.03 0.02 0.04 0.03 Hellfi re 207 0.00 0.00 0.06 − 0.14 0.07 −0.06 preacher Chorus 105 − 0.02 −0.02 0.03 0.14 0.04 −0.06 Intrepid 66 −0.06 0.28 − 0.03 0.07 − 0.06 0.05 reporter Madame 15 −0.47 0.26 −0.08 0.65 −0.14 0.16 sosostris Crazy prufrock 399 0.01 0.07 0.00 0.01 −0.01 −0.01 Nervous one 126 − 0.26 0.09 0.04 0.29 −0.07 0.07 Woman in bar 151 − 0.45 −0.01 − 0.24 0.73 − 0.11 0.20 The typist 54 0.14 0.42 −0.03 − 0.14 −0.02 −0.02 MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 63

These values show us that our method seems to work; that is, that it produces human-interpretable results that correspond suffi ciently to our intuitions to enable us to trust them. Yet, as we argue of all computational metrics, stylistic profi les only really become useful when they suggest something we didn’t already know. In this case, they proved useful by prompting us to reconsider our assignments of particular passages to specifi c characters. One nagging concern we encountered in devising our human interpretation of the poem was whether Tiresias and Crazy Prufrock were suffi ciently distinguishable to stand as independent characters. Given that both voices were marked by the same qualitative traits—wordiness, a deep familiarity with the literary tradition and a fond- ness for literary quotation—we sometimes wondered, along with critics like Calvin Bedient, 40 whether they weren’t simply projections of a single consciousness. Our stylistic profi les provide reason to consider the two characters distinct. Similar as the voices are in the literary dimension, they are strongly distinguished in colloquial ( p < 0.001), where Prufrock’s schizophrenic shifts across registers produce much higher values. We likewise debated whether Crazy Prufrock is speaking to himself or to another voice in the extended back-and-forth dialogue that occurs in the middle of “A Game of Chess.” In our WTCS interpretation, we described this passage as an exchange between Prufrock and another character, “Nervous One,” and data from our stylistic profi les reinforces our choice by strongly distinguishing the voices in the subjective, objective and colloquial dimensions (all p < 0.001). Stylistic profi les were perhaps most useful of all for testing our qualitative “clustering” of the poem, certainly the most subjective and intuitive interpretive procedure we employed to produce the WTCS edition. In a few instances, data that seemed to suggest a misreading in our interpretation in fact reinforced it. Despite strongly divergent style data for the second (77–110) and third (215–56) passages we attributed to Tiresias (“Tiresias 2” and “Tiresias 3” in the naming convention followed in the rest of the chapter), for instance, we remain convinced of our reading. In Tiresias 2, which describes a rich woman’s elaborate grooming ritual, the narrator’s presentation is strongly ironic: the evocation of “The chair she sat in, like a burnished throne,” borrowed from Enobarbus’s account of Cleopatra’s raft in Antony and Cleopatra , is deliberately overblown, serving to demonstrate the extreme disconnect between the cocoon of the dressing-room and the “Unreal City” beyond. This disjuncture is signaled through the painting that sits on the woman’s mantel, depicting a scene from the Philomela 64 A. HAMMOND ET AL. myth. Not even a pastoral rendering of the story can avoid evoking the brutality of Tereus’s rape, and in presenting his ekphrasis , Tiresias momen- tarily abandons his hyper-refi ned diction to comment with unadorned lexis on the persistence of cruelty in the modern world: “And still she cried, and still the world pursues,/‘Jug Jug’ to dirty ears” (102–3). By contrast, Tiresias 3, in which the self-assured “young man carbuncular” forces himself upon the passive “typist,” presents a much more direct account of a contemporary rape. While these scenes clearly respond to and mirror one another, they yield very different stylistic profi les. The language in Tiresias 3 is signifi cantly more colloquial than in Tiresias 2 (p < 0.01), refl ecting the fl atly sordid account of the typist’s rape. It is also markedly more subjective ( p < 0.01), refl ecting the more honest account of the typist’s feelings, as opposed to the ironic evocation of the rich woman’s hermetic emotional landscape, buffered on all sides by luxury. In this case, then, the divergent stylistic profi les simply highlight the chameleonic aspect of Tiresias’s nar- ratorial style, which adapts itself to the particular scene presented. 41 Elsewhere, however, style data led us to change our interpretation. Another discrepancy in passages attributed to Tiresias—between Tiresias 2 and Tiresias 5 (378–85)—uncovered an untenable reading. Tiresias 5 begins with a description that recalls the dressing scene (“A woman drew her long black hair out tight”); another link is established between the passages through the echo of the opening words of Tiresias 2, “At the violet hour,” in the description of “bats with baby faces in the violet light” (380). Yet while such reverberations were suffi cient to convince us of a connection, our stylistic profi les show little to suggest a common speaker. Prompted by this data, we reconsidered the passage, and noted that it echoes words not only from Tiresias, but also from numerous other voices in the poem. Its evocation of “towers/Tolling reminiscent bells, that kept the hours” (384), for instance, recalls two passages we attributed to Crazy Prufrock: “where Mary Woolnoth kept the hours” (Prufrock 2, 67) and “Falling towers” (Prufrock 15, 374). Since this passage deliberately mixes together fragments of voices from throughout the poem, we decided to attribute this passage to the non-personal entity we call “The Chorus.” Beyond testing particular interpretations, stylistic profi les can provide a starting point for evaluating writers’ representations of certain classes of characters. For instance, we were interested to see whether Eliot’s male or female voices are more mutually differentiated. Investigating the fi g- ures, we noted that his male characters are more vocally diverse, and that each of his female characters (Marie, Madame Sosostris, Nervous One, MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 65

Woman in Bar and The Typist) has a relatively high score for subjectivity, possibly indicating a stereotyped representation. 42 Certain female voices are quite distinct: Marie and Madame Sosostris, for example, register statistically signifi cant differences in abstract, colloquial and concrete (all p < 0.01). Yet, in a poem that differentiates so successfully between its voices—across all possible pairings of characters, only two pairings fail to register a single statistically signifi cant difference (p < 0.05)—it is telling that one of these indistinct pairings should be between female characters, Nervous One and Madame Sosostris. (The other is Crazy Prufrock and the non-personal Chorus.) Yet careful analysis is required before we jump to conclusions: their similarity may be due to Eliot’s failure to distinguish female voices, but it may also be due to these characters’ similar registers (both highly oral) or simply to the fact that there is insuffi cient data for Madame Sosostris, who speaks very little. Although the stylistic profi les we produced for The Waste Land were not able to answer these questions defi nitively, they were able to raise them with new urgency. As such, they were suffi ciently promising to prompt us to investigate their application in other modernist texts.

QUANTIFYING FREE INDIRECT DISCOURSE IN TO THE LIGHTHOUSE AND “THE DEAD” At the time of our investigation of The Waste Land , we were involved in another project focused on modernist dialogism, The Brown Stocking , which looked at free indirect discourse (FID) in Virginia Woolf’s To the Lighthouse and James Joyce’s “The Dead.” (The name of the project is taken from the fi nal chapter of Auerbach’s Mimesis , where he reads FID in To the Lighthouse as an example of modernist “multipersonal representation of consciousness.”) Though this project was not initially devised with stylistic profi les in mind, it benefi tted signifi cantly from a shift toward a style-based approach, further demonstrating the power of computational methods trained on large-scale datasets to vivify literary inquiry and contribute meaningfully to close reading. We began The Brown Stocking with three principal aims. First, we wanted to help our undergraduate students better understand To the Lighthouse by highlighting its principal interpretive dilemma: the vexed question of who is speaking at any given point. We pursued this through a TEI encoding exercise that asked students to annotate short passages from the novel. For each instance of character speech in their assigned passage, students were 66 A. HAMMOND ET AL. asked to indicate whether it was introduced as direct, indirect or free indirect discourse; whether it was spoken aloud or silently; and which character was speaking. Because there are often multiple valid interpretations of a given passage, we assigned each to four or fi ve students. We devised this as an exercise in computer-assisted close reading, and, in practice, students reported that the act of translating their implicit interpretations into explicit markup helped them to clarify their reading of the text. The next goal was to combine these interpretations into a digital edition of To the Lighthouse that would serve as a “reader’s map,” showing the vast array of possible interpretations of Woolf’s text and thus visualizing an active circuit of modernist dialogism: a dialogic novel prompting a dialogic scene of reader response. Following two rounds of annotation, each involving approximately 160 students and focusing respectively on the fi rst four and fi nal seven chapters of the novel, we published this edition on the project website, brownstocking.org . 43 Our fi nal goal was to devise a means of using these student annotations to train a machine-learning model that could detect FID automatically in untagged plain text. In pursuing this goal, we were consciously seeking to replicate Auerbach’s understanding of the “multipersonal representation of consciousness” as an aggregation of numerous distinct interpretations that, when combined, provide a “synthesized cosmic view.” In practice, however, this provided diffi cult because inter-annotator agreement was quite low, due to the complicated, multi-voiced nature of the text, in which Woolf uses FID so pervasively. We thus decided to perform another round of annotation on a modernist text with a more conventional use of FID—James Joyce’s “The Dead”—yet the added data brought us no closer to a machine-learning system for detecting FID. (We were, however, able to devise a relatively accurate rule-based system for identifying FID from grammatical and syntactic clues—and we produced a “reader’s map” edition for “The Dead” at livingdead.ca ). 44 The data proved immensely useful, however, in a task quite different from that for which it was initially collected: the further exploration of our method of stylistic profi ling. The fi rst research question we posed was a fundamental one related to the defi nition of FID. If FID has become today a reasonably familiar element of literary discourse, the history of the invention, detection, or critical elaboration of FID is suffi ciently curious to merit careful scrutiny. If we consider that the fi rst novels were produced in the sixteenth century, it took some two hundred years of literary history for FID to fi rst be employed; though Cervantes used direct and indirect discourse, it was MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 67 not until the time of Austen and Goethe that FID appeared in the novel. Following its invention, it took another century for critics to notice it. It is generally agreed that Adolph Tobler was the fi rst to identify the device, calling it “a peculiar mixture of direct and indirect speech” in 1892. 45 In the years that followed, FID became a focus of intense modernist critical scrutiny. Graham Pechey estimates no fewer than eighteen separate names were given to the device in the modernist period, among them “veiled speech” (Theodor Kalepky 1912), “free indirect style” (Charles Bally 1912), “pseudo-objective speech” (Leo Spitzer 1921) and “pseudo-objective discourse” (Mikhail Bakhtin, 1920s). 46 The one common notion in these various defi nitions of FID—a notion that retains its critical force today—is that FID is an “in-between” mode of discourse (a “peculiar combination ,” a “ pseudo ” or “ veiled ” form) existing on the continuum between pure narration and direct discourse. Given the delayed and uncertain process of defi ning FID—a process that was carried out in a haphazard and entirely qualitative manner—we were interested to see whether our quantitative method could support or refute the notion of FID’s “in-betweenness.” Our method for testing this defi nition, described in further technical detail elsewhere, 47 proceeded as follows. First, we located all the passages in To the Lighthouse and “The Dead” in which a majority of annotators identifi ed a span as FID, direct discourse, or narration. 48 Then, using the method described earlier in relation to The Waste Land , we built stylistic lexicons for both texts, and used these to generate stylistic profi les for narration, spoken direct discourse, silent (thought) direct discourse, and FID (Table 3.2 ). For both texts, our results largely conformed to expectations. In “The Dead,” FID is “in-between” in all six dimensions, most clearly in colloquial and subjective. In To the Lighthouse , FID occupies a middle position in four of six stylistic dimensions: it is more abstract than narration, but less abstract than directly rendered thought; more literary than narration but less so than direct speech or thought; less concrete than narration but more so than direct speech or thought; and so on. Exceptions occur in objective and colloquial, where FID is in an extreme position; yet in both cases, FID tracks closely with narration, and the particular divergences may simply refl ect a mannerism of Woolf’s narrator, who tends not to admit colloquialisms when mixing her language with that of her characters. Our work thus offers quantitative support for two long-held but seldom-tested hypothesis about FID: that it is an identifi able mode of discourse distinct from narration and direct discourse, and that it falls stylistically between these two poles. 49 68 A. HAMMONDETAL.

Table 3.2 Stylistic profi les for discourse types in To the Lighthouse and “The Dead”

Text Discourse Unique Stylistic dimensions words Objective Abstract Literary Colloquial Concrete Subjective

To the Lighthouse Narrator 765 0.00 0.00 0.00 0.00 0.00 0.00 FID 2916 0.08 0.16 0.02 −0.02 − 0.15 0.02 Thought 212 − 0.15 0.21 0.07 0.30 − 0.20 0.08 Speech 172 − 0.32 0.14 0.06 0.49 − 0.20 0.11 “The Dead” Narrator 1325 −0.01 0.02 0.08 0.04 0.00 0.09 FID 400 −0.13 0.19 0.10 0.19 − 0.15 0.11 Thought 57 − 0.43 0.18 0.12 0.74 − 0.30 0.22 Speech 651 −0.11 0.23 0.06 0.27 − 0.19 0.16 MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 69

Next, we investigated whether stylistic profi les would prove insightful in the mixed, murky waters of FID. Although the method worked well in The Waste Land , where speech is rendered mostly as direct discourse, we were unsure whether individual characters’ stylistic personalities would reveal themselves in FID, in which their speech is mixed with that of the narrator. Here, again, the results were promising. The stylistic profi les of the narrators of To the Lighthouse and “The Dead” in Table 3.2 highlight revealing differences: where Woolf’s narrator is consistently fl at, detached and objective, Joyce’s narrator scores higher values for literary, colloquial, and subjective. As Table 3.3 shows, stylistic profi les also provide insights into the FID of individual characters. Gabriel’s distinct manner—reserved, given to deep thoughts and literary quotation—emerges clearly in the profi le of his FID, which is notably less colloquial, more literary and more abstract than that of other characters. Profi les of Woolf’s FID are likewise revealing. Some of the rift between Mr. and Mrs. Ramsay is captured in their style values: Mr. Ramsay, lost in his world of philosophical speculation and scholarly research, is much more abstract, much more literary and far less concrete than Mrs. Ramsay. Most interesting in To the Lighthouse are the relationships of inter-character infl uence that the stylistic profi les suggest. Though their style profi les are quite dissimilar, Mr. and Mrs. Ramsay have much more in common with one another stylistically than they do with their children. Cam and James spend much of Part III of the novel pondering the infl uence of their parents, mourning the lost infl uence of their mother while bristling against the domineering authority of their father. This generational confl ict is expressed at the level of style: the Ramsays and their children speak very different languages, with the latter notably less objective, less abstract and more colloquial. Lily too ponders the infl uence of the elder Ramsays in Part III. She is particularly ambivalent toward Mrs. Ramsay, whom she admires deeply while resisting the conventional gender role she adopts in her family. Despite these misgivings, stylistic profi les suggest Lily is indeed Mrs. Ramsay’s stylistic heir: in all six dimensions, their profi les are nearly identical. The situation is very different for Mr. Ramsay and his would-be protégé, Charles Tansley. The young philosopher is desperate to belong to the Ramsays’ social and intellectual world, yet is bitterly aware of the barrier that his working-class origins present. Tansley’s failure to integrate himself into their sphere is as plain in the stylistic profi le data as in the plot of the novel. Stylistically, Tansley and Mr. Ramsay are extremely dissimilar with the exception of their shared philosophical penchant for the abstract. 70 A. HAMMONDETAL.

Table 3.3 Stylistic profi les for characters in To the Lighthouse and “The Dead”

Text Character Unique Stylistic dimensions words Objective Abstract Literary Colloquial Concrete Subjective

To the Lighthouse Mrs. 805 0.07 0.24 0.00 0.03 −0.22 0.03 Ramsay Mr. Ramsay 70 0.09 0.58 0.27 0.01 −0.49 0.00 William 248 − 0.01 0.19 0.03 0.14 −0.17 0.08 Bankes Lily Briscoe 1485 0.06 0.17 0.03 −0.02 −0.15 0.01 James 540 − 0.06 0.03 0.08 0.06 −0.03 0.02 Ramsay Cam 381 − 0.10 0.04 0.04 0.10 −0.06 0.00 Ramsay Charles 138 −0.07 0.21 − 0.07 0.22 −0.23 0.05 Tansley “The Dead” Gabriel 358 − 0.12 0.21 0.12 0.17 −0.17 0.10 Other 85 −0.30 0.06 0.02 0.44 −0.07 0.14 MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 71

Table 3.4 Stylistic profi les for various social groups in To the Lighthouse Social identity Unique Styles Factor Category words Objective Abstract Literary Colloquial Concrete Subjective Age Young 969 −0.03 0.06 0.04 0.04 −0.06 0.01 Old 2248 0.09 0.21 0.02 −0.02 −0.19 0.02 Class Lower 138 −0.07 0.21 −0.07 0.22 − 0.23 0.05 Higher 2844 0.08 0.16 0.03 −0.02 −0.15 0.02 Gender Female 2356 0.08 0.18 0.02 − 0.02 − 0.17 0.01 Male 878 0.02 0.14 0.06 0.03 − 0.12 0.03

Note: Bold indicates statistically signifi cant difference at the p < 0.01 level between two categories of the same factor.

Prompted by the disparity between Tansley and Mr. Ramsay, we pursued a deeper investigation into the infl uence of socioeconomic categories on character speech in To the Lighthouse (Table 3.4 ). Tansley is the only character of working-class background who is attributed FID in the novel; working-class characters such as Macalister or Mrs. Beckwith speak directly or not at all. Comparing his limited FID with that of characters of higher class, however, we fi nd a conventional power dynamic at work: higher-class characters are more authoritative, more literary, more concrete, less subjective and far less colloquial. The stylistic difference between age groups is similar to that between classes, though the key distinction for age is abstraction (words that require signifi cant cultural knowledge) whereas that for class is literariness. Perhaps most interesting is that while To the Lighthouse reproduces conventional power dynamics for class and age, it almost completely reverses them for gender. Compared to male characters, female characters are more objective, more abstract, less colloquial and less subjective—and Mr. Ramsay’s extreme values for literariness and concreteness likely explain why men rank slightly above women in these categories. These results may be taken by some critics as confi rmation of biases in Woolf’s authorial practice. As one of the most vigorous champions of feminism and female authorship of the modernist period, it will come as little surprise that she extended this struggle to the level of style, erasing and indeed reversing gendered linguistic power dynamics. On the other hand, those who have accused the upper-middle-class Woolf of insensitive or stereotyped representations of lower-class characters 50 will fi nd quantitative support in our stylistic profi les. As elsewhere, however, we urge readers to consider these fi gures not as the fi nal word, or defi nitive proof, 72 A. HAMMOND ET AL. but rather as prompts for further close reading. Indeed, Bakhtin and Auerbach championed modernist dialogism precisely because, by presenting markedly differentiated strata of socially infl ected speech, it modeled the lively interchange of democratic debate. From our perspective, the question of whether Woolf should be applauded or condemned for differentiating the speech of characters of different ages, classes and genders is one that requires more than quantitative data to answer. How mimetically accurate is her depiction of female or lower-class speech? What resources, qualitative and quantitative, might we need to draw upon in determining this? Where she departs from mimesis, how likely is it that she does so deliberately? If her departure is deliberate, what is she trying to achieve? If not, how might this lead us to re-evaluate her authorial practice, or modernist authorship more generally? Responding to these questions, raised by computational models trained on large data sets, requires all our resources as literary critics: intimate familiarity with literary history, knowledge of context, and the ability to read closely and carefully.

BIG DATA IN THE HERMENEUTIC CIRCLE In this chapter, we’ve focused on the way that analytic techniques trained on large datasets can animate interpretation of a few canonical modernist texts; as our subtitle suggests, we have looked at “close reading with big data.” As we refl ect on what we’ve learned in our research, our focus is shifting toward applying these techniques to ever-larger numbers of texts. In order to build the stylistic lexicons we used to produce stylistic profi les of voices in The Waste Land , To the Lighthouse and “The Dead,” we developed a technique for automatically separating character speech from narration in untagged plain text. 51 Applying our rule-based approach to identifying FID, and supplementing it with what we learned from investigating stylistic profi les in Eliot, Woolf and Joyce, we are now developing techniques for automatically identifying characters and classifying their speech as direct, indirect and free indirect discourse. Having demonstrated the usefulness of our method of stylistic profi les through close engagement with individual literary texts, we are in a position to begin an algorithmic investigation of the history of dialogism in English-language fi ction. Now that we are able to derive automatic dramatis personae for any novel or play, and to calculate a quantitative measure of the stylistic diversity that exists in each text, we will have a quantitative means gaining insight into several large-scale questions about dialogism. Are the works of modernist writers like Woolf, Joyce and Eliot—all of whom pursued dialogism as a conscious aim—really MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 73 the most dialogic in the literary record? How does dialogism map onto historical time; do periods of political turmoil correspond to changes in the stylistic diversity of fi ction? Which regions produce the most stylistically varied writing? Do changes in the dialogism of fi ction anticipate changes in non-fi ction? What previously ignored authors, periods and genres might our method consider as particularly dialogic? As we make this Auerbachian leap from the concrete “handle” of modernist dialogism to the largest scale of literary history, we expect our technique to raise new questions, to prompt investigations of new texts, and to alert us to unexpected writers, periods and genres—in other words, to supply us with an abundance of material that will require our most attentive close reading. In pursuing our research, we fi nd it useful to envision the role of computational analysis within the framework of Dilthey’s hermeneutic circle. Dilthey posits that literary interpretations emerge from interactions at different scales of meaning: the movement of the hermeneutic circle is propelled by the paradoxical fact that while we can understand the whole of a literary work only through careful consideration of its individual parts, so too can we know individual parts only through careful consideration of whole. In the hermeneutic circle, literary interpretation is a necessarily mobile, dynamic act of holding together various mutually interdepen- dent elements. From our perspective, the insights available at the scale of big data contribute to, and by no means invalidate, this dynamic. To shift metaphors somewhat, we see big data as a cog in the movement of the hermeneutic circle rather than a wrench thrown into the works. In our investigations into modernist dialogism, extrinsic features and human- interpretable stylistic profi les trained on massive datasets helped us to refi ne our interpretations, shed light on fi ne points of theme and characterization, and allowed us to probe basic defi nitions of literary terms. In each of these tasks, close and distant reading are complementary. Far from “inappropriate” in the context of big data, close reading remains the ground by which distant reading achieves its effects and demonstrates its usefulness.

NOTES 1. Matthew L. Jockers, Macroanalysis: Digital Methods and Literary History (Urbana: University of Illinois Press, 2013), 7. 2. Ibid. 3. Susan Hockey, “The History of Humanities Computing,” in A Companion to Digital Humanities , ed. Susan Schreibman, Ray Siemens and John Unsworth (Oxford: Blackwell, 2001), http://perma.cc/L3A8-EJHV . 74 A. HAMMOND ET AL.

4. Julia Flanders, “Detailism, Digital Texts, and the Problem of Pedantry,” TEXT Technology 14, no. 2 (2005): 57. 5. Stephen Ramsay, Reading Machines: Toward an Algorithmic Criticism (Urbana, IL: University of Illinois Press, 2011). 6. Tanya Clement, “Text Analysis, Data Mining, and Visualizations in Literary Scholarship,” in Literary Studies in the Digital Age: An Evolving Anthology , ed. Kenneth M. Price and Ray Siemens (MLA Commons, 2013), http:// perma.cc/2CED-BNEK . 7. Jockers, Macroanalysis , 151–3. Jockers’s more recent work employing sentiment analysis to identify six basic plot shapes has been variously attacked as fl awed, misguided and reductive, as well as praised as path breaking and insightful. Until Jockers publishes his results and the scholarly community has an opportunity to test them, it remains too early to take a side. See Eileen Clancy, “A Fabula of Syuzhet: A Contretemps of Digital Humanities and Sentiment Analysis,” Storify , May 3, 2015, https://storify.com/clan- cynewyork/contretemps-a-syuzhet . 8. Ryan Heuser and Long Le-Khac, A Quantitative Literary History of 2,958 Nineteenth-Century British Novels: The Semantic Cohort Method , Stanford Literary Lab Pamphlets 4 (Stanford, 2012), http://litlab.stanford.edu/ LiteraryLabPamphlet4.pdf . 9. Ted Underwood and Jordan Sellers, “The Emergence of Literary Diction,” Journal of Digital Humanities 1, no. 2 (Spring 2012), http://perma.cc/ K655-GMLG . 10. Jockers, Macroanalysis , 7. Jockers later (p. 26) goes further, arguing that he is “not suggesting a wholesale shelving of close reading” but in fact recommending “a blended approach.” Although he writes that close reading and distant reading are “not antithetical”—and indeed “share the same ultimate goal of informing our understanding of the literary record”—his work implies, at best, a model of close and distant reading as parallel interpretive strategies, whereas our argument is that they are most productive when positioned as a “feedback loop.” 11. Erich Auerbach, “Philology and Weltliteratur,” trans. Edward Said and Marie Said, The Centennial Review 13, no. 1 (Winter 1969): 10. 12. Ibid., 16. 13. Ibid., 13–14. 14. For more on Bakhtin, Auerbach, and a historicized account of the development of their theories of polyvocality, see Adam Hammond, “The Honest and Dishonest Critic: Style and Substance in Mikhail Bakhtin’s ‘Discourse in the Novel’ and Erich Auerbach’s Mimesis ,” Style 45.4 (Winter 2011): 638–53. 15. Ken Hirschkop, Mikhail Bakhtin: An Aesthetic for Democracy (Oxford ; New York: Oxford University Press, 1999). MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 75

16. Mikhail Bakhtin, “Discourse in the Novel,” in The Dialogic Imagination: Four Essays , ed. Michael Holquist, trans. Caryl Emerson and Michael Holquist, 18. paperback printing, University of Texas Press Slavic Series 1 (Austin, Tex: Univ. of Texas Press, 2011), 298. 17. Ibid., 285. 18. Ibid., 297. 19. Ibid., 298. 20. Erich Auerbach, Mimesis: The Representation of Reality in Western Literature , trans. Willard Trask (Garden City, NJ: Doubleday, 1957), 536. 21. Ibid., 549. 22. Ibid., 534. 23. Virginia Woolf and Anne O. Bell, The Diary of Virginia Woolf. Vol. 5: 1936–1941 , 1. Harvest ed. (New York, NY: Harcourt Brace & Company, 1985), 210. 24. For an infl uential contemporary genre analysis that privileges drama for its multi-voicedness, see Stephen Dedalus’s famous discussion in James Joyce, A Portrait of the Artist as a Young Man (New York: B. W. Huebsch, 1921), 251–2. 25. T. S. Eliot, The Waste Land, in Collected Poems, 1909–1962 (London: Faber and Faber, 1963), 82. 26. For more on the Eliot’s working title, see Craig Raine’s video essay in The Waste Land (London: TouchPress, Faber and Faber, 2011). 27. Adam Hammond and Julian Brooke, “He Do the Police in Different Voices: Exploring Voices in T. S. Eliot’s The Waste Land,” 2012, http://hedothepolice.org /. 28. Students provided their readings of the poem before being “taught” the poem in lecture. Most students enrolled in the class had previously studied the poem in their fi rst-year introductory classes. While their readings were not colored by the lectures in “The Digital Text,” we cannot discount the effect of earlier instruction. 29. Julian Brooke, Adam Hammond and Graeme Hirst, “Distinguishing Voices in The Waste Land Using Computational Stylistics,” Linguistic Issues in Language Technology 12.2 (October 2015): 1–43. 30. The actual size of the “window” is given as w in our paper “Distinguishing Voices in The Waste Land Using Computational Stylistics.” 31. The poems are W. H. Auden, “September 1, 1939”; Rupert Brooke, “Wagner”; T. S. Eliot, “The Love Song of J. Alfred Prufrock”; D. H. Lawrence, “Ballad of Another Ophelia”; Mina Loy, “Giovanni Franchi”; Wilfred Owen, “Strange Meeting”; William Shakespeare, “How Should I Your True Love Know?” (Ophelia’s song from Hamlet ); Stevie Smith, “Not Waving but Drowning”; Edmund Spenser, “Epithalamion”; Algernon Charles Swinburne, “Before the Beginning of Years”; Alfred, 76 A. HAMMOND ET AL.

Lord Tennyson, “The Coming of Arthur”; and Dylan Thomas, “A Saint About to Fall.” 32. For a more detailed explanation, see Brooke, Hammond and Hirst, “Distinguishing Voices in The Waste Land Using Computational Stylistics.” 33. Michael Levenson, A Genealogy of Modernism , (Cambridge: Cambridge University Press, 1984), 36. Though Levenson initially notes that a switch seems to occur between lines 7 and 8, and not between lines 4 and 5, his larger point is that the poem’s “overlapping principles of similarity undermine the attempt to draw boundaries around distinct speaking voices” and that “we can say with no certainty where one concludes and another begins” (171). We agree with Levenson that no defi nitive boundaries or certain conclusions can be drawn, but argue that the act of devising a reading—however provisional—is a worthwhile critical act, particularly in a classroom setting. 34. Jewel Spears Brooker and Joseph Bentley, Reading The Waste Land: Modernism and the Limits of Interpretation (Amherst: University of Massachusetts Press, 1990), 140. 35. Brooke, Hammond and Hirst, “Distinguishing Voices in The Waste Land Using Computational Stylistics.” 36. Julian Brooke and Graeme Hirst, “A Multi-Dimensional Bayesian Approach to Lexical Style,” Proceedings of the 13th Annual Conference of the North American Chapter of the Association for Computational Lingusitics , 2013, 673–79. 37. Julian Brooke and Graeme Hirst, “Hybrid Models for Lexical Acquisition of Correlated Styles” Proceedings of the 6th International Joint Conference on Natural Language Processing (2013): 82–90. 38. For details of the Project Gutenberg process, see Brooke, Hammond and Hirst. “Using Models of Lexical Style to Quantify Free Indirect Discourse in Modernist Fiction.” Digital Scholarship in the Humanities 2.2 (Advance Access, 3 February 2016). 39. Julian Brooke et al., “Unsupervised Multiword Segmentation of Large Corpora Using Prediction-Driven Decomposition of N-Grams” Proceedings of the 25th International Conference on Computational Linguistics (2014): 753–61. 40. Calvin Bedient, He Do the Police in Different Voices: The Waste Land and Its Protagonist (Chicago: University of Chicago Press, 1986). 41. For a detailed investigation of the “multiform” (140) character of Tiresias, see Jewel Spears Brooker and Joseph Bentley, Reading The Waste Land. 42. Among the critics to accuse Eliot of stereotyped representations of female characters are Sandra Gilbert and Susan Gubar, who argue that Eliot “transcribe[s] female language in order to transcend it” (No Man’s Land: The Place of the Woman Writer in the Twentieth Century , vol. 1: The War of the Words [New Haven: Yale University Press, 1988], 236). For more MODELING MODERNIST DIALOGISM: CLOSE READING WITH BIG DATA 77

recent efforts to reassess Eliot’s representation of gender, see Cassandra Laity and Nancy Gish, eds., Gender, Sexuality and Desire in T. S. Eliot (Cambridge: Cambridge University Press, 2004) and Rachel Potter, “Gender and Obscenity in The Waste Land ,” The Cambridge Companion to The Waste Land , ed. Gabrielle McIntire (Cambridge: Cambridge University Press, 2015), 133–46. 43. Adam Hammond and Julian Brooke, “The Brown Stocking: Exploring Voices in Virginia Woolf’s To the Lighthouse,” 2013, http://brownstocking.org /. 44. See the “What the Computer Said” section of brownstocking.org for details and examples. 45. Graham Pechey, Mikhail Bakhtin: The Word in the World , Critics of the Twentieth Century (London ; New York: Routledge, 2007), 208. 46. Pechey, Mikhail Bakhtin . 47. Brooke, Hammond, and Hirst, “Using Models of Lexical Style to Quantify Free Indirect Discourse in Modernist Fiction.” 48. Majority agreement can produce highly reliable interpretations even when the diffi culty of the task results in only moderate inter-annotator agreement. See Beata Beigman Klebanov and Eyal Beigman, “From Annotator Agreement to Noise Models,” Computational Linguistics 35, no. 4 (2009): 495–503. 49. While a sample of two texts is of course very limited, the fact that FID functions so similarly in two such dissimilar texts—a novel and a short story; an experimental and pervasive employment of the device versus a limited and more conventional one; one text by a female English writer, the other by an Irish male—suggests that the “in-betweenness” of FID will be found to apply more generally. 50. See, for instance, Mary M. Childers, “Virginia Woolf on the Outside Looking Down: Refl ections on the Class of Women,” Modern Fiction Studies 38.1 (Spring 1992): 61–79 and Alison Light, Mrs. Wolf and the Servants (New York: Bloomsbury, 2008). More balanced and sympathetic assessments of Woolf’s representation of class can be found in Melba Cuddy-Keane, Virginia Woolf, the Intellectual, and the Public Sphere (Cambridge: Cambridge University Press, 2003), especially pp. 52–4 and 100–6, and Jean Mills, “Virginia Woolf and the Politics of Class,” A Companion to Virginia Woolf , ed. Jessica Berman (London: Wiley Blackwell, forthcoming 2016), 219–32. 51. This feature is now publicly available as part of our software package, GutenTag , available at www.projectgutentag.org . See Julian Brooke, Adam Hammond, Graeme Hirst, “GutenTag: an NLP- driven Tool for Digital Humanities Research in the Project Gutenberg Corpus,” Workshop on Computational Linguistics for Literature (North American Association for Computational Linguistics, June 2015): 1–6. CHAPTER 4

Mapping Modernism’s Z-axis: A Model for Spatial Analysis in Modernist Studies

Alex Christie and Katie Tanigawa

Frederic Jameson’s proclamation that “our psychic experience, our cultural languages, are today dominated by categories of space rather than by categories ozf time, as in the preceding period of high modernism” 1 both characterized and shaped modernist studies’ focus on temporality. As Andrew Thacker notes in Moving Through Modernity: Space and Geography in Modernism , however, space is also an important, though frequently overlooked, consideration for modernist texts. Thacker argues that to analyze modernist works “is to consider the material spaces these texts discuss, and to disclose how the representational spaces of these texts refl ect, contest or endorse the geographical shaping of these topoi by various ideological representations of space.” 2 In other words, literary representations of space, particularly modernist representations of space, exist in conversation with geographical, cartographical and historical representations of these spaces. This chapter brings such scholarly attention together with a renewed consideration mapping modernist space through various digital means (represented by the myriad digital scholarship that map characters’

A. Christie () Brock University, St. Catharines, ON , Canada e-mail: [email protected] K. Tanigawa Brock University, Victoria, BC, Canada

© The Editor(s) (if applicable) and The Author(s) 2016 79 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_4 80 A. CHRISTIE AND K. TANIGAWA movements in modernist texts). 3 It presents a digital, three-dimensional mapping methodology—what we at the Modernist Versions Project call z-axis research—that allows for the analysis of literary, cartographical and historical representations of spaces that shape (and are in turn shaped by) the modernist texts that write them. In this chapter specifi cally, we restrict our gaze to outline the rationale, methodology, and interpretive fi ndings from maps of Djuna Barnes’s Nightwood and Jean Rhys’s Quartet to show one way the confl uence of digital humanities, modernist studies, and spatial analysis can productively expand literary interpretations of texts. Z-axis research scans or takes an already scanned image of an archival city map contemporary to the setting of a modernist text and warps the map in three dimensions according to the number of words narrated within a given location in the text. The warped map creates the peaks and valleys of literary representation and reveals how the text rewrites the modern city in ways that often alternately reify or contest the city’s popular, contemporary representations. In short, z-axis maps express the reader’s spatial experience of the modern city described by a given novel. Rather than primarily mapping where the action of a novel takes place, z-axis maps additionally emphasize subjective, embodied, and marginal experiences of the modern city, enabling critical attention to geography’s role in modernist literary expression. Thus far, we have deployed the z-axis methodology to model the Dublin of James Joyce’s Ulysses , the Paris of Jean Rhys’s Quartet and Djuna Barnes’s Nightwood (Fig. 4.1 ), and the London of Virginia Woolf’s Mrs. Dalloway. 4 Here, we discuss our fi ndings for Paris, focusing on a comparative geospatial reading of Jean Rhys and Djuna Barnes’s representations of marginal experiences of interwar Paris. Across both novels, the z-axis maps reveal each text’s spatial arguments concerning class, gender, sexuality and race that often rub against the ideologies of that space as represented by the 3D maps themselves.

MAPPING THE MODERNIST CITY Z-axis mapping anchors the geographic references from modernist novels in contemporary maps that express the culture to which that novel responds. By “z-axis,” we refer to socio-political experiences represented as a third axis of inquiry underpinned by longitudinal and latitu- dinal expressions of geographic data. Broadly defi ned, a z-axis approach expresses data through digital objects that render culture and politics as physical properties. Every map (z-axis or otherwise) comes with its own spatial and cartographic codes, its own procedures for representing and MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 81

Fig. 4.1 Z-axis map of Djuna Barnes’s Nightwood navigating space: these codes are designed, inscribed and navigated. They are legible in the material arrangement of a map and correspond to historically situated understandings of space in accordance with ethnicity, politics and class—in other words, culture. Under Western regimes of mapping, perspective is often visual as well as cultural: different maps offer different cultural perspectives on a given geography, reproducing power and politics at the level of design. As J. B. Harley explains this process in “Deconstructing the Map,”

The distinctions of class and power are engineered, reifi ed and legitimated in the map by means of cartographic signs. … Using all the tricks of the cartographic trade—size of symbol, thickness of line, height of lettering, hatching and shading, the addition of color—we can trace this reinforcing tendency in innumerable European maps. We can begin to see how maps, like art, become a mechanism ‘For defi ning social relationships, sustaining social rules, and strengthening social values.’ 5

Deploying maps as a site of geospatial interpretation therefore requires anchoring cultural and political critiques in historical expressions of such processes. While scholars cannot directly access the politics of the past, we can re-engage graphic expressions of those politics. 6 Because one of the principles of z-axis research is to visualize this conversation between 82 A. CHRISTIE AND K. TANIGAWA historical and literary space, we use archival maps contemporary to the setting of the novels in question. Remaining as close as possible to the temporal setting of the novel allows scholars to readily see the relationships among historical, cartographical and literary representations of space. As cartography scholars have long understood, maps, like texts, provide political and socially constructed renderings of the spaces they represent. Z-axis research unites cartographical theories with literary study to query the relationship between both representation systems (cartographic and textual). Z-axis research unites such cartographical theories with literary studies to produce maps that acknowledge the role socio-political conditions and textual representations play in representing experiences of space. To query how Barnes and Rhys’s descriptions of Paris respond to dominant patterns of urban circulation at the time, we expressed the geographic data for their novels through a popular interwar tourist map (Fig. 4.2 ).

Fig. 4.2 Paris Monumental et Métropolitain map (1932) (Note: Image is under public domain and is reproduced under a Wikimedia Commons License ( http:// commons.wikipedia.org )) MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 83

In so doing, we discovered that the map deploys visual warping and a two-and-a-half dimensional perspective to embed a capitalist logic into its cartographic expression. The Nouveau Paris Monumental map series 7 was a ubiquitous tourist map of Paris used in the nineteenth and twentieth centuries and included in pocketbook city guides. 8 The z-axis map uses an entry from the series made during the interwar period; its estimated publication date is 1932. Implementing a 2.5 dimensional perspective, the map eschews a unifi ed two-dimensional or three-dimension view of Paris in favor of a design that confl ates both visions. Monuments are viewed from a false-three dimensional perspective as they sit atop the fl at surfaces of Paris’s major boulevards and metro lines. The vertical shrinking of Parisian space occurs in tandem with the horizontal warping of Paris itself—areas surrounding key Parisian monuments appear larger on the map than they exist in Paris, whereas streets and areas that do not generate revenue are either shrunk or ignored altogether. Comparable to the novels we are studying, the map does not depict Paris as it actually existed at the time, but instead a partial and biased representation of a “monumental and metropolitan Paris.” For instance, the bulk of Barnes’s Nightwood occurs in a street just around Saint Sulpice and south of the Boulevard Saint-Germain—an area that barely even exists on the Monumental Map (Fig. 4.3 ). Barnes’s literary transformation of Paris confl icts with the cartographic warping seen in the map, transforming Parisian geography in accordance with the novel’s stratifi cation of class and sexuality. The z-axis maps not only challenge these ubiquitous, isomorphic and authoritative depictions of Paris; they also acknowledge Paris’ history as a city deeply enmeshed in experimental methods for remapping cultural currents through space. As modernist scholars such as Lara Schrivjer have

Fig. 4.3 Detail from map of Paris: Saint-Sulpice 84 A. CHRISTIE AND K. TANIGAWA recently documented, the commercialization and commodifi cation of the modern city was picked up by both Situationists and Surrealists, 9 who experimented with methods for remapping commercial expressions of the city. André Breton’s 1933 “Experimental Researches: On the irrational embellishment of the city” 10 extends the remapping of urban geography from a literary technique to a methodological practice. This surrealist game asked participants to conserve, displace, modify, transform or remove monumental areas in city using their imagination. Directly inspired by Breton, Situationist practitioners extended the modifi cation of a city to a built media practice in the 1950s and 1960s. Situationist psychogeog- raphy promised to re-map cultural currents of the city by cutting-up and rearranging different cartographic representations of that city. As Simon Sadler explains,

Rather than fl oat above the city as some sort of omnipotent, all- possessing eye, situationist cartography admitted that its overview of the city was reconstructed in the imagination, piecing together an experience of space that was actually terrestrial, fragmented, subjective, temporal, and cultural. … In short, the situationist maps described an urban navigational system that operated independently of Paris’s dominant patterns of circulation. 11

We have revisited this strain of experimental practice to push against top- down and realist geospatial epistemologies that normalize space to the exclusion of marginal perspectives. We strive instead to extend modernist experiments in remaking the city to contemporary digital practice, a z-axis approach that unpacks the social and cultural depth of archival maps that are otherwise read as only surface or image. In “Humanities Approaches to Graphical Display,” Johanna Drucker’s description of visualizing ambiguity resonates with these surrealist and situationist constructions of space; she calls for digital expressions of subjective and constructed data, explained as “the difference between putting many kinds of points on a map to show degrees of certainty by shades of color, degrees of crisp- ness, transparency, etc., and creating a map whose basic coordinate grid is constructed as an effect of these ambiguities.” 12 Visualizing ambiguity and indeterminacy as a core element of spatial experience is thus a chief aim of our maps. We understand the modernist city as mutable and n-dimensional in nature, as a multiplicity of overlapping and interfolding cities, rather than a singular or essential geographic space. This approach invites investigation into modernist literature’s representations of urban MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 85 life, comparing multiple spatial accounts of the same historical and geographic city. 13 And crucially, the subjective experience of modernist narration, as it constructs multiple situated and partial expressions, produces the multiple maps through which our z-axis readings operate.

METHODOLOGY: MARKUP Our geospatial approach to the texts we work with is directly inspired by modernist experiments in mapping and warping visual representations of the modern city to visualize spatial and subjective experience. While our mapping methods enact a theoretical approach to modernist geospatial expression, so too do our methods for working with the texts we map. In other words, we take seriously Lev Manovich’s notion that “a prototype is a theory” 14 and Alan Gaeley and Stan Ruecker’s appeal to digital humanists to accept that “the design of artifacts is a critical act” 15 in itself. In the case of z-axis research, the very mapping and warping of archival maps forward arguments about the connection between modernist literature and space and challenges the primacy digital humanities has placed on realist representations of texts. It does so by enumerating the ways space and modernist aesthetics mutually infl uence one another and by arguing that this mutual infl uence can and should be considered in our studies of the two. To generate the warped maps, we use a bifurcated process of textual markup and three-dimensional modeling of the historical map. We begin the markup process and work with an archival quality (600 dpi) scanned edition of a text. The digital scans are then run through Adobe’s Optical Character Recognition (OCR) software to generate machine-readable text. From here, we mark up the text in the XML editor oXygen, using extensible markup language (XML) to manually identify where spans of text are located. Locations are encoded both into an XML markup of the novel and an SVG copy of the archival map (with locations encoded through pixel location on the map, rather than longitude and latitude coordinates) (Fig. 4.4 ). 16 The marked-up text is then run through an extensible stylesheet language transformation (XSLT) that shows the number of alphanumeric characters located in each researcher-identifi ed location in the text. These numbers are placed into a spreadsheet, checked for accuracy, and reduced to a ratio of the total number of words in the text (Fig. 4.5 ). This fi nal ratio is what researchers use in the second stage of the methodology: mapping. 86 A. CHRISTIE AND K. TANIGAWA

Fig. 4.4 XML markup of text and map-based locations

Fig. 4.5 XSTL-transformed geo-data of Paris MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 87

Though our markup is fairly minimal—limited to identifying shifts in the location of the narratives—the process requires refl ection and guidelines. What constitutes a place? How do we mark a place that is not named or a place that exists only in the novel? How do we mark movement? For example, if a character is taking a cab, as Marya in Quartet often does, how do we know what parts of the narrative take place at which points in the cab ride? When a character imagines a scenario that is located in a place other than where they currently are, how do we identify where the narrative occurs? While many of these decisions we made are based on the needs of each individual novel, we devised a few guidelines. For instance, our defi nition of place for the markup was liberal. While a place would ideally be mappable on our “Landmarks of Paris” map, it seemed equally important to identify the unmappable places—for example, the slew of unnamed cafés in Quartet . Even if the data from these unmappable spaces never altered the z-axis maps, the importance of unidentifi able and vague spaces to Rhys’s work in particular, 17 warranted the acknowledgement of these places if only through their demarcation as place= “unknown”. Marking narrative movement, however, proved more diffi cult than deciding what constituted a place. This decision was subject to the discretion of each encoder. A strategy used to mark up Quartet was to assume (faultily) that the narrative encapsulated in the geographical journey occurred evenly between the beginning and end of the journey. While questions about the defi nition of place and movement shaped our markup, the most formative question was whether to privilege a character’s imagined space or his/her physical space. If a character imagines a scenario that occurs in a space other than where the character is physically, do we mark the place where the imagined scenario occurs or where the character is physically? In order to avoid privileging perceived “real” spaces in these novels, we decided to mark narrative space as opposed to character location in instances in which these two confl icted. Wherever the narrative goes, we mark. This decision became our guiding markup practice that led to the formation of the warped maps.

METHODOLOGY: MAPPING AND 3D MODELLING To express the two-dimensional features of the tourist map in 3D, the scanned historical map is converted into a displacement map, a process that heightens the black and white contrasts of the map to allow an algorithm to produce a three-dimensional map wherein the white regions 88 A. CHRISTIE AND K. TANIGAWA are raised. Displacement mapping is a computer graphics technique that uses a two-dimensional image to displace the geometric position of points on a three-dimensional surface. It uses differences in image contrast to calculate differences in elevation, adding depth to a fl at surface based on the color of the image (displacement map) applied to it. Converting the scanned map into a displacement map thus requires (fi rst) converting its color mode to grayscale and (second) increasing the image contrast, making blacks deeper and whites brighter (such that the displacement mapping algorithm can pick up on and calculate differences in shading). The displacement mapping algorithm raises the white areas of the ordnance map vertically, while depressing the areas in black. Shades of grey in-between those two extremes register as different elevations on the surface of the 3D map. In this sense, the displacement map is not so much stamped onto the 3D surface as the surface itself is displaced vertically around the map image. The 3D map thus turns differences in black/white contrast into differences in elevation, molding three-dimensional surfaces around the two-dimensional topography of the map image. This produces a three-dimensional map mesh, which can then be transformed using geo-data from a given novel (Figs. 4.6 , 4.7 , and 4.8 ). Each area on the map is located using the pixel location from the SVG map, with the radius taken from the XML coordinates embedded in the SVG fi le. The signifi cance ratio is then applied to determine the precise strength of the warping effect, applied to the 3D map mesh using either the bulge or the warp function using the 3D

Fig. 4.6 Displacement map of Paris MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 89

Fig. 4.8 Warped 3D map mesh of Paris

Fig. 4.7 3D map mesh of Paris modeling software Autodesk Mudbox. The result is a warped three- dimensional map that expresses the geography of a given novel through the transformed Nouveau Paris Monumental map. 18 We have embedded into this research methodology a layer of self- critique by testing the process with different modernist novels, presenting ourselves with limit cases, and regularly re-evaluating and changing the process. Beyond this self-critique, however, this research also answers Alan Liu’s call for digital humanities work that moves beyond critiquing the “tools, data, metadata” that these projects themselves form to critique “society, economics, politics, or culture” as well. 19 We map to render the 90 A. CHRISTIE AND K. TANIGAWA peaks and valleys (in other words, a three-dimensional indication of the various spaces a text might dwell on, avoid, shift, shrink, expand, invent or erase) of literary representation to show the multiple spatial narratives from and about the communities that inhabit these places. In the process, the maps reveal the ways modernism marginalizes, and in turn brings to the fore, key areas and key communities in the cities they chart. Understanding how modernist texts imagine and represent the spaces they inhabit requires more than just the map. To this end, the methodology considers Wendy Chun’s assessment of mapping in Programmed Visions as “the beginning rather than the end, of analysis.” 20 Z-axis research queries and accesses intersectional constructions of space by anchoring them in contemporary cultural materials. Deploying the Nouveau Paris Monumental map as a site of spatial interpretation anchors Barnes’s and Rhys’s marginal Paris in an historical expression of the normative patterns of urban circulation to which they responds. From this perspective, the maps generated from our computational reading of the text serves as but a starting point for the interpretive analysis at which humanities scholars excel.

DJUNA BARNES’S QUEER LATIN QUARTER Nightwood ’s queer Paris is constructed through the subjective and embodied opposition between normalized and marginal experiences of the city. From the fi rst page of Barnes’s 1936 novel, geography serves as a narrative code for both constructing and calling into question the identity of her characters, specifi cally through distinctions of class and sexuality. The opening of the book describes the birth of a young child, Guido, whose father believes him destined for nobility. The child’s desired noble status is written onto his newborn body in the novel’s opening scene. Guido’s mother, Hedvig is described “lying upon a canopied bed of a rich spectac- ular crimson, the valance stamped with the bifurcated wings of the House of Hapsburg, the feather coverlet an envelope of satin on which, in massive and tarnished gold threads, stood the Volkbein arms … .” 21 While Hedvig is physically situated alongside markers of nobility, the following scene confl ates Guido’s birth with the geographic landscape that his nobility is meant to possess: “Turning upon this fi eld, which shook to the clatter of morning horses in the street beyond, with the gross splendor of a general saluting the fl ag, she named him Felix, thrust him from her, and died.” 22 Guido’s birth facilitates a geographic confl ation between Hedvig’s position amongst markers of nobility with her vision of the land over which her son MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 91 is meant to inherit military ownership, enacting Guido’s supposed noble status through the spatial description of the novel. Such confl ations of character and setting structure class distinctions in Barnes’s novel; upper- class characters possess and control geographic territory, imprinting their title onto the land and tying the reproductive renewal of their title to continued possession of that land. Marginal characters instead mask their identities in un-owned public spaces, assembling spaces of resistance to divorce sexual activity from reproductive ties to land. By way of contrast to the newborn Guido, Barnes’s Doctor Matthew O’Connor (the queer doctor who alternately identifi es as homosexual or inverted) lives near the Saint-Sulpice church, but describes himself as inhabiting a collection of hotels, cafes and restaurants in and around this area titled the “Doctor’s city.” By constructing this alternate, marginal version of the Latin Quarter, the Doctor constructs a space to articulate his marginal sexual identity. For Barnes’s impoverished and homosexual doctor, geographic possession does not support and sustain a heteronormative identity; instead, the lack of ownership of public places affords the expression of an indeterminate queer identifi cation. In one encounter in the café de la Mairie du Vie (his local haunt), the doctor makes the case for space as a key aspect of the experience of queer Paris, arguing, “If you think that certain things do not show from what district they come, yea, even to an arrondisement , then you are not out gunning for particular game, but simply any catch, and I’ll have nothing to do with you!” 23 For Barnes’s doctor, the anonymity of public yet overlooked Parisian places enables the construction of a space of male queerness. Reading Barnes’s novel through z-axis mapping anchors close critical readings of Barnes’s queer Paris in geographic expressions of Parisian space. Rather than indicating where queer activity takes place for Barnes, this approach instead considers how place itself functions to construct and express queer experience in the text. 24 For Barnes, the queer experience of interwar Paris is one highly infl ected by distinctions of class. The aesthetic features of the warped map reveal not only a Paris structured through distinctions of class and sexuality, whose markers of identity are fundamentally enmeshed in the spatial experience of Parisian geography. The warped map visualizes Barnes’s Paris as sharply divided by class, with cumulative warping effects visible in and around the Latin Quarter on the Left Bank and granular, isolated warping near the Champs-Élysées and the Opera on the Right Bank (Fig. 4.9 ). These aesthetic differences correspond to class distinctions that mark different narrative encounters in Barnes’s novel: episodes taking place on the Right Bank 92 A. CHRISTIE AND K. TANIGAWA

Fig. 4.9 Z-axis map of Djuna Barnes’s Nightwood almost exclusively involve Jenny and take pace at distinct locations (in the Opera, at Jenny’s house, etc.). On the other hand, while isolated encounters between lower-class characters do take place on the Right Bank (the doctor’s fi rst encounter with Robin occurs in the hotel Recamier, for instance), Barnes’s queer characters further wander in and around the Latin Quarter, where encounters often occur at indeterminate locations and intersections, on sidewalks and in nearby unidentifi ed cafés. Rather than being a marker of realist geographic fi delity, indicating that Barnes or the reader does not know where these encounters take place, it is precisely the fact that the locations of these encounters are unstable and indeterminate that is central to their status as queer. Lawrence Schehr explains how the indeterminacy of public space was central to the queer experience of modern Paris, writing:

[T]he sheer size of the city, the accessibility of public transportation, the number of people, and the general heterogeneity of the population allowed for a kind of freedom or opened to live (disponibilité , to use a word central to Gide), at least of the male individual, not present in the country. People living in a village or even a small town could deduce where someone else was going simply by observing the path he or she took. No such prediction could be made in Paris. 25 MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 93

This non-realist reading of Parisian space reveals public and indeterminate locations as zones of possibility, whose lack of determinacy is precisely what enables the possibility for actively pursuing queer desire. It in turn affords a corollary re-reading (or re-viewing) of the z-axis map; rather than simply indicating that Barnes’s references on the Left Bank tend to be more vague, the map reveals the spatial coordinates of the Left Bank as fundamentally warped, as a zone in which (hetero)normative conceptions of space are bent and contorted. This non-realist spatial hermeneutic allows a modernist re-viewing of the radius of the coordinates used to produce each warping effect; rather than viewing the radius as a simple visual marker of specifi city, it is instead seen to operate in a Bohrsian, quantum relativistic mode, in which radii indicate zones of possibility, freedom and indeterminacy. 26 The base of the Right Bank remains fl at, neutral and normalized, whereas the base of the Left Bank is fundamentally warped, a space in which defamiliarizing descriptions of space enable codifi ed expressions of queer desire. In the instance of Barnes’s Paris, warped space is both characterized by and produces the conditions for fl uid and unstable identifi cations. Visualizing the social construction of modernist space, then, the z-axis map reveals a Paris whose coordinate grid is fundamentally enmeshed in cultural experience. Since geographic experience is ultimately constructed through spatial representation, there is no stable, normal or real geography to which warped space refers (even normalized space is warped, although the effects of normalization often cause it to appear fl at). Indeed, as already indicated, the Nouveau Paris Monumental Map is warped according to the capitalist construction of tourist Paris: wealthy and profi t-generating areas are expanded on the map, while impoverished and marginal areas are shrunk. Expressing the doctor’s city by warping the monumental city blends material markers of cultural and economic difference, materially expressing the marginality of Barnes’s locations. By cob- bling together proximate interstitial places in order to form a new space, the Doctor constructs his own sense of space and place that operates simultaneously alongside and against the normalized Paris of the Nouveau Paris Monumental map (Fig. 4.10 ). For Barnes’s characters, queer Paris does not emerge solely in, at, or through specifi c locations, but is rather accessed by slipping into spaces of possibility that run just alongside the margins and lacunae of a heteronormative Paris. Wandering, strolling, chatting on the sidewalk, staring with longing into a crowded bar at night (perhaps even slipping between its folds where one’s actions can become 94 A. CHRISTIE AND K. TANIGAWA

Fig. 4.10 The doctor’s city in Nightwood anonymous and untraceable)—these are the modes by which queer Paris is structured, always existing beside another Paris, while still occupying the same geographic space. Both Parises are inextricable in Barnes’s text; the doctor’s city cannot exist without the shadow of the looming Église Saint-Sulpice to mask its contours. Rather than attempting to achieve immediate or unmediated access to the politics of Barnes’s time, z-axis mapping instead reengages material expressions of those politics. In so doing, the digital transformation of cultural and historical materials enables geographically situated readings of how and where Nightwood responds to specifi c spatial practices of interwar Paris. To this end, encoding Barnes’s novel for geographic references does not (primarily) function as a process for identifying places Barnes mentions in the novel, but instead facilitates a hermeneutic that reads place itself as a codifi ed mode of narrative expression. As Amy Wells-Lynn notes, the garden and fountain in the back of Nora’s house in Paris functions as a specifi c, coded reference to the garden and broken fountain at Natalie Barney’s salon. She explains: “Barnes’s garden allusions are equally sexually coded. Salon participants would know that the rue Jacob garden is a scene of lesbian fl irtation.” 27 Identifying scenes that take place at or MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 95 around Nora’s garden therefore does not pursue a realist impulse for geographic accuracy, but instead identifi es instances in which geography is as central to understanding Barnes’s writing as character, plot and dialogue. Marking up Barnes’s novel in XML facilitates a process of decoding rather than encoding; it enables close, detailed attention to modernist writing that deploys geographic codes as markers of queer activity. At the same time, these codes do not primarily reference or identify physical places, but instead exposes space as an expressive medium through which Barnes’s strategically discloses aspects of her characters’ identities. Journeys through the city thus function not solely as narrated travels to a particular destination, but simultaneously serve as events through which specifi c sexual dynamics play out across the encoded landscape of Barnes’s Paris. This is particularly true of the carriage ride scene, during which the wealthy Jenny Petherbridge initiates a sexualized assault on Robin Vote. The title of the chapter in which the scene occurs, “The Squatter” explicitly spatializes sexual desire, characterizing Jenny’s attachment to Robin as squatting on the property of Robin and Nora’s love. 28 Echoing the entanglement of sexuality and geography, the impetus for the carriage ride recalls the geographic possession associated with heterosexual reproduction constructed in the novel’s opening scene. During a party at Jenny’s house, the Marchesa de Spada comments that: “everyone in the room had been doing on from interminable sources since the world began, and would continue to reappear, but that there was one person who had come to the end of her existence and would return no more.” 29 As she says this, the Marchesa glances at Robin who is talking to a child, which causes Jenny to begin trembling and promptly order her company out for a carriage ride. The Marchesa’s heterosexuality is tied to land both through her words and her noble title. Just as a marquis is charged to protect the borders of a kingdom, so does the Marchesa banish Robin from heteronormative ties to the land: Robin’s desire does not allow the possibility for procreation. Reproduction is tied to geography through the knot of class and sexuality, and it is against this backdrop that the carriage ride across Parisian space unfolds. The child that Jenny and Robin cannot produce catalyzes Jenny’s attack on Robin, and it is no mistake that this event occurs while both women are traveling across an indeterminate geographic space (rather than safely inhabiting a fi xed location). Barnes’s geographic references become increasingly vague as the carriage enters increasingly impoverished neighborhoods (Fig. 4.11 ). The journey begins at Jenny’s house, just south of the Champs-Élysées, and continues into the Bois 96 A. CHRISTIE AND K. TANIGAWA

Fig. 4.11 Barnes’s version of the carriage ride in Nightwood de Boulogne, down through the woods, and out into “the lower parts of town” (likely the fi fteenth and sixteenth arrondissements ). It is in the working-class and immigrant “lower parts of town” that Jenny’s outburst occurs, geographically tying her frustrated desire to impoverished markers of class. Spatial expression overtakes narrative as the carriage ride moves into its denouement: the scene does not end with an explicit resolution to the attack, but instead at Nora’s house, where “Robin jumped before the carriage stopped, but Jenny was close behind her, following her as far as the garden.” 30 The subsumption or erasure of narrative into geography overturns a realist paradigm of place-based referentiality, ungrounding plot from place. Accounting for Barnes’s encoded allusion to lesbian desire that ends the scene therefore requires conceiving space as an expressive medium, not seeing Barnes’s novel set in Paris, but instead Barnes’s Paris set in the novel. In the following chapter, the doctor offers a spatial account of the carriage ride scene that differs markedly from that of Barnes. Re-telling his experience of the carriage ride to Nora, the doctor elides sharp distinctions of class from the physical and sexual journey across the Parisian cityscape. He describes the entire event taking place in the Bois de Boulogne, excis- ing the lower arrondissements from the journey altogether. The doctor’s MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 97 inclusion of the journey across the Right Bank, coupled with his elision of the lower parts of town, constructs an alternate connection between Jenny and Robin’s houses. Barnes links the two locations through sharp distinctions in class (traveling south through the Bois de Boulogne), whereas the doctor’s account remains fi rmly on the Right Bank, minimizing geographic markers of class difference from the episode. Across these divergent accounts of the attack, Barnes and the doctor both reshape the geography of Paris to differently infl ect Jenny and Robin’s encounter. Barnes’s account of the journey entangles class and lesbian desire, concluding with a spatially coded reference to Natalie Barney’s garden, while the Doctor disentangles class from desire, concluding not with a spatial code, but a reference to the fi gure of the dog (often used in the novel to register the violent expression of repressed desire). These divergent accounts of the carriage ride scene express diverse and semi-permeable articulations of queer desire: the homosexual doctor does not deploy Barnes’s spatial allusions to lesbian activity and avoids entan- gling class with queerness. While these different accounts do correspond to diverse articulations of queer desire, it is crucial not to suggest they neatly separate across an easy dichotomy of lesbian/homosexual or female/male. As Amy Wells-Lynn warns: “Reevaluating expatriate female space in historical and gendered contexts is important, but we must be careful about essentializing the space or falling into stale male/female dichotomies.” 31 Nightwood is not located neatly within one discrete version of Paris, but instead operates through a variety of mutable and transformable versions and visions of the city that correspond to shifting marginal experiences of urban life. Nightwood’s Paris does not function primarily as a geographic setting in which the novel takes place (although this is certainly the case), but further serves as a malleable, contingent spatial medium through which marginal experiences are expressed. Through its resemblance to the actual geography of Paris, Nightwood ’s confl ation of space and place enables the partial elision of marginal experiences for their very transgressive nature, masking them beneath the guide of geographic realism. 32

JEAN RHYS’S URBAN IMPRISONMENT Quartet traces the struggles of the protagonist, Marya, after her husband is jailed. Without her husband’s income, Marya is destitute. In her desper- ation, she turns to the Mr. and Mrs. Heidler, who offer to house and care for her. Soon, however, Marya becomes Mr. Heidler’s mistress, and her 98 A. CHRISTIE AND K. TANIGAWA

Fig. 4.12 Rhys’s representation of the Observatoire in Quartet manipulation and exploitation at the hands of both Mr. and Mrs. Heidler lead to her unraveling. The z-axis map of Quartet shows at least three specifi c ways in which the city of Paris refl ects the narrative. First, the map pinpoints where the narrative privileges spaces marginalized on the historical base map, such as the areas surrounding the monuments. More importantly, the map also shows how the geography of Paris changes to refl ect Marya’s evolving psyche. Finally, the map reveals a Paris that becomes a gendered and classed prison—a veritable hell for the female protagonist— and produces a reading that rubs against the notion that Marya constructs Paris as a “feminine city.” While the Monumental Map privileges monuments, Rhys’s narrative instead privileges spaces obscured and marginalized by this popular map. In Fig. 4.12 , one can see a large mound representing a signifi cant number of words spent around the Observatoire. What is crucial to note is that the none of the narration is located in the Observatoire. Rather, all the narrative located in this region takes place in the Heidlers’ studio, which is behind this popular landmark. Another key example of Marya’s construction of the Parisian city can be seen in her presence around—but never in —the Luxembourg gardens. As the map shows, she spends considerable amounts of time in cafés in the area, MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 99 but her diminishing social and class status due to her gender and to her sexual practices bar her from the classed spaces privileged in the 1930s monumental map. In other words, this map shows how Rhys’s narrative privileges spaces that are traditionally marginalized in social and economic discourse and that are in turn inscriptively marginalized in popular maps at the time, revealing her work in conversation not just with other early twentieth-century novels, but with the ideologies that infl uenced actual mapping practices of her time as well. Using z-axis analysis on Quartet also presented the question of how Rhys engages with and breaks from totalizing mapping practices that create what Richard Zeikowitz calls the “masculine city.” 33 Zeikowitz uses Le Corbusier’s detailed systematic description of Paris’ streets as a quint- essential example of the literary, masculine city in order to defi ne Rhys’s construction of what Zeikowitz calls a “feminine city” (Zeikowitz 1). Instead of describing, with measured accuracy, the angles of streets and precise intersections, Zeikowitz notes:

The Paris Marya writes undermines Le Corbusier’s masculine-based cityscape[. …] Marya does not layout ordered streets and right angles, “governed by the principles of geometry.” She does not set out to defi ne Paris, to enclose it and fi x its meaning; rather, she creates it as she experiences it. 34

Zeikowitz’s observation reveals one of the key diffi culties of mapping Quartet: the lack of order and fi xity refuse or certainly make diffi cult the process of pinning parts of the narrative to a single point in the city. This is most clearly seen in Marya’s vague wanderings along the Boulevard Saint Michel or the Boulevard du Montparnasse. Deborah Parsons’ analysis of Rhys’s treatment of space in her work certainly aligns with Zeikowitz’s assessment of Rhys’s transgressive narration of space. In Street Walking the Metropolis , Parsons writes about the relationship between space, gender and modernism and notes that for Rhys’s works especially, “the streets of the city and the pilgrimage through them become internalized, as city and psyche become one.” 35 In the context of mapping Rhys’s works, this not only means that her characters’ psyches are revealed through the cities they inhabit; it also means that the cities themselves are shaped by her characters’ psyches. Parsons’ observation about the connection between space and psyche resonates particularly for Marya, as shown in the map below (Fig. 4.13 ). This map shows two distinct clusters of narrative engagement. One cluster appears around Montmartre: the area of Paris fi rst inhabited by 100 A. CHRISTIE AND K. TANIGAWA

Fig. 4.13 Map of Marya’s narrative engagement with Paris in Quartet

Marya and the only space where she is ever “very near to being happy.” 36 However, upon the imprisonment of her husband, Marya descends from this wealthier section of town to live on the more bohemian and less wealthy Left Bank with the Heidlers. The physical movement of the narrative not only charts the economic descent of Marya into poverty; it also refl ects her increasingly fragmented consciousness. Whereas maps of other novels, such as Nightwood , show how the narrative links one location to the other even just with slight raises of the map that connect the key, raised areas, Quartet provides few links between Marya’s life on the Right Bank and the Left. This geographical fragmentation articulated by the narrative and depicted on the map mirrors Marya’s increasing discon- nection from her life in Montmartre, as well as Marya’s growing sense of psychological fragmentation. In other words, the z-axis map shows that Marya’s attempt to create a “feminine” Paris (in the sense invoked by Zeikowitz and Parsons) is ultimately overwhelmed by a Paris that becomes a gendered prison and “a hostile, alienating urban environment” 37 for the female protagonist. Despite Marya’s attempt to create a feminine city through her wanderings, the places where the narrative lingers are both signifi cant both in terms of duration and frequency (as illustrated by the map) and in terms of meaning. In Fig. 4.14 the largest mound shows that the bulk of the narrative takes place around the Place Denfert-Rochereau, which hosts the MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 101

Fig. 4.14 Quartet map showing place Denfert-Rochereau entrance to the catacombs, but more signifi cantly, was fi rst named Place d’Enfer—the place of hell. The setting is particularly appropriate given Marya’s movement from relative happiness prior to Stephan’s arrest as indicated by the narrator’s description of Marya’s life with Stephan: “It was a fantastic life, but it kept on its legs so to speak. There was no catastrophe. And eventually Marya stopped questioning and was happy.” 38 After Stephan’s imprisonment, Marya’s life shifts to one of despair and inner turmoil. The imagery in the narration that follows Stephan’s arrest refl ects this turn with words like “desolation” 39 and “sinner,” 40 and visions of torture 41 and madness. 42 Lois Heidler even remarks that if Marya were to leave the Heidler’s studio, which she eventually does, she would have “gone to the devil.” 43 In fact, after Marya leaves the Heidlers’s studio to live in a hotel room paid for by Mr. Heidler, Marya says of her new accommodations, “A bedroom in hell might look rather like this one.” 44 Without the context of her personal struggle or the imagery, Place Denfert-Rochereau never reforms back into Place d’Enfer, and the city of Paris never houses the “place of hell.” Just as Marya does not escape the geographical confi nes of Paris, she never escapes her hell in the novel. Thus, Paris not only becomes a kind 102 A. CHRISTIE AND K. TANIGAWA

Fig. 4.15 Quartet map showing the Palais de Justice and Santé Prison of hell, but a kind of jail as well. This fi gurative incarceration as a result of her gendered poverty and exploitation is also mirrored by the amount of time the narrative spends in literal, juridical penitentiaries, as shown by the substantial mounds where the Palais de Justice and Santé Prison are located (Fig. 4.15 ). What the map does not show is that these mounds represent events that occur early in the novel. As the novel progresses, a switch occurs: the juridical prisons become places for freedom and comfort for Marya, 45 whereas her experiences in the hotels and streets of Paris are increasingly described in terms of imprisonment. As she reduces her visits to her husband in jail, and as her struggles with the Heidlers grow, the novel invokes images of caged animals 46 and jail cells. 47 In other words, as Marya fi nds herself increasingly controlled and exploited by the Heidlers, her orientation in these spaces are not only refl ective of her experiences, but transform these spaces on the z-axis map accordingly. In a testament to the gendered nature of transnational and economic mobility, by the end of the novel, Marya languishes in her prison and hell while her male counterparts escape. Her husband, exiled from Paris, fl ees to South America to fi nd MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 103 employment; Heidler escapes from the adulterous relationship relatively unscathed. In other words, through the trajectory of the narrative and the spaces Rhys privileges, we see a new Paris emerge: one that exists in stark contrast both to the popular images of a sexually liberated and bohemian city of lights and to the Paris Monumental et Métropolitain Map. In contrast to popular Rhys’ Paris condemns the protagonist who breaks gendered sexual norms. Instead of a Paris constituted primarily of public monuments and a rich cultural history, we see a Paris constituted by poverty and a mentality of class exclusion.

MODERNIST PARIS ON THE Z-AXIS Beyond these interpretations of Nightwood and Quarte t, what our analyses ultimately reveal is that z-axis research envisions place as more than coordinates; instead, place, as geographer Doreen Massey notes, is both “the product of interrelations; as constituted through interactions, from the immensity of the global to the intimately tiny” and “the sphere of the possibility of the existence of multiplicity.” 48 Like the modernist literature we study, spaces themselves contain multiple meanings and layers that call for reading and analysis. In this sense, spaces transform into other texts, and as texts, spaces shape, and are in turn shaped by, the works in which they exist. The metaphors, images, and meanings attached to space culminate to produce a new version of the city, one whose mounds of representation and valleys of elision infl ect our experiences as readers of maps, text and cities. Moreover, what makes modernist literary space particularly interesting, diffi cult and important to visualize is that it is not simply constructed by either the text or its geographical referent. Rather, literary space springs from the conversations between its simultaneous formation by the text and the multiplicities of histories that constitute its existence beyond the text. In turn, the tool exposes digital objects to modernist subjects, revealing their status not as realist referents to objective truth, but rather interpretive objects that invite a plurality of humanist visions.

NOTES 1. Fredric Jameson, Postmodernism, Or, the Cultural Logic of Late Capitalism (Durham: Duke University Press, 1991), 16. 2. Andrew Thacker, Moving through Modernity: Space and Geography in Modernism (Manchester: Manchester University Press, 2003), 21. 104 A. CHRISTIE AND K. TANIGAWA

3. Z-axis research builds from the pioneering work of spatial modernist studies projects such as the WatsonWalk App, Mapping Memoirs of Montparnasse, and Walking Ulysses, as well as projects such as Neatline and the Map of Early Modern London. 4. This research has been undertaken with the Modernist Versions Project (MVP) and Implementing New Knowledge Environments (INKE) teams in the Electronic Textual Cultures Lab (ETCL) and the Maker Lab at the University of Victoria. 5. John Brian Harley, “Deconstructing the map,” Cartographica: The international journal for geographic information and geovisualization 26.2 (1989): 1–20. 6. Harley writes: “In the map itself, social structures are often disguised beneath an abstract, instrumental space, or incarcerated in the coordinates of computer mapping. And in the technical literature of cartography they are also ignored, notwithstanding the fact that they may be as important as surveying, compilation, or design in producing the statements that cartography makes about the world and its landscapes. Such an interplay of social and technical rules is a universal feature of cartographic knowledge”. Ibid., 8. 7. Paris Monumental et Métropolitan. Map. 1:2000 Paris: Robelin. 1932. 8. A 1946 instantiation of the Nouveau Paris Métropolitan map comes folded inside A. Leconte’s pocketbook guide Plan Commode de Paris , housed at the Newberry Library. 9. Lara Schrijver, “Utopia and/or Spectacle? Rethinking Urban Interventions through the Legacy of Modernism and the Situationist City.” Architectural Theory Review 16.3 (2011): 246. 10. André Breton. “Sur certaines possibilities d’embellissement irrationnel d’une ville.” Le surréalisme au service de la révolution . No. 6. P. 018. Web. 8 Jan. 2014. http://melusine.univ-paris3.fr/Surr_au_service_dela_Rev/ Surr_Service_Rev6.htm 11. Simon Sadler, The Situationist City . (Cambridge, MA: MIT Press. 1999), 82–8. 12. Johanna Drucker. “Humanities Approaches to Graphical Display,” Digital Humanities Quarterly 5.1. (2011). 13. At once aligned with and departing from the practice of Franco Moretti’s distant reading, our maps track trends and currents in modernist geospatial expression that appear through both pattern and instance. Our mapping methods and results are in conversation with geospatial work at the Stanford Literary Lab and the Scholars’ Lab at the University of Virginia. Franco Moretti, Graphs, Maps, Trees: Abstract Models for a Literary History , (New York: Verso, 2007). 14. Lev Manovich 2007. Q&A session at the Digital Humanities 2007 Conference, Urbana-Champaign, IL, June 2007. MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 105

15. Alan Galey and Stan Ruecker, “How a Prototype Argues,” Literary and Linguistic Computing 25.4 (2010): 407. 16. The suggestion for using pixel location came to us from the Humanities Computing and Media Centre (HCMC) and work on the Map of Early Modern London (MoEML) based there; we would like to particularly thank Martin Holmes for his suggestions of best practices for this project’s use of XML and for creating the XSLT we used to transform our data. 17. Richard Zeikowitz notes the importance of vague and ambiguous space when he invokes Michel de Certeau in a discussion of Marya’s creation of space: “Marya writes what de Certeau aptly describes as a ‘long poem of walking’—a poem that ‘manipulates spatial organizations, no matter how panoptic they may be’; her poem of Paris ‘creates shadows and ambiguities within […] [spatial organization’” (qtd. in Richard E. Zeikowitz, “Writing a Feminine Paris in Jean Rhys’s Quartet, ” Journal of Modern Literature 28.2 (2005): 3). 18. As of this writing, the Modernist Versions Project is beta-testing an automated mapping process that uses the Stanford Name Entity Recognizer (NER) (developed by the Stanford Natural Language Processing Group) and python to dynamically generate a warped map that shows the location and frequency of place names mentioned in a given novel. 19. Alan Liu, “Where is Cultural Criticism in the Digital Humanities?” in Debates in the Digital Humanities, ed. Matthew K. Gold (Minneapolis: University of Minnesota Press, 2012), 490–509. 20. Wendy Hui Kyong Chun. Programmed Visions: Software and Memory (Cambridge, MA: MIT Press, 2011): 177. 21. Djuna Barnes. Nightwood (Faber & Faber: London, 1936): 11. 22. Ibid.,11–12. 23. Ibid.,154. 24. In this sense, z-axis mapping understands space as does Amy Wells-Lynn when she describes geocriticism, writing: “[A] geocritical approach is more than just taking an inventory of the locations mentioned in a text; it analyzes how the geography functions as a coded language communicating sexual meaning. ” Amy Wells- Lynn, 78. 25. Lawrence Schehr, French Gay Modernism . (Urbana: University of Illinois Press, 2004): 3. 26. Henning Bech explains the geographic import of potential homosexual activity in When Men Meet: Homosexuality and Modernity . Bech writes: “That fellow members of his species also live there is obviously a factor extremely conductive, though not necessary, to such a concentration; in order for a homosexual neighbourhood to actualize, the individuals will in any case have to leave their dwellings and enter the city of other people’s places. But living close to others also has a signifi cance for one’s awareness 106 A. CHRISTIE AND K. TANIGAWA

of the potentialities of community and opportunity, and in urban living this is often just as important as actualization itself.” Henning Bech. When Men Meet: Homosexuality and Modernity (Chicago, IL: University of Chicago Press, 1997): 116. 27. Amy Wells-Lynn, “The Intertextual, Sexually-Coded Rue Jacob: A Geocritical Approach to Djuna Barnes, Natalie Barney, and Radclyffe Hall,” South Central Review 22.3 (2005): 101. 28. Barnes writes: “As, from the solid archives of usage, she had stolen or appropriated the dignity of speech, so she appropriated the most passionate love that she knew, Nora’s for Robin. She was a ‘squatter’ by instinct.” Djuna Barnes, 103. 29. Djuna Barnes, Nightwood , 105. 30. Ibid.,114. 31. Wells-Lynn, “The Intertextual, Sexually-Coded Rue Jacob: A Geocritical Approach to Djuna Barnes, Natalie Barney, and Radclyffe Hall,” 81. 32. Lawrence Schehr unpacks the complex role of the other across realist and modernist modes of representation in Figures of Alterity . As he argues, realism’s incomplete attempt to assimilate the other within a knowable, universalist framework gives rise to modernist forms of representation through which other occupies epistemological lacunae that central to its very status as other. For Barnes, the concealment of queer activity through spatial markers that register as geographic locations serves as an epistemological tactic central to that activity’s status as queer. These activities function through the very slippage of knowledge/ignorance, identifi cation/ misprision that is key to their status as marginal. As Schehr explains: “Instead of reducing the other to a version of the same, by which it is considered an inferior version of that which shows identity, realist narrative attempts a double movement: an extension of narrative toward the other and an inclusion of the other within a universal. Realist narrative seeks to maintain the particular nature of that which it discovers, describes, or represents while making itself the universal discourse that contains all others.” (13–14) And he continues: “In [other cases], the very space of representation is reformulated because the variable subjects bring their own laws of representation and form: as it moves toward a universal, or at least a sum, realism is its own undoing” (14). Lawrence Schehr, Figures of Alterity: French realism and its others . (Stanford: Stanford University Press, 2003): 13–14. 33. Richard E. Zeikowitz, “Writing a Feminine Paris in Jean Rhys’s Quartet, ” Journal of Modern Literature 28.2 (2005): 1. 34. Ibid., 3. 35. Deborah L. Parsons. Streetwalking the Metropolis: Women, the City, and Modernity (Oxford: Oxford University Press, 2000), 145. MAPPING MODERNISM’S Z-AXIS: A MODEL FOR SPATIAL ANALYSIS... 107

36. Jean Rhys. Quartet: Postures (London: Andre Deutsch, 1969), 16. 37. Thomas F. Staley, “The Emergence of a Form: Style and Consciousness in Jean Rhys’s Quartet ,” in Critical Perspectives on Jean Rhys (Washington, DC: Three Continents Press, 1990), 130. 38. Jean Rhys. Quartet: Postures , 22. 39. Ibid., 95. 40. Ibid., 101. 41. Ibid., 103. 42. Ibid., 103. 43. Ibid., 82. 44. Ibid., 93. 45. Ibid., 124–5. 46. Ibid., 74, 90, 136. 47. Ibid., 67, 186. 48. Massey, Doreen B. Massey. For Space . (Thousand Oaks, CA; London: Sage, 2006), 9. CHAPTER 5

Textbase as Machine: Graphing Feminism and Modernism with OrlandoVision

Kathryn Holland and Jana Smith Elford

The “United Procession of Women,” or “Mud March” of February 1907, was a turning point in the social life of feminism. The fi rst large-scale suffrage demonstration in Britain, it was organized by the moderate National Union of Women’s Suffrage Societies (NUWSS), but as a “joint women’s franchise demonstration,” it convened members of diverse, sometimes confl icting feminist groups—including the more militant Women’s Social and Political Union, some of whom would soon break off to form the Women’s Freedom League. The procession brought approximately 3,000 participants into the city’s streets, creating a new interface between the feminist movement and wider public; the era of large-scale feminist demonstrations was underway. 1 At this event, members of diverse feminist communities converge to advance their work on one of the movement’s central issues: female suffrage. A material source from the “Mud March”—the event’s promo- tional fl yer—documents this moment within decades of agitation, suggesting how the feminist activism of the time was marked by not only

K. Holland () MacEwan University, Edmonton, AB , Canada email: [email protected] J. S. Elford University of Alberta, Edmonton, AB, Canada email: [email protected] © The Editor(s) (if applicable) and The Author(s) 2016 109 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_5 110 K. HOLLAND AND J.S. ELFORD political divisions but also points of contact and outreach. The fl yer lists the leaders of the march: English and Irish speakers of multiple generations and political parties who were members of the NUWSS (Millicent Garrett Fawcett and Jane Strachey), Women’s Liberal Federation (Eva McLaren), Women’s Trades’ Council (Eva Gore Booth), and National British Women’s Temperance Association (Margaret Crosfi eld) among other groups (Fig. 5.1 ). 2 Other primary and scholarly sources indicate that these women’s feminist activities extended past platform suffrage politics to literary and cultural networks. Millicent Garrett Fawcett and Jane Strachey participated in the founding of Newnham College, one of two women’s colleges at Cambridge University; it opened in 1871 and would be transmuted most famously in Virginia Woolf’s A Room of One’s Own (1929). 3 Eva Gore- Booth and Esther Roper were pacifi sts but went to Dublin in 1916, to support rebels of the Easter Rising who included Gore-Booth’s sister Constance Markievicz. That year, Gore-Booth also edited the inaugural issue of Urania , a little magazine dedicated to explorations of identities beyond the gender binary; she created the magazine’s motto, “Sex is an Accident.” 4 W. B. Yeats knew the Gore-Booth family and meditated on the personal lives and political beliefs of the two sisters, in the poem “In Memory of Eva Gore-Booth and Con Markievicz” (1927). 5 Whereas the “Mud March” fl yer documents one moment when feminist groups came together to stimulate new public views of one of their major issues, suffrage, the bodies of writing by and about featured speakers at the march suggest some of feminism’s pervasive reach in modernist networks. We begin with this vignette because the fl yer, a visual representation of the interlinked fi gures in the suffrage movement, gestures towards the dense and multi-layered political, literary and cultural networks that constitute modernism, and towards the importance of feminist institutions in them. Using graphs drawn from a comprehensive feminist literary history, this chapter aims to advance understandings of how the digital may illuminate feminist modernism. Our study identifi es the interplay between abstract representations of authors’ connections and texts—narrative and material—by and about such authors. It demonstrates that graphs offer distinctive insights about the signifi cance of varied sites of feminist activity and the writing created in such sites. While our case studies focus on Newnham College (the women’s college that opened in 1871) and the NUWSS (the national political federation that aggregated local societies from 1897 onward), they also observe the ways cross-period, collaborative digital research reveals parts of modernism’s heterogeneity and overlaps TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 111

Fig. 5.1 Flyer for the United Procession of Women suffrage march, London, 9 February 1907 (Image reproduced courtesy Elizabeth Crawford) 112 K. HOLLAND AND J.S. ELFORD with late-Victorian feminist politics and literature. 6 We argue that readings of material artefacts and narrative texts—fi ctional, non-fi ctional, and scholarly—change via network graphs that illuminate relationships formed in feminist institutions of the period. Visualizations are provocative sources in the practice of modernist literary history. This chapter responds to recent critiques of modernist scholarship. In her introduction to the Modern Fiction Studies special issue on “Women’s Fiction, Modernist Studies, and Feminism,” Anne Fernald observes that “work on women writers abounds but defi nitions of modernist studies consistently neglect or underserve women.” 7 To comprehend the “uneven, surprising, and profound impact of modernity,” she argues, scholars must not only examine the work of diverse individuals and communities of women authors but also situate sustained critical and theoretical studies of gender and the “feminist institutions of modernism” at the center of emerging research across the fi eld (229). Similarly, in a special issue on “The Future of Women in Modernism” for Literature Compass, Barbara Green advocates for the print- and digital-based study of “the network, the dialogue the conversation” in feminist periodicals that were stimulated by contemporaneous media technologies and material contexts. 8 Feminist subjects and methodologies in modernist studies are currently developing in new venues, in response to work by these and other scholars. 9 Our research therefore joins ongoing examinations of the impact of feminism by looking particularly at how digital network graphs together with primary and secondary texts, some widely available and others diffi cult to access, alter understandings of institutions in modernist culture. This methodology applies a feminist perspective to Lawrence Rainey’s earlier argument that diverse venues and groups were part of making modernism “a social reality, a confi guration of agents and practices that converge in the production, marketing, and publicization of an idiom, a shareable language in the family of twentieth-century tongues.” 10

BACKGROUND: ORLANDO: WOMEN’S WRITING IN THE BRITISH ISLES FROM THE BEGINNINGS TO THE PRESENT AND O RLANDOVISION ( OVIZ ) The graphs we generated are drawn from the born-digital, interactive textbase of original scholarly writing, Orlando: Women’s Writing in the British Isles from the Beginnings to the Present (ed. Brown, Clements and Grundy). 11 TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 113

The encoded entries in it are created by teams of scholars across periods and movements, and the textbase contains entries on more than 1,300 canonical and lesser-known female and male authors, British and international. The textbase is thus unlike digital humanities sources that delve exclusively into modernist literature and culture—such as those created by the Modernist Journals Project, Modernist Versions Project, Editing Modernism in Canada and the Modernist Archives Publishing Project. 12 And whereas modernist visualization studies frequently explore primary texts directly, we machine-read graphs from a secondary textbase to test the multiple ways born-digital scholarly writing both pairs with its primary subjects and informs subsequent research. 13 The subjects and periods in the Orlando textbase make it conducive to analyze of points of contact between individuals and associations that bridge the modernist and Victorian periods, along with those bridging the center and peripheries of modernism. The textbase that we draw from in this study is produced by the Orlando Project, an interdisciplinary and collaborative feminist intervention in literary history, cultural studies and digital humanities. The names of the project and textbase spring from modernist literature and suggest their expansive scopes beyond the movement and period: they investigate the writing and lives of women and men from the sixth century BCE onward. They are inspired by Virginia Woolf’s “rich, rollicking, future-embracing” Orlando: A Biography (1928), 14 which embeds a history of women’s writing in the title character’s centuries-long experiments with poetry, with gender and with history. The interpretive markup, which shapes the content of each textbase entry and determines the context of links throughout the textbase, is made visible by the tagging structure that undergirds the textbase, which is accessible to the user. The tagsets, or Document Type Defi nitions (DTDs) created by Orlando team members represent and explore the complex conditions of writers’ lives and bodies of work. As such, they give access to the feminist paradigm that situates women’s writing as central to cultural and historical models of literary formation (Figs. 5.2 and 5.3 ). Tags appear more than 2.5 million times in the textbase, which currently contains more than 8.25 million words. This encoding ensures that each individual author- entry has at least ten separate links to other individuals, either in entries or events. It is a non-linear complex of narratives comprising a literary history built from multiple perspectives and based on topics and themes of the tagsets, rather than notions of the siloed author or discrete historical periods. To develop our analysis we use the project’s visualization tool OrlandoVision (OViz ), which takes the textbase as its sizeable dataset. 114 K. HOLLAND AND J.S. ELFORD

Fig. 5.2 Links screen for Vernon Lee entry in the Orlando Project

Fig. 5.3 Life document type defi nition (subtags not visible)

Regarding the exploration of primary texts, Franco Moretti has asserted that we can analyze only a “minimal fraction of our literary fi eld” if we undertake traditional methods of reading; this is because of the vast quantity of sources that “can only be understood by stitching together separate bits of knowledge about individual cases.” 15 The same can be said of Orlando , which cannot be read by a single user in its entirety because of its size and arrangement. OViz users machine-read the vast amounts of information from the textbase in abstract form, representing the text as node- edge network graphs where people are nodes connected by TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 115

Fig. 5.4 Selection tools on Right , text box at bottom , and visualization pane in top left edges. The tool’s interface encourages distant and close readings. When users click on a particular edge or node, the textpane under each graph excerpted from relevant Orlando entries explains the context of the link, listing which tags encode the narrative information (Fig. 5.4 ). The tool’s limitations include its focus on such genres as the novel and poetry, over periodical publications or non-fi ctional material. It is also limited by its inability to represent anything other than a person as a node. With people only represented as nodes, connections between organizations that do not mention other people are left out of the graph, because a connection is not evident to the tool. Nonetheless, the strengths of OViz are its responsive- ness to user inquiries, ability to access the entire textbase in depth without sacrifi cing breadth, its depiction of literature as deeply connected with society, and its fundamentally feminist re-centering of literary history.

Methodology for Using the “Paratextual Machine” We use the Orlando textbase and its graphing tool to identify how encoded scholarship and visualization together spur new approaches to literary texts and contribute to new understandings of locations of feminist activity in modernism. In this process we also examine the complementary potential 116 K. HOLLAND AND J.S. ELFORD of distant and close reading practices. This process aligns with Jacqueline Wernimont’s argument that the textbase entries and visible DTDs, which are at times speculative rather than exclusively declarative, “generat[e] a feminist and materialist hermeneutic space through which a reading of primary texts is enabled.” With a focus on the tagsets, Wernimont calls it “a paratextual machine that enables users to contextualize and read primary texts.” 16 Bringing the graphing tool into work with the entries' markup and prose, we seek to advance Wernimont’s discussion about the relationships between features of the textbase and its subjects. We combined our understandings of late Victorian and modernist literary history, developed in our multiple scholarly projects, with our awareness of the Orlando dataset. We then used the textbase and visualization tool to generate graphs of signifi cant feminist institutions in the modernist period. The subjects of our case studies, the NUWSS and Newnham College, have been recognized as feminist institutions in historical studies of the early twentieth century, but were not yet explored substantially in studies of modernist literature and culture. 17 The research questions for this study addressed how a cross-period, digital literary history, in which social and political issues feature prominently, could be leveraged to investigate the places of feminist activism and writing in modernism. Combined readings of graphs, narratives and material sources enable us to identify the signifi cance of these feminist institutions in modernist culture. The textpane clarifi ed the nature of each edge and alerted us to the substantial critical information in the textbase prose from which it is drawn, repeatedly turning us back to the full Orlando entry. After reading the graphs, we returned to these primary sources for close reading, applying new questions about their textual features, as well as their production and reception in their original cultural milieu. This approach is similar, for example, to Hoyt Long and Richard So’s combined modes of interpretation in their use of network diagrams of publication data to assess the relationship between literary activity and form in the work of poets contributing to a large, international set of modernist periodicals. As they state in their analysis of bibliographic information, the “ability to generate empirical evidence at such high orders of magnitude [is not] an excuse to reject traditional hermeneutics or cultural critique but [offers] a means to introduce new categories of analysis … for elucidating the network effects of literary affi liation on poetic style and form.” 18 The graphs we generated with OViz are an experiment in such a method. They confi rm that users can wield the abstract and narrative information provided by OViz , TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 117

drawing from Orlando itself, to build new interpretations of primary texts and the cultural institutions of which they were part.

Case Study 1: National Union of Women’s Suffrage Societies Our fi rst case study concerns a major site of convergence between late Victorian and modern feminist writers and activists: the National Union of Women’s Suffrage Societies. It is a feminist institution of modernism. Late Victorian and modernist feminist writers were involved in this organization, either as central, organizing members, on the periphery, or as critics of the organization. Others were politically involved members of the Fabian Society and the Independent Labour Party. We anticipated that our searches of a suffrage-dominated organization would reveal this multiplicity of social and literary perspectives, especially as the NUWSS became increasingly focused on large-scale suffrage demonstrations. Despite our awareness of the irony of decentering contemporary historicized feminism’s focus on suffrage with a search for a suffrage organization, we expected that the specifi city of a search for the NUWSS, examined in the larger context of a revisionist history of women’s writing, could contribute to an alternate understanding of feminism and its infl uence more broadly on the literature of the modernist period. In particular, it shows how a suffrage organization could serve as gathering place for a wide range of people and groups, including ostensibly unconnected writers of modern or proto-modern fi c- tion or social critics from a variety of political perspectives, motivating forms of concerted, collective action or other, diffuse forms of social change. The NUWSS’s distinctive history and structure make it a rich case study for the investigation of feminist exchanges and communities in modernism. It united many disparate local organizations into a cohesive whole, bringing awareness of the women’s suffrage cause to the national stage, and signaling a new phase of feminist activism. The NUWSS was a unique organization in its bi-partisan, education-focused activity, and it united cross-generational women, many of whom rejected the militant tactics of the suffragettes. Despite its initial policies of public education, government pressure and non- violence, members of the organization held a wide variety of perspectives, some of which shifted over time, and several would hold joint membership in both the NUWSS and the more militant Women’s Social and Political Union (WSPU). 19 As the fl yer at the start of this chapter indicates, partici- pation in the NUWSS-organized march of 9 February 1907 was one fl ash- point that drew women of different political stripes and formal affi liations. 118 K. HOLLAND AND J.S. ELFORD

We began our searches by selecting our keywords (“National Union of Women’s Suffrage Societies” and “NUWSS”) and deciding to undertake a “Broad” search, which links all textbase entries that mention the NUWSS and all the people to whom they are connected, in order to reveal the larger indirect connections between the NUWSS and women’s writing culture. We did this to avoid focusing on suffrage as though it were an exclusive representation of feminist during the period, and to demonstrate the degree to which feminism is embedded in modernist literary culture with affi liations and infl uences that would not otherwise be seen. To further indicate our interest in the connections between literary history and feminist activity, we narrowed our graph by selecting only writers who had entries in the published Orlando textbase, fi ltering out individuals with fewer signifi cant links to the constellation of women writers actually involved in feminist activity. Initial searches resulted in a heavily populated graph featuring several signifi cant fi gures dominating a graph with many purple and green links, which signify Orlando textbase tags (Fig. 5.5 ). Purple refers primarily to the women writers’ texts, either their history or the , while green signifi es aspects of their biography, including the , , and tags. Any lighter green or yellow tags, which also populate the graph, though less in number, refer primarily to women’s familial connections , though also to their . The layout of the graph depicts several large starbursts rimming the periphery and extending into the center of the graph. The nodes at the center of these starbursts denote individuals including Millicent Garrett Fawcett, the founder of the NUWSS, and running clockwise from Fawcett, Isabella Ormston Ford, Sarah Grand, Mary Gawthorpe, Katharine Tynan, Eva Gore-Booth, Dora Marsden, Rebecca West, Storm Jameson, Virginia Woolf, Julia Strachey, Mary Agnes Hamilton and Rose Macaulay. Towards the center-left of the graph, near Fawcett, are smaller starbursts that include Beatrice Harraden, Eleanor Rathbone, Maude Royden, Constance Garnett, Emmeline Pethick- Lawrence and the three Pankhursts (Emmeline, Christabel and Sylvia). Many of the fi gures are notable for their signifi cance to scholars of feminism, women’s modern literature and the women’s suffrage campaign, but are not actually members of the NUWSS. By selecting the particular links that connect these women to each other, we can see in the textpane window Orlando ’s description of the specifi c links in the graph, which reveal that while several people in this graph were NUWSS members, several TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 119

Fig. 5.5 Initial network graph of the NUWSS more were not members, and instead engaged with the NUWSS either in opposition to its tactics (such as top members of the WSPU) or are simply present in the graph by virtue of their connections to other members of the group. For instance, while Millicent Fawcett, Isabella Ford, Maude Royden, Eleanor Rathbone and Ray Strachey were key, executive members of the NUWSS during its long tenure, individuals like Dora Marsden, Beatrice Harraden, Cecily Hamilton and the three Pankhursts are present in the graph because, as members of the WSPU, they largely opposed the non-militant tactics of the NUWSS. 20 Writer Mary Gawthorpe is in the graph because of her connections to both the NUWSS and the WSPU: she initially founded the Leeds branch of the NUWSS, but subsequently became a paid organizer of the WSPU; she was reputedly respected by both suffragists and militants alike, and is a signifi cant bridge between both suffrage organizations and beyond them. Other individuals, like Storm Jameson, Julia Strachey and Mary Agnes Hamilton, are only present in the graph because they are connected to fi gures like Dora Marsden, Ray Strachey and Virginia Woolf who themselves have direct connections to the NUWSS. The graph thus offers information about points of contact 120 K. HOLLAND AND J.S. ELFORD across divergent groups and the relationships that developed outside the sustained, formal campaigns of the suffrage movement. The expansive nature of feminism in the modernist period becomes clear as we further explore the NUWSS graph, the Orlando textbase and other sources. One edge in the graph reveals that Storm Jameson was a close friend of the journalist Dora Marsden, editor of The Egoist , and contributed review essays and creative work to the journal from 1914 onwards. 21 Marsden had edited and established The Egoist in January 1914, which emerged in the wake of her previous two journalistic efforts, The Freewoman (est. November 1911) and The New Freewoman (est. June 1913). 22 Marsden and her colleague, Mary Gawthorpe, a former member of the NUWSS, had both been intensely involved with the WSPU at its beginning, but began to distance themselves from the organization believing its vision too narrow and possibly even misguided, given the pervasiveness of corrupt government institutions. Wanting a women’s journal focused more on the full scope of women’s concerns—in Rebecca West’s words, the “profounder aspects of Feminism,” 23 —and not solely on the suffrage struggle, The Freewoman was born when Gawthorpe donated ₤500 to Marsden. As Robert Scholes has noted in his introduction to these journals on The Modernist Journals Project , both The Freewoman and The New Freewoman “encouraged debate [… and] also allowed space for views they opposed,” including a wide range of perspectives on female sexuality, female subjectivity and more. 24 All three journals, though primarily social and political, made room in each issue for poetry, fi ction and reviews, increasing in proportion in The New Freewoman and again in The Egoist . They included a wide range of contributors, ranging from social and sexual critics to feminists and suffragettes, as well as key modernists fi gures. The graph, though ostensibly about the NUWSS, reveals a web of links representing Marsden’s connections to both central and peripheral modernist and feminist fi gures by way of these journals, some describing the writing process, including , or , and others, like the tag, denoting political activity (Fig. 5.6 ). Ezra Pound, for instance, is connected to Marsden in because she employed him as the Poetry Editor of The New Freewoman , while H. G. Wells and Rebecca West link to Marsden via the tag because they were also members of the Freewoman Discussion Circle, a political group affi liated with the journal that discussed feminism, socialism and anarchism, among other things. The appearance of a wide range of feminist and modernist fi gures in a graph of the NUWSS suggests that the feminism of the period confronts and pervades other TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 121

Fig. 5.6 Graph of NUWSS focused on Marsden

modernist institutions and individuals, effecting the establishment of journals and the development of feminism well beyond the core feminist activism of the female suffrage movement. It also reveals signifi cant information about the place of Marsden’s modernist publication within its political and creative environment, and in the feminist movement. When we supplement the connections rendered visible in the graph with the text of Marsden’s fi rst article in The Freewoman , entitled “Bondwomen,” we gain further insights into the specifi cities of Marsden’s wide-ranging feminism at this point in her career, and her particular infl u- ence on the modernist period. In her article “Bondwomen,” Marsden calls for women to declare and thus embrace their individuality, saying, “if she is an individual she is free, and will act like those who are free.” 25 These “Freewomen” as Marsden called them, were in dialogic opposition to “Bondwomen,” who, Marsden contended, “are not individuals […. T]hey round off the personality of some other individual, rather than create or cultivate their own” (1). Marsden called on “Freewomen” to reject their inferior servitude as “Bondwomen,” and instead “take [their] 122 K. HOLLAND AND J.S. ELFORD place among the masters” by acknowledging their freedom, and “setting up their own standards and living up to them, […] putting behind them for ever their role of complacent self-sacrifi ce” as mothers and wives. This, Marsden argued, is the meaning “behind the feminist movement” (1–2). Marsden’s defi nition of feminism and the women’s movement, formed in the context of her experience with the NUWSS and the WSPU, displays the often-overlooked currents and cross-currents that constituted the feminism of the period. Her radical challenge to the narrow focus of the suffrage campaign and the techniques of these suffrage organizations, which often appealed to the populace to grant women the vote by virtue of their apparently “natural” roles as wives, mothers or managers of the domestic sphere, was formed in the context of her disagreements with these suffrage organizations, and came out of her experience with them. Although a large number of suffragists and suffragettes held nuanced opinions on this front despite their campaign rhetoric, Marsden’s strand of feminism refused an appeal to sexual difference to justify women’s enfranchisement, and instead expressed the fundamental need for self-recognition fi rst of women’s own individuality. Her emphasis on the individual was both infl uenced by and in turn infl uenced the modernist movement, which can be seen in this fi rst issue of The Freewoman as well as in Marsden’s subsequent magazines, particularly The Egoist, which became even “more modernistic and exclusive” over time. 26 Marsden’s feminist individuality, which she expressed in this fi rst issue of The Freewoman, thus precedes her subsequent magazine’s eventual underscoring of a particular kind of modernist subjectivity, which had strong ties to the feminist movement. It also alters established views of Marsden’s involvement with these feminist groups, making us rethink her position in the social and cultural milieu of the time and the signifi cance of her Freewoman editorials. The presence of Marsden in a graph of the NUWSS, along with her connections through both literary and political phenomena, speaks to the effect of a feminist institution of modernism like the NUWSS on the art, culture and literature of the modernist period. The feminism of the NUWSS was not stand-alone; instead, it encountered and confronted alternate versions of feminism, which in turn deepened and expanded the feminism of the period. Rather than simply a movement narrowly focused on gaining the franchise and related political rights, this institution of feminism addressed issues of individuality, subjectivity and even sexuality, with deep connections to the art and literature of modernism. Indeed, the feminism of the NUWSS and WSPU spurred a particular kind of modernist subjectivity. TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 123

Case Study 2: Newnham College Our second case study for this essay also combines narrative sources with a graph to develop new interpretations of the feminist dynamics of modernism. We focus on Newnham College, University of Cambridge, whose cultural signifi cance as an educational center is interwoven with the numerous kinds of feminist politics and literature that developed inside and by way of the college. We generated an OViz graph for the college by using “Newnham” as our keyword set to a “Narrow” search to explore the points of contact among fi gures whose entries or mentions in the textbase discuss Newnham, but whose associations are not readily apparent in their writing and the established reception of such writing. The graph contains 153 nodes and 187 edges, one “fl oating” cluster of fi fteen nodes connected by sixteen edges representing particular aspects of Newnham’s early history (Fig. 5.7 ). A second, larger cluster of nodes in the center of the graph contains 105 nodes connected by 140 edges (Fig. 5.8 ). With them we extend our argument that the network graph does not simply rehearse accounts of feminist activities and relationships in modernism that have been examined in scholarly narratives. Instead, it prompts users to change such narratives

Fig. 5.7 Cluster 1 in OViz graph of Newnham College 124 K. HOLLAND AND J.S. ELFORD

Fig. 5.8 Cluster 2 in OViz graph of Newnham College, with view of Woolf, Harrison and Mirrlees nodes by visualizing unfamiliar points of contact that then inform new analyzes of primary literature and the culture to which it belongs. In the fi rst “fl oating” cluster (Fig. 5.8 ), one edge, which links Olive Schreiner and the socialist writer Katharine Bruce Glasier via , marks the former’s infl uence on the latter’s feminism and prompts new approaches to Glasier’s novels. 27 Schreiner visited Newnham as a guest lec- turer when Glasier was a student in the late 1880s. The textpane includes Glasier’s description of Schreiner’s lecture: Glasier writes that Schreiner “encouraged every bit of courageous aspiration or rebellion she found in us.” 28 Glasier’s “rebellion” involved identifying herself by her education, adding BA to the end of her name, though women at Cambridge did not win full degree status for more than fi fty years after she studied there. The single edge supports readings of the people and events involved in signifi cant cross-generational feminist relationships at Newnham and their impact in the public culture around the college. Glasier examined the importance of the college for feminist social con- science at the turn of the century. She wrote two novels with heroines who TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 125 embrace socialism and feminism after studying at fi ctionalized Newnham- like institutions. The novels Aimée Furniss, Scholar (1896) and Margret: A Twentieth-Century Novel (fi rst serialized in the Weekly Time and Echo , 1902–03) represent these institutions as important for theoretical learning, but her feminist characters demonstrate the necessity of practical labor, outside of the institution, alongside theoretical feminist beliefs. Aimée Furniss, Scholar opens with a scene that features the eponymous heroine triumphantly reading the Greek tragedy Agamemnon “with no thought for grammar or dictionary," 29 but when she concludes her reading, she scorns at the thought that “the master-artist […] appeal[ed] through all time to—[…] Brookfi eld” (9), the small industrial town in which she has recently become a High School teacher (11). Although the novel assumes the importance of Aimée’s feminism as a prerequisite for any kind of social change, Aimée Furniss, Scholar suggests that feminism is meaningless if only on the level of pure theory. Instead, it demonstrates that feminist solidarity—and intimacy among women—might be the impetus for practical action and thus the solution for meaningful social transformation. The novel’s climax occurs when Aimée prevents Annie Deardon, a destitute pregnant woman, from becoming a prostitute. The two women make vows of eternal affection—“your sorrow is my sorrow, […] your home is my home, till death do us part” (103)—and exchange rings. They then commit to work together to advance the socialist cause, while together raising Annie’s child. Glasier’s polemical fi ction alludes to the college’s tacit, multilayered feminism, where feminist agency developed in concert with female intimacy and socialist politics, pointing to the expansive social and creative relations promoted at the college. Glasier’s writing demonstrates that the cultural signifi cance of Newnham, as a feminist institution of modernism, can be studied via obscure texts. The network graph and Orlando textbase scholarship are crucial to our identifi cation of Glasier’s role with others in establishing Newnham’s cultural signifi cance because her novels are out of print and comparatively inaccessible. When we return to Glasier’s fi ction and read it in the context of her experiences at Newnham, after locating her with many others in the graph, we confi rm that Newnham was a signifi cant feminist institution because of the way it informed women’s thinking in their writing careers, when they created feminist fi ction to promote wider social and political change. The second, larger cluster in this graph stimulates new perspectives on Newnham as a gathering place for authors who were not activists but 126 K. HOLLAND AND J.S. ELFORD incorporated feminist concerns into their practices and bodies of writing. Three authors—Virginia Woolf, Jane Harrison and Hope Mirrlees—feature as nodes with high degree centrality in the cluster (Fig. 5.8 ). Woolf and Harrison are discussed in much existing scholarship, and because of that coverage their positions in the graph are predictable. In such prose studies, Woolf’s most familiar connections to Newnham comprise her October 1928 lecture at the college that inspired parts of her feminist polemic A Room of One’s Own , and her allusion to Harrison in that text. 30 Harrison was not a “declared feminist,” but her Newnham-based Classics scholarship and work as a public intellectual asserted the prominence of women and female principles in ancient art and culture. 31 Harrison’s life narrative intersects with Newnham’s: she was one of its fi rst resident students in the early 1870s, was one of the fi rst alumni to become a research fellow there in 1900, and left Cambridge permanently in 1922 after a vote against the establishment of women’s degrees and the destruction of the college’s Clough Memorial Gates by rioting male students. 32 The two are connected to the lesser-known writer Hope Mirrlees, a Newnham graduate and intimate friend of Harrison whose poetry was published by Woolf’s Hogarth Press. Her poetry has been recovered in recent modernist studies including the Orlando textbase entry on her work, the fi rst version of which was published in 2006. 33 Mirrlees wrote the cosmopolitan Paris: A Poem after her time studying Russian in Paris with Harrison, and she dedicated it to Harrison. In it, the speaker gives thickly layered impressions of past literary and historical events and fi gures during a jolting, hallucinatory tour of the city from day to night. Those impressions mingle with the speaker’s visions of the present postwar scene. 34 The poem about the nightmare city was published by Woolf’s Hogarth Press in 1920—two years before Hogarth issued T. S. Eliot’s The Waste Land . Paris , which is being recognized increasingly as a vital modernist text, is not set at Newnham and was created outside the college. Yet it is a striking example of literature infl uenced, written and published by women who established relationships with each other at the college. This cluster in the Newnham graph indicates that the college is a major nexus in the authors’ writing, which show Newnham’s infl uence and represent it as a site for considerations of the radical potential of female community. Edges between the authors convey this. Woolf and Harrison are linked via edges for and tags, which attend to Woolf’s use of Harrison as the spectral “J—H—” in the garden of the fi ctional Fernham College in A Room of One’s Own . Mirrlees is linked to TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 127

Woolf by , which captures information about Virginia and Leonard Woolf’s visit with Mirrlees and Harrison in Paris. Mirrlees and Harrison are linked in the graph by seven edges that communicate varied facets of their relationship. The tag marks Harrison’s role as Mirrlees’s teacher and mentor when the latter studied at Newnham between 1910 and 1913. A edge represents Mirrlees’s years of work on her biography of Harrison that she began after Harrison’s 1928 death but left unfi nished, and an edge for represents Mirrlees’s placement of Harrison’s papers and her unfi nished biography at Newnham College. This information regarding Woolf and Harrison not new: the individual edges mark widely-discussed aspects of these authors’ texts and encounters. But their links to Mirrlees, which repeatedly involve Newnham, are less familiar. And when we analyze the information collectively and simultaneously in the larger Newnham graph, new highlights emerge that reinforce our argument about the college as a feminist institution of modernism. The authors’ places in the graph, situated among ones representing 150 other fi gures associated with the college, demonstrate that their individual bodies of writing and lives are integrated in the college’s culture, changing how we orient their work within the period. Their Newnham-based relationships shaped texts that circulated through the modernist network in which the college participated. Newnham must be considered in accurate assessments of forces shaping the texts’ production and reception in the culture of the time. Edges that link the three fi gures’ nodes to others contain unpredictable information about how and when authors’ affi nities developed at Newnham. These parts of the OViz graph also suggest the value of cross-period scholarship to modernist studies because the information they contain resists notions of siloed authors and discrete historical periods. Harrison’s node, for example, is connected to George Eliot’s via the tag. 35 The textpane quotes Harrison’s little-studied memoir, Reminiscences of a Student’s Life (Hogarth Press, 1925), where the author describes her own “senseless […] excitement” while anticipat- ing Eliot’s visit to Newnham because of her admiration for Eliot’s writing. The textpane explains that Harrison found that “[t]he ecstasy was too much” when Eliot complimented Harrison on her William Morris wallpaper. Studies of Harrison often depict her as a transitional fi gure whose work, much of which was written in the early twentieth century, is part of modernist culture. 36 Our use of OViz attends to something different: hitherto-unexamined aspects of Harrison’s reading practices and views on 128 K. HOLLAND AND J.S. ELFORD

Victorian writers and artists, which are illuminated by the graph and excerpts from the Orlando textbase that draw from a primary source (Harrison’s memoir) that is diffi cult to access and not read widely. The edge between nodes representing Harrison and a Victorian whom she revered does not detract from Harrison’s place in modernism. Much like this graph’s representation of Schreiner and Glasier, the edge turns attention to dialogues at Newnham College, it quotes signifi cant but under-explored writing, and it confi rms tenacious points of contact among politically-minded women writers across the Victorian and modernist periods. Like Glasier, Mirrlees, and Harrison, Woolf produced writing related to Newnham College as a feminist institution that facilitated creative female companionship. The interplay between literature and feminist politics in the college is manifest in Woolf’s “A Woman’s College from Outside.” The short story is set at Newnham, which is named directly in the text. Woolf wrote it in 1920, at the time that she published Mirrlees’s Paris and when she explored primarily masculine traditions at Cambridge in her novel Jacob’s Room (1922). She kept an early draft of the story in a Jacob’s Room typescript and titled it “chapter X,” planning to include it in the novel, but the narrative was published in Atalanta’s Garland (1926), a volume sold to support the Edinburgh University Women’s Union. 37 “A Woman’s College from Outside” participates in contemporaneous feminist discourses by representing Newnham as a lively site for the development of women’s agency. Its protagonist is Angela Williams, a young working-class woman who is enraptured by Newnham. The “Outside” of the title refers to Angela, the natural space around the college, and the wider world of which it is part. Angela is drawn to another student, Alice Avery: fascinated by Alice’s knowledge and poise, Angela imagines kiss- ing her. Her excitement concerns not only her sapphic idea of Alice but also her sense, involving anxiety and longing, of Newnham as a gateway to the “new world.” She holds her perceptions “glowing to her breast, a thing not to be touched, thought of, or spoken about, but left to glow there.” 38 Another scene that is rhythmically narrated conveys a sensual energy between fellows and students, and the rich atmosphere of the place itself.

Elderly women slept, who would on waking clasp the ivory rod of offi ce. Now smooth and colourless, reposing deeply, they lay surrounded, lay supported, by the bodies of youth recumbent or grouped at the window; pour- ing forth into the garden this […] laughter of mind and body fl oating away TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 129

rules, hours, discipline: immensely fertilising, yet formless, chaotic, trailing and tufting the rose-bushes with shreds of vapour. (141)

The garden is the center of the college and a space that signifi es women’s vitality. The boundless, jovial laughter of Newnham women, moving through the garden, is as much as a central focus of the narrative as Angela. Along with the protagonist’s consciousness and the intimacy depicted between characters, it resists notions of Newnham as an arid or insular place. The work of a canonical modernist, “A Woman’s College from Outside” is widely available but not widely studied. It thereby differs from the texts connected to the NUWSS and Newnham we discuss earlier in this chapter. But like the writing of other authors in the network graph, the story’s signifi cance lies partly in how it transforms understandings of its author’s career in relation to a feminist institution not outside but inside modernism. Woolf wrote the Newnham story years before her best-known work about modern women’s intimate relationships and their private and public intellectual lives. “A Woman’s College” is a direct precursor of texts that now are, with others, at the core of modernist literary history. It anticipates the characterization of Clarissa Parry, Sally Seton and Elizabeth Dalloway in Mrs. Dalloway (1925); Lily Briscoe’s perspectives in To the Lighthouse (1927); and the depiction of Fernham in A Room of One’s Own . This study of “A Woman’s College” and work by varied writers suggests how Newnham resonated in authors’ and audiences’ imaginations and situates Newnham with other colleges featured in the fi ctional and historical texts of the time. 39 Just as the NUWSS was an institution where writers converged to explored different forms of feminism in society, Newnham College was not limited to the advancement of a single feminist issue. It was a site of women’s higher education, founded and sustained by feminists, which consistently informed explorations of that movement’s intersections with socialism, sapphism and postwar cosmopolitanism.

CONCLUSION We began this chapter with a reading of a material text from 1907. The “Mud March” fl yer provides evidence of how feminist groups advanced one aspect of their cause that spanned the later-nineteenth and early-twentieth centuries, and our use of it changes our awareness of the multiple positions occupied by feminists in modernist culture. Flexible network graphs, 130 K. HOLLAND AND J.S. ELFORD which are digital texts of our own period, move the practice of literary history forward by not only reminding us of what we already know from narrative accounts of modernism but also visualizing unfamiliar information drawn from major datasets that are interpreted fruitfully beside digital and print narratives. This study identifi es a feminism embedded in multiple parts of modernism, in contact with fi gures not always associated with gender politics, which emerged via interpretations of OViz graphs drawn from the Orlando textbase and positioned alongside the primary texts to which the graphs refer. The NUWSS and Newnham College depict much about the multilayered nature of feminism of the period, particularly the clear and the tacit relationships among authors that infl uenced the development of novels, short fi ction, life writing and essays. Distant and close readings of sources about cultural institutions—the college and union of the suffrage societies—reveal connections between literature and politics of feminist modernism. In the essay “Sketch of the Past,” Virginia Woolf muses about how we engage with and represent different kinds of histories. Speculating about the future as well as the past, she imagines a time in which “Instead of remembering here a scene and there a sound, I shall fi t a plug into the wall; and listen in to the past.” 40 There still exists no machine for accessing the unspoken thoughts and memories of historical individuals. Yet digital tools allow for the exploration of scholarly data about women’s writing alongside their particular fi ctional and non-fi ctional writing in other media. Network graphs specifi cally make that data legible and open to interpretation in unique, powerful ways. In this sense, we may be closer than Woolf imagined to fi tting “a plug into the wall; and listen[ing] in to the past.”

NOTES 1. The Women’s Social and Political Union, which established the Women’s Press that same month to market publications and branded goods featuring its signature colors, went on to hold a Hyde Park rally in June 1908 that drew more than 300,000 people. 2. Elizabeth Crawford, ed., facsimile of fl yer for Joint Women’s Franchise Demonstration (London, 1907), Woman and her Sphere , November 21, 2011, web, 5 May 2015. 3. Susan Brown, Patricia Clements and Isobel Grundy, eds., Millicent Garrett Fawcett entry: Life screen within Orlando: Women’s Writing in the British TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 131

Isles from the Beginnings to the Present (Cambridge: Cambridge University Press Online, 2006–15), web, October 2, 2015; and Virginia Woolf, A Room of One’s Own (London: Hogarth Press, 1929). 4. Susan Brown, Patricia Clements and Isobel Grundy, eds., Eva Gore-Booth entry: Life screen within Orlando: Women’s Writing in the British Isles from the Beginnings to the Present (Cambridge: Cambridge University Press Online, 2006–15) web, October 2, 2015. 5. W. B. Yeats, “In Memory of Eva Gore-Booth and Con Markievicz,” The Winding Stair and Other Poems (London: Macmillan, 1933): 5. 6. This area of study is advanced digitally and in print by the transitional focus of Yellow Nineties Online ( www.1890s.ca ) and Elizabeth Carolyn Miller’s characterization of “long modernism” in her book Slow Print: Literary Radicalism and Late Victorian Culture (Stanford University Press, 2013). 7. Anne Fernald, “Women’s Fiction, Modernist Studies, and Feminism,” Modern Fiction Studies (Summer 2013): 230. 8. Barbara Green, “Recovering Feminist Criticism: Repetition, Women Writers, and Feminist Periodical Studies,” Literature Compass 10.1 (January 2013): 58–9, web, May 5, 2015. 9. These include the journal Feminist Modernist Studies , co-edited by Cassandra Laity and Anne Fernald, which will be launched by Routledge in January 2017; and “Theory, Method, Aesthetics: A Forum on Feminism and Modernist Studies,” the prospective online cluster for Modernism/ modernity ’s Print-Plus Platform formulated by Urmila Seshagiri. 10. Lawrence Rainey, Institutions of Modernism: Literary Elites and Public Culture (New Haven and London: Yale University Press, 1998): 4. 11. We have experience with the textbase as contributors and users. Kathryn Holland is the Orlando Project’s Senior Research Associate and draws from entries in the textbase in her argument about feminist exchanges in “Late Victorian and Modern Feminist Intertexts: The Strachey Women’s Roles in A Room of One’s Own and Three Guineas ,” Tulsa Studies in Women’s Literature 32.1 (Spring 2013): 75–98. Jana Smith Elford, Research Associate with Orlando, has written on “Network Analysis and the Historical Recovery of Women: A Case Study of the Fabian News ,” forthcoming in the Journal of Modern Periodical Studies 6.1 (2015). 12. Modernist Journals Project, modjourn.org /; Modernist Versions Project, web.uvic.ca/~mvp1922: Editing Modernism in Canada, editingmodernism.ca/; Modernist Archives Publishing Project, modernistarchives.com /. 13. See articles in the Journal of Modern Periodical Studies issue on “Visualizing Periodical Networks,” 5.1 (2014), ed. J. Stephen Murphy, and the online Twentieth-Century Literary Letters Project ( modmaps.net/tcllp /), which is part of the Mapping Modernisms collection ( modmaps.net /). 132 K. HOLLAND AND J.S. ELFORD

14. Susan Brown, Patricia Clements, and Isobel Grundy, “‘The Most Unaccountable of Machinery: The Orlando Project Produces a Textbase of One’s Own,” Interdisciplinary/Multidisciplinary Woolf: Selected Papers from the 22nd Annual International Conference on Virginia Woolf , ed. Ann Martin and Kathryn Holland (Clemson, SC: Clemson University Press, 2013): 208. 15. Franco Moretti, Graphs, Maps, Trees: Abstract Models for a Literary History. (London: Verso, 2005): 3–4. 16. Jacqueline Wernimont, “Whence Feminism?: Assessing Feminist Interventions in Digital Literary Archives,” DHQ: Digital Humanities Quarterly 7.1 (2013): n. pag., Web, December 7, 2014. 17. Newnham College’s place in Virginia Woolf’s writing has been well-documented in such studies as Naomi Black’s Virginia Woolf as Feminist (Ithaca: Cornell University Press, 2004). For an exploration of the NUWSS’s cultural signifi cance, see Maria DiCenzo, Leila Ryan and Lucy Delap, Feminist Media History: Suffrage, Periodicals, and the Public Sphere (London: Palgrave Macmillan, 2011). 18. Hoyt Long and Richard So, “Network Analysis and the Sociology of Modernism,” boundary 2 40.2 (2013), 148–9. Long and So also discuss their methodology, which includes “a two-layer process that relies on the derivation of specifi c network metrics and traditional close reading of texts,” in “Network Science and Literary History,” Leonardo 46.3 (2013): 274. 19. Sandra Stanley Holton, “National Union of Women’s Suffrage Societies,” Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004–15): n. pag., web, May 3, 2015. 20. Dora Marsden later abandoned the militancy of the WSPU. 21. Susan Brown, Patricia Clements and Isobel Grundy, eds., Storm Jameson entry: Writing screen within Orlando: Women’s Writing in the British Isles from the Beginnings to the Present (Cambridge: Cambridge University Press Online, 2006–15) web, May 2, 2015. 22. Susan Brown, Patricia Clements and Isobel Grundy, eds., Dora Marsden entry: Writing screen within Orlando: Women’s Writing in the British Isles from the Beginnings to the Present (Cambridge: Cambridge University Press Online, 2006–15) web, May 2, 2015. 23. Rebecca West, “The Freewoman,” Time and Tide, July 16, 1926, reprinted in Dale Spender, ed., Time and Tide Wait for No Man (London: Pandora Press, 1984): 63–8. 24. Scholes, “General Introduction to the Marsden Magazines.” Modernist Journals Project . Brown and Tulsa Universities, n. pag., web. November 20, 2015. TEXTBASE AS MACHINE: GRAPHING FEMINISM AND MODERNISM... 133

25. “Bondwomen.” The Freewoman: A Weekly Feminist Review 1.1 (November 23, 1911): 1–2. 26. J. Stephen Murphy and Mark Gaipa, “‘You Might Also Like …’: Magazine Networks and Modernist Tastemaking in the Dora Masden Magazines,” The Journal of Modern Periodical Studies 5.1 (2014): 27–68. 27. Glasier published her fi rst several works under her birth name, Katharine St. John Conway, even though she was by that point married. She only later adopted her husband’s surname, becoming Katharine Bruce Glasier. 28. Qtd. in Susan Brown, Patricia Clements and Isobel Grundy, eds., Katharine Bruce Glasier entry: Life screen within Orlando: Women’s Writing in the British Isles from the Beginnings to the Present (Cambridge: Cambridge University Press Online, 2006–15) web, May 2, 2015. 29. Katharine St. John Conway (Katharine Bruce Glasier), Aimée Furniss, Scholar (London: Clarion Press, 1896): 9. 30. See, for example, Anne Fernald, Virginia Woolf: Feminism and the Reader (Basingstoke: Palgrave Macmillan, 2006). 31. Jean Mills, Virginia Woolf, Jane Ellen Harrison, and the Spirit of Classical Modernism (Columbus and New York: The Ohio State University Press, 2014): 6, 32–3. 32. Susan Brown, Patricia Clements and Isobel Grundy, eds., Jane Ellen Harrison entry: Life screen within Orlando: Women’s Writing in the British Isles from the Beginnings to the Present (Cambridge: Cambridge University Press Online, 2006–15) web, October 2, 2015. 33. See Susan Brown, Patricia Clements and Isobel Grundy, eds., Hope Mirrlees entry: Writing screen within Orlando: Women’s Writing in the British Isles from the Beginnings to the Present (Cambridge: Cambridge University Press Online, 2006–15) web, October 2, 2015; Julia Briggs, “Hope Mirrlees and Continental Modernism” and Briggs’s notes to her edition of the poem, Gender in Modernism: New Geographies, Complex Intersections , ed. Bonnie Kime Scott (Bloomington: University of Indiana Press, 2007): 261–306; and Sandeep Parmar, ed., Hope Mirrlees: Collected Poems , by Hope Mirrlees (London: Carcanet Press, 2011). 34. Briggs identifi es the many allusions in Mirrlees’s poem in her footnotes for her edition of the text in Gender in Modernism: New Geographies, Complex Intersections . 35. The tag choice is an example of what Wernimont identifi es as the speculative, generative nature of the Orlando DTDs: Harrison’s discussion of her interaction with George Eliot, with Morris wallpaper in the background, could also be encoded with the tag. 36. See Mills, for example. 134 K. HOLLAND AND J.S. ELFORD

37. Susan Dick discusses the story’s 1920 composition and 1926 publication in her endnotes for The Complete Shorter Fiction of Virginia Woolf , by Virginia Woolf, ed. Susan Dick (London: Hogarth Press, 1985): 294–5. 38. Virginia Woolf, “A Woman’s College from Outside,” The Complete Shorter Fiction of Virginia Woolf , ed. Susan Dick (London: Hogarth Press, 1985): 141. 39. These include University College, Dublin in James Joyce’s A Portrait of the Artist as a Young Man (1916), Merton College and Trinity College, Oxford in F. Scott Fitzgerald’s The Great Gatsby (1925), and Girton College, Cambridge in Rosamond Lehmann’s Dusty Answer (1927). 40. Virginia Woolf, “Sketch of the Past,” Moments of Being: Unpublished Autobiographical Writings , ed. Jeanne Schulkind (London: Pimlico, 2002): 8. CHAPTER 6

Remediation and the Development of Modernist Forms in The Western Home Monthly

Hannah McGregor and Nicholas van Orden

In May of 1901, The Western Home Monthly printed a short anecdote, sandwiched between a note to their readers and an advertisement for Lumière Dry Plates, entitled “That Nothing be Wasted.” “There seems to be no limit,” it begins, “to the fun to be had with a phonograph. One man writes that he has a perfect record of the barking of his dog, and the dog enjoys hearing the record play as much as his master does.” 1 It continues:

This reminds me of the story now going the rounds in the newspapers— perhaps you’ve heard it—concerning the economics of the pork packing

Special thanks fi rst to our collaborators, Harvey Quamen and Matt Bouchard, to the excellent people at University of Alberta Libraries Digital Initiatives, and to EMiC UA. Thanks as well to our editors for their thorough and thoughtful feedback. This paper was made possible through the generous support of the Social Sciences and Humanities Research Council of Canada. All images from The Western Home Monthly appear courtesy of Peel’s Prairie Provinces (peel. library.ualberta.ca), a digital initiative of the University of Alberta Libraries.

H. McGregor ( ) • N. van Orden University of Alberta, Edmonton, AB, Canada email: [email protected] ; [email protected]

industry in Chicago. Every part and parcel of Mr. Pig is made use of with the greatest of care and ingenuity.… Even the dying squeals, heretofore wasted on the midnight air, are now preserved on phonograph cylinders. It may be only a story, but it shows the modern tendency of manufacturers to utilize every by-product. (8)

With its self-conscious references to media innovation and the circulation of information, and its simultaneous suspicion and embracing of the modern, this anecdote powerfully evokes the magazine’s ongoing relation to modernity, one that registers at the levels of cultural critique, self-conscious technological sophistication and a surprisingly timely remediation of the new media landscape of the moment. This uneasy combination of the modern and the anti-modern, the conservative and the novel, is characteristic of the category into which The Western Home Monthly most obviously fi ts—the middlebrow. As John Guillory defi nes it, “[m]iddlebrow culture is the ambivalent mediation of high culture within the fi eld of the mass cultural.” 2 Mass cultural is precisely what this magazine wanted to be—a note to the readers in May 1901 claimed, “It is our ambition to make The Western Home Monthly the representative publication of the great middle classes” 3 —and arguably what it became by the 1930s. The Western Home Monthly was published by the Stovel Printing Company of Winnipeg, Manitoba, in Western Canada, between 1899 and 1932. Stovel Printing was founded in 1889 by the Stovel brothers, who had moved from Mount Forest, Ontario to Winnipeg four years prior. It quickly became the largest English-language printer and publisher in the city, and while it never achieved the circulation of New York or London-based magazines, it certainly made its mark. By September 26, 1932, Time magazine could describe it as the most widely-circulating household magazine in Canada, the top of the “Big Five” and the only one not published in Toronto. 4 In addition to standing at the forefront of print culture in Western Canada, the magazine also published work by an international collection of authors such as Victor Rousseau, Laura Goodman Salverson and Sinclair Lewis, whose writing introduced many of the aesthetic innovations of realism and modernism to a mainstream readership. With a comparatively modest circulation of 180,000, decidedly non-cosmopolitan origins, and an ongoing interest in the agricultural, it might seem a far stretch to discuss The Western Home Monthly as a modern magazine, let alone a modernist magazine. And yet, REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 137 modernism developed on the pages of this magazine as much as on the pages of any other. It is a basic tenet of modern periodical studies that “little magazines”— self-consciously avant-garde periodicals of small circulation based out of major metropolises of modernist production 5 —were a primary venue for the development of modernist aesthetics and communities of thinkers and writers. In fact, Robert Scholes and Clifford Wulfman, in the title of the fi rst chapter of Modernism in the Magazines , refer to Ezra Pound as the “Founder of Periodical Studies.” 6 And yet periodical studies as a fi eld predates the rise of modern periodical studies, much as the periodical press predates the little magazine. As scholars interested in media history have begun to challenge the strict divisions between Victorian and modern print culture, so too have they explored the blurred borders between modernism and the middlebrow. Patrick Collier, for example, points out the aesthetic continuities between “the image-collages of the Illustrated London News and the shaped and gathered fragments of The Waste Land ,” refusing the narrative of radical rupture between the Victorian middlebrow and the modernist avant garde. 7 Similarly, Kirsten MacLeod’s work looks to the 1890s—a decade commonly associated with “the birth of the mass-market magazine”—for “the contemporaneous effl orescence of a more experimental and amateurish form of print” that looks an awful lot like the little magazines of the twentieth century. 8 Middlebrow periodicals like The Western Home Monthly , bridging the gaps between “the sentimental and the political, realism and experimentation, and modernity and the Victorian,” 9 are an ideal site for exploring how modernism and modernity were being remediated in mainstream and non-cosmopolitan print. Although patterns of mediation and remediation can be remarkably hard to track, especially across the vast archives of long-term serial publication, emergent methods of digitization and distant reading open up periodical archives to new kinds of analysis, of which we make ample use in this chapter. This chapter will proceed in four parts. First, we will articulate our argument for putting The Western Home Monthly in conversation with modernism by exploring the links that have been drawn recently between modernism, the middlebrow and new media studies. Second, we will outline the method through which The Western Home Monthly was digitized and the tools we used in our analysis. The third section will demonstrate how our distant-reading methods helped us to better understand the formal dimensions of the magazine, particularly in terms of the intersec- 138 H. MCGREGOR AND N. ORDEN tions between new media, advertising and the fragment across the hypermediated and remediated pages of our digitized archive. In our fourth section, we will analyze a single issue of The Western Home Monthly , showing how a combination of distant and close readings helps us to understand the place of an agrarian middlebrow magazine within the transnational and intermedial phenomenon of modernist culture. In the process, we will demonstrate how the digital remediation of The Western Home Monthly has brought us closer to an understanding of the magazine’s own practice of remediating a modern media ecology that, by necessity, includes emerging modernist forms.

“MODERNISM IS A MUCH-ABUSED WORD” Even a glancing survey of recent scholarship in modernist studies reveals that the term is still much debated, and any attempt to summarize these debates exceeds the purview of this chapter. We instead turn to Douglas Mao and Rebecca L. Walkowitz’s account of “The New Modernist Studies,” 10 a fi eld that they defi ne in terms of the expansion of the traditional temporal, geographical and methodological boundaries of modernism. 11 Attention to spaces beyond the familiar cosmopolitan nodes of modernist production like New York and London (737), to the fl uid boundaries “between high art and popular forms of culture” (738), and to the central role that mass media played in the circulation of modernist thought (742) has served to push back against a narrow critical lens that would associate modernism with Blast but not Cosmopolitan , with Preview but not The Western Home Monthly . Key to these expanding defi nitions of modernism has been a body of criticism exploring modernism’s deep engagement with mass media. From Collier’s link between “the imagist poetics of the fragment … [and] the visually and textually fragmented modern newspaper page” 12 to Sean Latham’s interest in the magazine as part of the new media landscape of the twentieth century, akin to radio and cinema, 13 periodical scholars have clearly linked modernism and new media. If we can think of magazines as new media, Latham continues, then it is valuable to read them using the vocabulary of new media, particularly in terms of their affordance and emergence , by which he means the “action possibilities” of “complex systems capable of producing meaning through the unplanned and even unexpected interaction of their components” (1, 3). Latham’s argument, to which we will return in more detail below, is essentially that maga- REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 139 zines differ from other print media in their non-linear and intermedial layout; they contain an almost hypertextual quality “that invites the reader to construct connections” within individual issues and across different iterations of the same magazine (2). Rather than treating the magazine as a series of discrete textual objects that can be mined for content—an approach that has dominated in the past, though it has been largely repu- diated by media studies—Latham treats it as a system composed of diverse and overlapping units, a “chaotic mixture of advertisements, images, texts, headers, captions, maps, and photographs” that readers can navigate in diverse and unpredictable ways (3). While Latham’s focus here is on the modernist magazine, these features are arguably even more evident in the middlebrow magazine. As Sullivan and Blanch argue, the periodical can be seen as “the exemplary middlebrow artifact—the ‘composite text’— demanding and constructing a range of reading practices in response to its discursive collage and complexity.” 14 The Western Home Monthly is just such an exemplary middlebrow print artefact, one that mediates the newness of modernity through a conservative, even Victorian, frame. This balance of the new and the old, the experimental and the traditional, is key to one of the magazine medium’s defi ning features—its seriality. As James Mussell has argued:

No single issue exists in isolation but instead is haunted by the larger serial of which it is a part. This larger serial structure is invoked through the repetition of certain formal features, issue after issue. It insists on formal continuity, repeated from the past and projected onwards into the future, providing a mediating framework whose purpose is to reconcile difference by presenting new content in a form already known to readers. This new content, whether the next instalment of story, a one-off essay on a new subject, or a piece of news, is always tempered, regulated within a formal framework that readers have seen before. 15

This mediating framework, as Mussell describes it, can be structural (recurring features or columns), formal (typeface and page layout), and material (paper quality and print technologies). The identity of a particular periodical is thus abstracted in that it exists outside of any individual iteration, and decidedly material in that it is instantiated within the magazine issues as physical objects (348). If the new medium of the magazine is one of the ways in which modernism was making itself new, that newness could not only be a radical break with the past, but must also rely on the familiarity and repetition that might be associated with more conservative forms. 140 H. MCGREGOR AND N. ORDEN

It is possible to explore the shifting, complex relationship of the old and the new, the conservative and the innovative, at the level of form and content across The Western Home Monthly ’s thirty-three-year run. The early issues, with their ornate mastheads, anonymous editorial voice, and straightforward four-column format, look distinctly Victorian, a sensibility still in evidence as late as 1919. This seeming conservatism is belied by the magazine’s deliberate material and rhetorical incorporation of new media. This remediation of the early twentieth century’s new media landscape within The Western Home Monthly is central to our argument. 16 Reading the magazine not as a collection of different texts but as a coherent media object allows us to trace a genealogy of emergent modernist forms across a digitized archive of 24,170 pages and 33,099,536 words—an archive that provides a singular opportunity and a host of methodological challenges. While at times the magazine’s interest in modernism is linguistically explicit, such as the June 1925 editorial feature on “Modernism” by the Rev. John Mackay, Principal of Manitoba College in Winnipeg 17 —from which the title of this section is taken—or the March 1932 article on Canadian painters working in Montparnasse, 18 we are not interested in documenting individual occurrences of words and phrases already generally associated with modernism. Instead, we begin from the argument that modernism happened on the pages of magazines as the new medium negotiated the nineteenth-century affordances of the periodical and its relation to the new media landscape of the modern era. To emphasize the differences between the cultural institution that is the print magazine titled The Western Home Monthly and the digitized version of this archive, we refer to the magazine as The Western Home Monthly and to the digitized version as the WHM .

WORKING WITH THE DIGITAL WHM A common concern about the study of print culture in digital form is that the process of digital remediation fundamentally shifts the nature of the object of study. The digitized magazine page is not, and cannot be treated as, a proxy or replacement for the material page; what we can say about the digitized page, using digital tools, is not equivalent to what we could say about the material page. We by no means deny this argument. Instead, like Mussell, we’re interested in how digitizing periodicals leads to an opportunity to rethink print differently. 19 “What appears to be a defi cit, a misrepresentation, in digital resources," Mussell concludes, REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 141 is actually difference, introduced through transformation” (17–18). For researchers, the process of working through this transformation can itself be benefi cial. Elsewhere, McGregor has argued that digitization affords a unique perspective on periodicals, addressing one of the key challenges in periodical studies, that of dealing with large, heterogeneous archives. A digitized periodical can be more easily analyzed for the patterns of repetition and difference that characterize the serial text. 20 The Western Home Monthly ’s digitization was made possible through a partnership between Editing Modernism in Canada at the University of Alberta (EMiC UA), the research group with which we are affi liated; the Manitoba Legislative Library, where the magazine’s archive is housed; the Digital Initiatives offi ce at the University of Alberta Libraries; and Peel’s Prairie Provinces, a U of A Libraries digital collection of pre-1953 materials from Canada’s West. The various contributors to this digitization are indicative of the collaborative nature of remediating processes—processes that generally deal with “source” and “output” texts across different media and that therefore require input from scholars, artists and techni- cians with expertise in those disparate media. The resulting resource is a full-color digital edition of all extant issues of the magazine; it will be the fi rst household magazine hosted on the Peel’s Prairie Provinces website. The digital WHM contains 348 issues, including: two issues from each of 1901 and 1903, ten issues from 1904, and every issue between 1905 and 1932 except for January 1916, September 1919, and March 1922. A special illustrated issue titled The 1914 War supplements the twelve issues from 1915. The WHM is a rich and expansive resource that was largely shaped by the limitations and capacities of digital remediation. Though now common practice for libraries and literary scholars, the digitization of archival materials can be approached in many ways—from photos hastily snapped under the watchful gaze of special collections librarians to the careful scanning of millions of manuscript pages. Explaining how The Western Home Monthly was remediated into the WHM helps to reveal the range of possible questions we were able to consider in our initial exploration of this valuable resource. Even something as seemingly banal as the number of individual fi les contained in the archive has an enormous impact on the feasibility of storing, sharing and processing the material. Peel’s Prairie Provinces outsourced digitization of The Western Home Monthly to Backstage Library Works of Provo, Utah. Backstage provides “preservation-quality digital images,” as well as optical character recognition 142 H. MCGREGOR AND N. ORDEN

(OCR) and archival-standard metadata markup. 21 The digital WHM that Backstage produced includes 348 PDF and METS fi les (one for each issue) and 24,170 jp2 and ALTO fi les (one of each for every scanned page). In total, the digital WHM consists of 66GB of data across more than 50,000 fi les. METS/ALTO, the markup schema used to encode the magazine, is an archival standard originally developed for the digitization of newspapers. The Library of Congress’s Metadata Encoding and Transmission Standard (METS) is an XML-based schema used to describe the hierarchical structure of digital objects, to preserve the names and locations of the various fi les that comprise an object, and to maintain digitization metadata pertinent to that object. The METS schema is roughly analogous to the bindings of a print publication—METS fi les ensure the structural integrity of a digital object. In our work, the METS fi les allowed us to reconstruct approxima- tions of individual issues of the magazine from among the thousands of scanned pages. Whereas METS data determines the structure of the digital object, the Analyzed Layout and Text Object XML schema (ALTO) contains metadata relating to the stylistic elements of the original print material, such as font descriptions, location data (indicating the position of each word or image on the page), and the original textual content. ALTO fi les are roughly analogous to detailed maps of each digitized page within a digital object. ALTO fi les provided us with most of the data presented here. The ALTO schema is also maintained by The Library of Congress. Used together, METS/ALTO metadata is a powerful tool for the arrangement, maintenance and preservation of digitized objects. Our work would have been impossible without the METS/ALTO metadata Backstage compiled. The digital WHM METS and ALTO fi les contain enormous quantities of detailed information. For example, in Bonnycastle Dale’s April 1916 article titled “Some of Our Adventures in Pursuit of the Finny Ones,” the three words “an old bird” produce the following ALTO XML data:

REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 143

In addition to headers that note crucial namespace, measurement, and OCR information, the ALTO fi les for each scanned page identify individual text blocks (such as columns, article titles and image captions), individual text lines within each text block, and individual strings of content (words) within each text line. Each block, line and string is distinguished by a unique identifi cation number. The horizontal position (“HPOS”), vertical position (“VPOS”), width and height of every element are also noted. Information about the OCR reliability of every character within each word is included within each string tag. The spaces between words are represented by unique “SP” identifi ers, with the VPOS, HPOS and width of each space faithfully recorded. In addition to text blocks, the ALTO fi les label each illustration and advertisement using “ComposedBlock” identifi ers. The size and position of each illustration is indicated. Every ComposedBlock “TYPE=‘Advertisement,’” like the standard text blocks, contains information about all of the lines, strings and pieces of textual content within each ad. One of the key limitations of the WHM is that images within advertisements are not identifi ed in this ALTO XML schema. In total, information about more than 56,000 advertisements, 33,000 illustrations, 24,000 pages and 33,000,000 words is contained in more than 314,000,000 lines of XML code. Our fi rst challenge was to extract all of the textual content from the ALTO fi les so that we could run topic modeling and text-mining programs on the content of the entire collection. A simple PHP script and some command line wizardry, courtesy EMiC UA collaborator Matt Bouchard, produced 24,170 plain text fi les comprised of the ALTO fi les’ “Content” attribute. Further work on the command line, and with Perl (Matt Bouchard again; see modmag.ca/publications/readingmodernism/), concatenated these plain text fi les into appropriately named issues and digests (digests being all of the issues from a calendar year). We next combined all of the ALTO fi les into issues and digests to facilitate broad analyses of the metadata (for example, to calculate the number of advertisements or illustrations in a given year). The possibilities for working with these fi les are seemingly endless. In addition to the work discussed below, location metadata (VPOS and HPOS, which are measured to the tenth of a millimeter) could be used to track the shifting position of advertisements across the issues and digests; changing ratios of text, image and white space might be used to construct arguments about design aesthetics; and information about typefaces and font sizes could lead to visualizations of the mediating frameworks that Mussell argues are central to the serial identity of periodicals. As the remain- 144 H. MCGREGOR AND N. ORDEN der of this paper will demonstrate, even a comparatively simple use of the affordances of METS/ALTO can reveal valuable new information. Our goals in this fi rst examination of the digital WHM include an analysis of the relationships between emerging media technologies and the development of modernist formal characteristics such as textual fragmentation, collage and narrative disruption. We considered questions such as: How did the relationship between modernism and media develop on the pages of The Western Home Monthly ? How can digital reading methods help us discover, understand, and represent this relationship? What do parallels between the remediation of the magazine (into a digital archive) and remediation within the magazine suggest about the interplay between media technologies, representations of visual and textual data and digital humanities-infl ected analyses of literary and cultural artefacts? To focus this project and to leverage the most basic affordances of the METS/ ALTO fi les, we used text mining data and topic models to juxtapose the appearance of new media in the magazine as a whole to the appearance of those media in advertisements. We also extracted simple statistical data from the ALTO fi les in order to visualize trends and shifts in the magazine’s publication, layout, and general form. Figure 6.1 , for example, produced using Microsoft’s Excel’s graphing function, compares the total number of pages, advertisements and illustrations across all 348 issues of the digital WHM . Many discussions about the

Fig. 6.1 Advertisements, illustrations and pages in each year of the Western Home Monthly REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 145 development of new printing technologies, local events in Winnipeg and world-historical trends emerged from these simple graphical representations. For example, is the precipitous decline in advertising between 1929 and 1930 a refl ection of the stock market crash in October 1929? Was the sudden drop in illustrations around 1916 caused by the fi re that ravaged the Stovel Printing Company’s main plant? Is the steady rise in page numbers during the fi rst years of the twentieth century the result of increased circulation, changes in printing technologies or easier access to materials (both raw and intellectual)? Many similar questions will hopefully be explored by other scholars with other interests and expertise. Our decisions to use MALLET for topic modeling and RapidMiner for text mining were infl uenced by a range of factors, including our experience using the tools, the size of the data and the research questions that motivated our analysis. Based on a set of user-defi ned parameters, MALLET (MAchine Learning for LanguagE Toolkit) analyzes massive quantities of data and produces “topics”—lists of words that are statistically likely to occur in proximity to each other within a document or across a set of documents. For example, we would expect the words “western,” “home,” and “monthly” to feature prominently in MALLET’s reading of the digital WHM —which they do. MALLET uses the well-known LDA (Latent Dirichlet Allocation) algorithm, a process which has recently come under some scrutiny, most notably in Andrea Lancichinetti et al.’s “High-Reproducibility and High-Accuracy Method for Automated Topic Classifi cation.” 22 There has been little response to this critique, and we are confi dent that our careful use of MALLET avoided many of the most obvious problems raised in Lancichinetti’s paper. Having experimented with various parameters, as is often necessary when topic modeling, we found that the most revealing topics MALLET produced from the WHM were comprised of twenty words when MALLET was trained to fi nd one hundred topics. Using fewer topics produced vague results and using more than one hundred created unnecessary levels of granularity in the data. Topics longer than twenty words began to include statistically insignifi cant results, grouping together words only distantly related to each other. Topics produced by MALLET reveal within minutes patterns and themes across the content of the WHM that would have taken months to discern through traditional reading practices (if we had been able to discover them at all). As Robert K. Nelson explains in his excellent “Mining the Dispatch ” project, “Topic modeling…allows us to step back from individual documents and look at larger patterns among all the documents, to 146 H. MCGREGOR AND N. ORDEN practice not close but distant reading , to borrow Franco Moretti’s memorable phrase” (emphasis in original). 23 For example, MALLET’s topics reveal the patterns of repetition and difference that characterize the serial identity of The Western Home Monthly . The four most statistically relevant topics, which maintain their dominance throughout the 348 issues, focus on the magazine’s identity as well as overarching concerns with time (year/s, time, day), gender (man, woman, girl), labor (make/made), nation and home. Across the magazine’s thirty-two year run, topics relating to new media emerge and, in some cases, disappear—phonographs, radio and then fi lm feature prominently at different times and cluster within a vocabulary of advertising that includes the brand names Kotex, Palmolive, Edison, Heintzman, Crisco and Quaker. MALLET’s topics reveal the interplay between the traditional and the modern, between the familiar structures of the magazine and the introduction of new media through the discourse of advertising. Visit modmag.ca/publications/ readingmodernism for full-color visualizations of the topic modeling data discussed here. The juxtaposition of general trends and specifi c words produced by MALLET is echoed in the text-mining data produced by RapidMiner. RapidMiner is a powerful analytics, text-, and data-mining platform used extensively across a wide range of industries. Our use of RapidMiner, though suffi cient for our research questions, barely scratched the surface of its text-mining capabilities; version 5.3 (freely available through Sourceforge) was more than adequate for our needs. Using the “Process Documents From Files” operator, we broke each of the WHM digest fi les into individual “tokens” (words), transformed the tokens into lower-case letters, removed common stop-words and tokens shorter than three letters or longer than fi ve hundred, and applied the built-in Porter stemming algorithm (which works iteratively through the material to reduce words to their root forms—i.e. “working,” and “worker” to the root “work”). We copied the RapidMiner output into an enormous Excel spreadsheet for ease of access and manipulation. This same process was repeated on the digest fi les containing only text from the WHM ’s advertisements. Powerful though spreadsheets are, scanning 2.4 million cells of data to discover patterns and trends is unmanageable. Another EMiC UA collaborator, Dr. Harvey Quamen, wrote a wonderful PHP script to extract data from our spreadsheets and topic models in a fashion suitable for input into the powerful R statistical software environment (see modmag.ca/publications/ readingmodernism/). Many of the images presented here were produced REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 147 using ggplot2, a data visualization package for R. The simpler line and bar graphs were made using Windows Excel’s built-in graphing functionality. Combining several reading strategies allowed us to analyze the enormous quantity of material contained in the WHM . As Nelson notes, “Topic modeling and other distant reading methods are most valuable not when they allow us to see patterns that we can easily explain but when they reveal patterns that we can’t, patterns that surprise us and that prompt interesting and useful research questions” (par. 15). Distant reading tools can also be used to supplement traditional close or attentive reading practices. Topic models might suggest keywords deserving of particular attention; for example, knowing the importance of gendered terms such as man, woman and girl across the WHM corpus alters the signifi cance of these terms when they are read within the context of particular editorials, serialized literary texts and advertisements. Similarly, RapidMiner’s statistical data reveals the emergence of new technologies, such as radio or cinema, and suggests specifi c digests or issues to examine using more traditional methods of scholarly enquiry. Collections of material comprised of several million words can only be parsed effectively using a combination of reading practices. Current objections to distant reading as a legitimate mode of critical engagement often fail to consider not only the historical contingency of all reading, but also the benefi ts of hybrid analyses that embrace distant readings, close readings and everything in between.

REMEDIATION The digital remediation of The Western Home Monthly has been surprisingly revealing of those mediating frameworks that allow periodicals, including magazines and newspapers, to establish their identities across years of shifting content. We can discuss these repeated features—monthly columns, familiar writers, typefaces and column widths—in terms of Jay David Bolter and Richard Grusin’s theory of remediation, which emphasizes the relationship between immediacy and hypermediacy . Immediacy, or “the transparent presentation of the real” might take the form of those monthly features that become invisible through their familiarity. 24 Conversely, hypermediacy, or “the enjoyment of the opacity of media themselves” (21), is constituted in moments when the mediated nature of the magazine is foregrounded—often in the process of trying to train readers to use this new media object and, as a result, to see it as immediate once again. 148 H. MCGREGOR AND N. ORDEN

The July 1919 edition of the editorial titled “A Chat With Our Readers” provides a strong example of the interplay between immediacy and hypermediacy. This “Chat” section varies across the many issues in which it appears between 1909 and 1921, sometimes including letters from readers, other times directly addressing readers on topics the editors situate as particularly important. In the July 1919 issue, they focus on “The Value of Advertising:"

[M]any of the things we count today as necessities or simple luxuries could not be made and sold at their reasonable prices except as advertising has created a broad market for them, making millions of sales at little prices and little profi ts. And so you owe very much to advertising. You owe much to the people of yesterday, who have read and been infl uenced by past advertising, and so have made possible the economies and varieties, and wide distribution of merchandise that you enjoy. 25

This page in July 1919 (Fig. 6.2 ) looks much like the corresponding page in the preceding and following issues. Its formal features constitute those invisible protocols or mediating frameworks that disappear in the process of reading—from the title and volume information in the header, to the page layout, to the blank subscription form at the bottom. These protocols—issue numbering, subscription rates and layout—can be read as examples of Bolter and Grusin’s immediacy, the naturalized and seemingly transparent presentation of formal media characteristics. In this issue, however, the editors’ explanation about how these dimensions of the magazine as a medium are meant to be read shift the magazine’s advertisements from immediacy to hypermediacy. In the process of trying to teach readers how to use the magazine properly, the editors apply pressure to the medium, rendering it temporarily opaque rather than transparent and revealing how strongly the magazine was invested in and moved by advertising. It also gives us an opportunity to investigate what kind of impacts an ever-increasing emphasis on advertising as a medium may have had on the magazine. The Western Home Monthly underwent a series of important shifts between 1919 and 1920. Full-color covers were introduced in January 1920, and the use of color gradually extended into glossy advertisements throughout the magazine. In this period, there was also a rapid increase in features on, and advertisements for, radios. REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 149

Fig. 6.2 “A chat with our readers” in the WHM , July 1919 (Image courtesy of Peel’s Prairie Provinces (peel.library.ualberta.ca), a digital initiative of the University of Alberta Libraries) 150 H. MCGREGOR AND N. ORDEN

Fig. 6.3 The fall of the phonograph and the rise of radio in WHM advertisements

As demonstrated by Fig. 6.3 (produced using the ggplot 2 package for R), the shift away from the phonograph and toward the radio as the dominant new technology occurred rapidly and was largely driven by advertising, the space in which readers’ relationships to new technologies was most actively negotiated. These gestures toward modernity (color printing, new forms of media) continue to share space with signs of the old; the sheer range of printing techniques at work, and of technologies alluded to or directly mediated within the magazine’s pages, make it a fascinating object. The contrasts are often striking, with experimental photo-collages of “Motherhood of Many Lands” (May 1929) followed by traditional woodcut-illustrated advertisements for “Canada’s Pulp and Paper Industry” (June 1929). The catalyst for much of this change was advertising, and the mosaic of different printing techniques served in themselves as advertisements for Stovel Printing’s ability to keep pace with Winnipeg as the city moved forward into modernity. The advertisements in the magazines are also at the forefront of its modernization—they are the fi rst color features, the fi rst to incorporate glossy paper and the fi rst to introduce new media into their hypermediated aesthetics. Another impact that advertising had on the magazine becomes apparent in a major shift in its layout. At the beginning of 1919, the magazine followed the same layout protocols it had since 1899: editorial features were presented sequentially and in their entirety, extra spaces fi lled up REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 151

Fig. 6.4 Graph demonstrating sudden increase in fragmentation in the WHM in 1919 with miscellaneous items including random photographs and anecdotes likely pulled from other publications. Beginning in October 1919, the magazine’s layout underwent a dramatic shift, offering the fi rst few pages of each feature and then continuing the piece on later pages, with the distinctly modernist outcome of fragmented narratives and collage-like spreads. Figure 6.4 (produced using Microsoft’s Excel’s graphing function) demonstrates the marked rise of this technique, charted through the magazine’s use of the phrase “continued on,” which signaled the discontinuous content scattered across an issue. While this dynamic remained prevalent through the rest of the magazine’s run, Fig. 6.5 offers a clear example of the collage-like pages that resulted from this new non-linear layout. These facing pages show how the accumulation of multiple fragments onto a single page frees up more full-page space for advertisements. The resulting collage brings together a travel narrative about hunting on the Pacifi c Coast, a whimsical story of fairies bringing seeds to life, an adventure tale set on the Dakota Plains and an article advising women on comfortable modern dress. The impact of this shift in page layout was an increased emphasis on what Latham calls “emergence.” Emergence, according to N. Katherine 152 H. MCGREGOR AND N. ORDEN

Fig. 6.5 A collage of text leaves more space for advertising in October 1919 issue of the WHM (Image courtesy of Peel’s Prairie Provinces (peel.library.ualberta.ca), a digital initiative of the University of Alberta Libraries) REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 153

Hayles, “is any behavior or property that cannot be found in either a system’s individual components or their additive properties, but that arises … from the interaction of a system’s components.” 26 Rather than reading magazines as mechanically reproduced and lexically stable print objects, Latham argues that the magazine’s “wide affordances”—how readers can interact with it in unpredictable ways—produce divergent and individual readings, the result of which is “the phenomenon of emergence: the creation of meanings and behaviors generated by the multiple ways in which textons [strings of signs] can interact with one another” (4). He concludes by drawing modern magazines into a media history that includes literary modernism as well as hypertext and other new media practices (4). The wide affordances of non-linear layouts, the resulting emergence and the aesthetic modernism of The Western Home Monthly appear to have been motivated not by philosophical or political attempts to undermine the hegemony of traditional linear narratives, but by the more practical and material concerns of advertising and print technologies. While Latham’s focus remains on the avant-garde aesthetics of modernism, his arguments prove equally true in the case of middlebrow and mass magazines—an observation that further blurs the lines between modernism and mass culture. In Latham’s terms, distant reading can help produce different possible “paths” through the digital WHM , paths which highlight the meanings and behaviors generated by the multiple ways in which “textons,” such as pages, illustrations and advertisements, interact with one another. Figure 6.6 (produced using Microsoft’s Excel’s graphing function) shows paths representing the number of advertisements, illustrations, pages and words (divided by one thousand) annually across the WHM . The importance of advertising is immediately evident in this image; ads dominate nearly every year. In general, the number of advertisements, illustrations, pages and words published change in direct relation to each other, rising and falling roughly in unison. As noted above, world historical events (such the stock market crash) and changes in printing technologies (such as the incorporation of color images) can be mapped closely onto these paths, dramatic events often precipitating sudden shifts. Combining close and distant reading practices is crucial to interpreting this information and discerning these connections. For example, careful analysis of several issues of the WHM around 1925 reveals that the increase in advertising between 1924 and 1925 was likely related to the incorporation of color advertisements throughout the magazine. Starting with two-tone colors early in the year, advertisements in 1925 quickly developed into 154 H. MCGREGOR AND N. ORDEN full-color glossy inserts featuring everything from fl ooring and furniture to bacon and biscuits (see modmag.ca/publications/readingmodernism for a full-color example). During that same period, the number of pages and words increased in proportion to the increase in advertisements, but the number of illustrations stagnated—suggesting that space previously allotted to illustrations was being absorbed by the increase in color advertising. These closely linked metrics in the mid-1920s are particularly interesting because this period represents arguably the most formally settled period in the magazine’s history. Although ads, illustrations, pages and words were generally directly related in earlier years, their relation was often not proportional—the massive increase in advertising between 1907 and 1908 was accompanied by only small increases in page numbers and illustrations. From 1922 to roughly 1930, however, the form of the magazine remained relatively stable; Fig. 6.6 shows all four paths following similar trajectories. This period corresponds almost exactly with the dramatic rise in the fragmentation of the texts within The Western Home Monthly —use of the “continued on …” convention increased from 74 in 1921 to a high of 383 in 1929. This combination of relative formal stability and increased fragmentation suggests that this period in the magazine echoed the development of typically modernist literary forms, particularly narrative fragmentation and the mixing of genres and voices. Many promi-

Fig. 6.6 Advertisements, illustrations, pages and words (/1000) in the WHM REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 155 nent modernist texts featuring these same characteristics were published during the same period, including: Eliot’s The Waste Land (1920), Joyce’s Ulysses (1922), Woolf’s To The Lighthouse (1927) and Faulkner’s As I Lay Dying (1930). While it is impossible to claim any direct link between The Western Home Monthly and these various writers, strong formal parallels suggest lines of continuity between the middlebrow magazine and literary modernism. We argued above that the digital remediation of The Western Home Monthly puts us in a better position to understand the interplay of immediacy and hypermediacy on the magazine’s pages, and we have shown how this dimension of the magazine was constantly pushed by advertising and the development of new print technologies. Modern magazines like The Western Home Monthly are dense with remediation, incorporating old media like newspapers, woodcuts and paintings, as well as new media like advertising, photography, radio and fi lm. Despite the methodological challenges that an enormous archive like this one poses to us, it also offers an unprecedented opportunity to track the dynamics of remediation across pages, issues and years. In the process, it allows us to formulate more clearly an argument about how the dynamic processes of remediation led to the adoption of distinctively modernist forms on the pages of a middlebrow magazine.

REMEDIATING NEW AND OLD MEDIA In the fi nal section of this paper, we turn to a single magazine issue as a case study for reading modernist forms against the dynamic of remediation in The Western Home Monthly . In the third installment of Martha Ostenso’s novel Wild Geese , printed in the October 1925 issue of The Western Home Monthly , the four young protagonists gather for an illicit party:

They sat about in the sitting room for an hour listening to Mark’s phonograph. Judith had heard one in the home of an Icelander, but it had a horn and had not produced the alluring music that she listened to now. Her eyes grew dark and absent as she let her emotions drift with the spirit of the dance. A waltz played, and she feared that she would cry before it came to an end. Lind and Mark danced a little, and Jude watched them enraptured. It was all so new to her, and yet it seemed the thing to which she belonged. 27 156 H. MCGREGOR AND N. ORDEN

For the sheltered farm girl Judith, the phonograph stands in for all things modern: youth, sexual freedom, art and the newness of technology and experience. The phonograph, and the world that it signifi es beyond the farm and its deprivations, is one of a range of images that signal Judith’s temptation away from tradition and familial duty and into an illicit pre- marital relationship with the city-dwelling Sven. As old-fashioned as this scene might seem, Wild Geese was far from an old-fashioned novel in 1925. Faye Hammill has convincingly demonstrated that Canadian cultural com- mentators of the 1920s classed the novel as “objectionably realistic” in a period when “realism” tended to be equated “with explicitness about sex.” 28 While the dominant popular literary mode of the time was sentimental romance, Wild Geese is rife with themes that refuse the sentimental, including “illegitimate sexual relationships, jealousy, cruelty, and the violent overthrowing of patriarchal authority” (87). Stigmatized in its time, the novel has since been consecrated as an early work of prairie realism that helped pave the way for the intertwined movements of realism and modernism in Canadian literature. Understanding Wild Geese as a novel wedded neither to nostalgia nor to tradition, we must still ask why the phonograph plays such a central role. As Fig. 6.3 demonstrates, in 1925 the phonograph was far from a dominant technology, having fallen off drastically in appearances in the magazine as of 1920. The dominant new medium at the time was radio. The scatterplot in Fig. 6.7 (produced using the ggplot2 package for R) draws on topic modeling results to show with what other terms radio frequently co-occurred. This graph demonstrates two important trends: fi rst, that the rise of interest in radio happened quite rapidly—as indicated by the steep trajectory of the “Radio topic (65)” box; and second, that this rise coincided so neatly with the serialization of three particular novels that the word “radio” clusters with character names, which fall within the box labeled “Wild Geese, The Flame of Courage, Mantrap topics (18, 3, 0)”. Topics 18, 3, and 0 correspond, respectively, to the novels Wild Geese by Martha Ostenso (August 1925–January 1926), The Flame of Courage by George Gibbs (February–July 1926), and Mantrap by Sinclair Lewis (August–December 1926). The signifi cant overlap between the novel topics and radio topic shows that many of these terms co-occurred frequently. In the October 1925 issue of The Western Home Monthly , the word phonograph appears seven times, while the word radio appears 187 times. Beyond its single instance in the novel, “phonograph” appears in two other items: an article entitled “Advances in Radio,” in which it stands REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 157

Fig. 6.7 Scatterplot showing distribution of new media topics in the WHM in as the medium of the past whose progression might serve as a predic- tor for the future of the radio; and an advertisement for J. J. H. Maclean & Co. Limited, a piano and organ manufacturer offering phonographs on clearance. This smattering of references to the phonograph is densely surrounded on all sides by radios—from double-paged color advertisements to passing notes in the Classifi eds, to an abundance of editorial content debating the future of the technology, its appropriate uses, and its social impacts. An article comparing Quebec to the rest of Canada lists 158 H. MCGREGOR AND N. ORDEN radio, along with “[j]azz, joy rides, gasoline fi lling stations fi tted out with the pomp of pleasure gardens, [and] cross-word puzzles” as “part of the immense equipment of an overmaterialized civilization” that is corroding our “personal and national potency.” 29 Hardly a page goes by in which the magazine isn’t overtly working through the rising cultural dominance of this new medium. In this moment radio is remarkably hypermediated. In their introduction to an article entitled “What a Radio Manufacturer Thinks about Radio Editorials,” the anonymous editorial “we” insists that the era of radios being an experimental, homemade apparatus are long passed; radios have instead become another standard household good, reliably mass- produced by corporations. 30 This overt editorial framing of what radios are suggests a moment in which the signifi cance of this new medium is actively contested, a suggestion reinforced by the multiple overlapping and confl icting versions of the radio remediated in this issue alone. A single representative advertisement (Fig. 6.8 ) connects the hypermediacy of the radio to the scene in Wild Geese that features Mark and his phonograph. The two-page full-color advertisement was new to The Western Home Monthly in 1925; it represented an unprecedented level of technological sophistication for the magazine and was among the most expensive advertising spaces they offered. In its combination of text and image, modernity and tradition, this ad for Fada Radios exemplifi es almost everything that Richard Ohmann argues is characteristic of modern advertising from the early twentieth century. To begin, it combines all three of the forms of visual abstraction that Ohmann outlines: the icon, the symbol and the index. 31 The icon, the object itself abstracted from the surrounding world, appears in the form of the two radios in the lower left-hand side of the ad. The symbol, or trademark, appears in the form of the grand piano icon that accompanies the slogan “The Grand Piano of the Radio World.” The index, “an image of people, places, or occasions to be somehow associated with the product and its use” (182), effectively ties icon and symbol together, showing how a radio can be like a grand piano. In this stylized illustration, well-dressed men and women sit and stand around an elegantly appointed room observing a concert. The grand piano in this room echoes the symbol below, while the sweeping line that frames the index draws the viewer’s eye back to the brand name, insistently linking the two. The text more clearly explains this link: “To engage artists for the entertainment of one’s friends was once the privilege of only the wealthy. The advent of radio, however, enabled every home to share this pleasure.” REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 159

Fig. 6.8 Two-page full-color advertisement for Fada Radios in the WHM , October 1925 (Image courtesy of Peel’s Prairie Provinces (peel.library.ualberta. ca), a digital initiative of the University of Alberta Libraries) 160 H. MCGREGOR AND N. ORDEN

The value of modernity is expressed here through an aspirational gesture toward traditional forms of wealth and elegance, confl ating democracy with commodity acquisition. In this way the advertisement refl ects what Edward P. Comentale has described as a “logic of commodifi cation” and “faith in the activity of production” that characterized the aesthetics of the early twentieth century, including modernism and the avant garde. 32 At the same time, this image suggests the ambivalence toward modernity that Ohmann argues is typical of advertising of the period (206). While the “now” of the radio is more advanced, more accessible, more available to the reader, the “then” of the elegant gathering is implicitly more desirable. The modern technology that puts that elegance in the reach of the viewer is a sign of the same modernity that makes that elegance a thing of the past. The similarity between this Fada advertisement and the passage from Wild Geese is evident: young people gathered around a scene of music- making in an image that simultaneously harkens to an absent past (the elegant home concert, the newness of the phonograph) and points to the modernity of the present (the new technology of the radio, the rejection of traditional mores and agrarian lifestyles). Through this consonance of imagery, the outdated medium of the phonograph is drawn into conversation with the contested new medium of the radio, much as the phonograph is used as a parallel for the radio in the magazine’s discussions of new technology in the twentieth century. The result is that a seemingly nostalgic scene is pulled into a collage of competing meanings through which multiple possible courses can be charted. The collage-like texture that we discussed above as a property of individual magazine pages post- 1919 can thus be read across an entire magazine issue, not only the page. And just as the collage-like page represented a shift to modernist formal aesthetics prompted by print technologies and the dominance of advertisements, here we can see how the aesthetic experimentation at the level of ads and page layout can shift the meaning of literary texts as well. If, as Jessica Pressman argues, modernism is invested in “renovating the past through media” (emphasis in original), 33 then the relationship between past and present played out in the remediation of a new medium like the radio constitutes a modernist moment in The Western Home Monthly . There are many such moments to be discovered, and the methods outlined in this paper are key to such discovery. As much as remediation in The Western Home Monthly has been our focus, we have simultaneously remained cognizant of the effects that the remediation of the maga- REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 161 zine into the digital WHM has had on our arguments and conclusions. Our analyses have combined moments of immediacy and close reading with the distinctly hypermediated data of distant reading. We set out to investigate two questions: what does modernism look like on the pages of The Western Home Monthly , and how have digital reading methods helped us to understand it better? Methods of distant reading allow us to navigate the enormity of a resource like the digital WHM , fi nding meaning in its lines of code that in turn lead us back into the magazine. As J. Stephen Murphy explains, “visualization is not simply a fast-forward button, doing the work we would otherwise do much more slowly. Visualization also helps researchers direct their research by helping them see relationships among data that would be otherwise obscured.” 34 Modeling the magazine as data through digital remediation has given us a way into an archive that could otherwise be approached only through sampling, allowing us to identify patterns that in turn have led to answers about the development of modernist forms in The Western Home Monthly .

NOTES 1. “That Nothing be Wasted,” The Western Home Monthly (May 1901): 8. 2. John Guillory, “The Ordeal of Middlebrow Culture,” review of The Western Canon: The Books and School of the Ages , by Harold Bloom, Transition 67 (1995): 87. 3. “To Our Readers,” The Western Home Monthly (May 1901): 8. 4. “Press: Maple Leaf Magazines,” Time (Sept. 26, 1932), Time.com . 5. Our thanks to Shawna Ross for this concise wording. 6. Robert Scholes and Clifford Wulfman, Modernism in the Magazines: An Introduction (New Haven & London: Yale UP, 2010): 1. 7. Patrick Collier, “Imperial/Modernist Forms in the Illustrated London News ,” Modernism/modernity 19.3 (2012): 511. 8. Kirsten MacLeod, “American Little Magazines of the 1890s and the Rise of the Professional-Managerial Class,” in Magazines and/as Media: The Aesthetics and Politics of Serial Form , ed. Faye Hammill, Paul Hjartarson and Hannah McGregor, spec. issue of English Studies in Canada 41.1 (2015): 41. 9. Melissa Sullivan and Sophie Blanch, “Introduction: The Middlebrow— Within or Without Modernism,” Modernist Cultures 6.1 (2011): 4. 10. Douglas Mao and Rebecca L. Walkowitz, “The New Modernist Studies,” PMLA 123.3 (2008): 737–48. 162 H. MCGREGOR AND N. ORDEN

11. Our own research emerges from the collaborative Editing Modernism in Canada project ( http://editingmodernism.ca /), a SSHRC Strategic Knowledge Cluster project that focused on the interwoven concerns of New Modernist Studies, particularly the transnational reach of modernism beyond cosmopolitan centers. Hannah McGregor is also involved in “Nations of Print,” a research project focused on the periodical production of Canada, Scotland, and Australia—non-cosmopolitan nodes of modernist and middlebrow production. 12. Collier, 4. 13. Sean Latham, “Affordance and Emergence: Magazines as New Media,” What is a Journal? Towards a Theory of Periodical Studies , MLA Convention 2013, Special Session 384 (21 Dec. 2012): 1. 14. Sullivan and Blanch, 6. 15. James Mussell, “Repetition: Or, ‘In Our Last,’” Victorian Periodicals Review 48.3 (2015): 347. 16. Jay David Bolter and Richard Grusin, Remediation: Understanding New Media (Cambridge: MIT Press, 2000). Our defi nition of remediation draws on Bolter and Grusin’s work, in which they defi ne remediation as “the representation of one medium in another” (45), and the dialectic between old and new media that results (50). The process of remediation is in turn intertwined with the tensions between immediacy and hypermediacy, those “twin preoccupations of contemporary media” (21) that, as we will explore below, are key to how magazines mediate new materials through a framework of repetition. 17. John Mackay, “Modernism.” The Western Home Monthly June 1925: 13, 75–76. Print. 13. The “modernism” to which Mackay refers is actually one side of a theological debate raging in the Presbyterian church in the 1920s and 1930s that distinguished between modernism and fundamentalism as two possible interpretations of scripture; many of the textual occurrences of “modernism” and “modernist” within the pages of The Western Home Monthly hold this meaning, indicating why a straightforward search for particular keywords is not the most productive way to engage with the magazine. 18. Francis Dickie, “Montparnasse—Heart of the Artistic World,” The Western Home Monthly (March 1932): 18–19. 19. Mussell, 3. 20. Hannah McGregor, “Remediation as Reading: Digitising The Western Home Monthly ,” Archives and Manuscripts 42.3 (2014): 249. 21. “Digitization Services,” Backstage Library Works (Apr. 27, 2015): pars. 2–3. REMEDIATION AND THE DEVELOPMENT OF MODERNIST FORMS IN THE... 163

22. Andrea Lancichinetti et al., “High-Reproducibility and High-Accuracy Method for Automated Topic Classifi cation,” Physical Review 5.1 (Jan. 29, 2015). 23. Robert K. Nelson, “Introduction,” Mining the Dispatch (Apr. 27, 2015): par. 10. 24. Bolter and Grusin, 21. See also Lisa Gitelman, Always Already New: Media, History, and the Data of Culture (Cambridge: MIT Press, 2006), who argues that successful media must render many of their features invisible “in favor of attention to the phenomena, ‘the content’” (6). These features include a medium’s “protocols,” or the “vast clutter of normative rules and default conditions, which gather and adhere like a nebulous array around a technological nucleus” (7). 25. “A Chat With Our Readers: The Value of Advertising,” The Western Home Monthly (July 1919): 1. 26. Qtd. in Latham, 3. 27. Martha Ostenso, “Wild Geese. Third Installment,” The Western Home Monthly (October 1925): 9. 28. Faye Hammill, “The Sensations of the 1920s: Martha Ostenso’s Wild Geese and Mazo de la Roche’s Jalna ,” Studies in Canadian Literature 28.2 (2003): 86. 29. W. F. Osborne, “Quebec in Relation to Canada,” The Western Home Monthly (October 1925): 13. 30. S. D. Armour, “What a Radio Manufacturer Thinks about Radio Editorials,” The Western Home Monthly (October 1925): 41a. 31. Richard Ohmann, Selling Culture: Magazines, Markets, and Class at the Turn of the Century (London & New York: Verso, 1996): 180–4. 32. Edward P. Comentale, Modernism, Cultural Production, and the British Avant-Garde (Cambridge: Cambridge University Press, 2004): 6. 33. Jessica Pressman, Digital Modernism: Making it New in New Media (Oxford Scholarship Online, 2014): 4. 34. J. Stephen Murphy, “Introduction: ‘Visualizing Periodical Networks,’” The Journal of Modern Periodical Studies 5.1 (2014): vii. CHAPTER 7

Stylistic Perspective Across Kenneth Fearing’s Poetry: A Statistical Analysis

Wayne E. Arnold

Frequently anthologized author Kenneth Fearing (1902–61) predominantly wrote poetry during the inter-war period and was often labeled as an outly- ing Communist Party sympathizer. 1 While writing poetry and short stories dominated his early career, Fearing shifted most of his attention to novels by the late 1930s. In the 1940s and 1950s, his poetic success fl oundered, and accusations arose from critics that this later poetry appeared merely a reworking of his more impactful work of the late 1920s and early 1930s. Fearing received much attention for his early work due to the assumed connection with communism; 2 after he began publishing novels—the most successful being The Big Clock (1946)—his status as a dominant modernist poet dwindled. In order to determine if Fearing underwent a stylistic change or merely rehashed his previous poetic successes, this study attempts to perform a textual analysis with a focus on Fearing’s style across the spectrum of his published poetry. I establish the hypothesis that Fearing does undergo a stylistic shift and that his later work neither replicates nor imitates his earlier poetry; instead, the data reveals that Fearing did indeed develop and enter a new phase of poetic composition. Critics of Fearing’s poetry have focused the bulk of their attention on his fi rst two published collections, Angel Arms (1929) and Poems (1935).

W. E. Arnold () University of Kitakyushu, Kitakyushu, Japan email: [email protected]

Both publications contain poems for which Fearing is most well known, “Jack Knuckle’s Falters,” “1935,” and “Dirge,” to name a few. In all, Fearing published fi ve collections of original poetry, with the three subsequent works entitled, Dead Reckoning: A Book of Poetry (1938), Afternoon of a Pawnbroker and Other Poems (1943) and Stranger at Coney Island and Other Poems (1948). In selecting the poems for analysis, I incorporate Collected Poems (1940) since it contains twenty original poems. Additional support for including Collected Poems arises from the 5,779 word tokens comprising these verses. This total is merely fi fty tokens fewer than the sum in Angel Arms and 365 tokens more than Coney Island. Poems excluded from the analysis include twenty-one poems that Fearing published in his university days, poems that were published in magazines but never added into his poetry collections, as well as the four poems that Fearing included in New and Selected Poems (1956). In total, there are 142 poems that comprise the data set under analysis. Fearing’s premature death from malignant melanoma at the age of fi fty-nine, came thirteen years after his last published collection of new poetry; in total, this study examines the poems that span nineteen years 3 of Fearing’s poetic output.

LITERARY CRITICISM ON FEARING Fearing’s work has failed to garner widespread critical attention. Those reviewers and critics who have focused on his publications tend to appor- tion his poetry into two categories: the productive and the ineffective, or, an early and later, redundant period of writing. In the 1920s, Fearing was considered by some to have “a personality of his own,” 4 with an “irony [that] is very special, unique in the history of American Poetry.” 5 “His manner is hard and tense,” wrote one reviewer of Angel Arms , “it is dry, closely-bitten, almost anti-poetic.” 6 By his next publication, however, critics were claiming that his style appeared surprisingly replayed: “Fearing’s second book is not very different from his fi rst […,] a form based entirely on parallelism of line and phrase,” noted one reviewer. 7 Many critics have nevertheless praised his lyrical ability to convey the consumer culture of the age, which he often demonstrated through lexeme repetition. Functionally, repetition acts as a form of cohesion 8 ; repetition, as well as rhythm, serve as “involvement strategies,” 9 aiming to allure the listener or reader. It would be diffi cult to argue that Fearing’s repetition functions to create rhythm in his poems—as he eschewed rhyme 10 —although, in some poems a rhyming presence is undeniable. Repetition in advertising STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 167 is a fundamental element of marketing utilized by Fearing, who “probably learned [repetition] not from the study of classical prosody but from the observation of advertising techniques.” 11 Fearing’s strategic implementation of this methodology does not imply that his repetition was wholly unique, as Ruth Finnegan, in Oral Poetry: Its Nature, Signifi cance and Social Context (1977), has posited: “the most marked feature of poetry is surely repetition.” 12 As time passed though, the growing criticisms against his work encouraged the perception that Fearing merely attempted to rehash his former style; such claims have, in some senses, become almost a standard analysis for his poetry of the 1940s and 1950s. By the early 1940s—indeed, even by 1937, some were noting that Fearing had quickly narrowed his attention to “a single medium” 13 —critics were becoming dubious of the poet’s ability to continue forging into new territory. Himself a regular reviewer, Fearing was conscious of the criticisms accumulating concerning his publications. Long-time Wisconsin friend Marya Zaturenska, wife of literary critic Horace Gregory, recorded two notable events in her diary during 1939 and 1940 concerning Fearing’s awareness of his public reception. The fi rst entry recalls Fearing, at a party, “complaining of how badly the critics had treated him—how Horace should have defended him.” 14 The second occurred at a dinner, when Zaturenska recounts that Fearing “[said] he is going to give up poetry and turn to novel writing”; doubting his claim, she refl ects, “of course I didn’t believe him about giving up poetry, but he did seem gloomy and depressed.” 15 Censures followed the poet’s work into the 1940s, as when one critic, reviewing Afternoon of a Pawnbroker , argued that “[t]he satirical explosions that characterized his early work are subsid- ing. He is neither as sharp nor as fl ashing as he was”; the reviewer continues, “[i]f he is not always as clear as he was in his previous books, it may be because his vision is changing and he is not certain himself of what he sees.” 16 Written in 1943, these charges against Fearing are representative of a number of critics who were, not surprisingly, seeking a continuation of Fearing’s past accomplishments. The years 1938–40 mark the turning point in reviewers’ acceptance of Fearing: with “the publication of Collected Poems of Kenneth Fearing (1940) it was clear that Fearing […] had found a medium in which to write and was content to repeat the earlier successes of his writings with slight variations on a central theme.” 17 By the time Fearing published New and Selected Poems in 1956, it seems that even his early, successful poetry was destined to be meshed together with his later, less popular work as being composed of regurgitated slogans and 168 W.E. ARNOLD now no-longer-shocking imagery. As one 1957 review suggests, “Fearing’s style, for all its apparent freedom from traditional imitations, is clogged with mannerisms and archly stylized habits: you are forever confronted by the momentous rhetorical question, the cosmically signifi cant repetition, the meaningfully twisted popular slogan.” 18 Fortunately, not all critics bemoaned the presumed rehashing in Fearing’s later poetic work. Weldon Kees observed in his 1941 review of Complete Poems of Kenneth Fearing , that “when his sour wit shows signs of having curdled, and when his repetitions and lists, forceful and effective in limited amounts, become tiresome and mechanical, degenerating into a facile and overwrought shrillness, there are still more than a few poems that are exactly what their author wished them to be; they are valuable and exciting.” 19 Nevertheless, Fearing is viewed as falling into a regular style that is branded “by certain words and images that also contribute to the unity of his work as much as his consistent dark moods and ironic tone and free-verse style.” 20 Stylistically, Fearing’s poetry is representative of the period, as Alan Wald observes, “Fearing surely tilted toward modernism in applying ironic strategies to cancel expectations evoked by representational art.” 21 In these reviews, rhetorical repetition 22 is linked with Fearing’s style of presenting his thematic content; but as the data in this project will demonstrate, such presumptions fall short in accurately encapsulate Fearing’s oeuvre. More recent critics have been more accepting of the “less popular” Fearing poetry, and in the 1990s a minor Fearing revival took place with the publication of Kenneth Fearing Complete Poems (1994), edited by Robert M. Ryley. Additionally, a noteworthy thematic analysis appeared with Mark Halliday’s “Damned Good Poet: Kenneth Fearing” (2001); Halliday perpetuates Ryley’s observation that Fearing’s later work avoided the larger political events of the mid century. Halliday suggests that Fearing “seems to have felt increasingly jaded and skeptical about poetry’s chance to participate in national life.” 23 While not implicating Fearing in a weakened style or repetitive format of previous work, Halliday’s comment highlights that Fearing’s latter poetry fails to create (through poetic allusion) a new and necessary juncture in his poetic output. Fearing’s earlier style of “image-making by listing and repetition” 24 provided him the case- ment to unveil his signature image of the period. After Fearing’s passing in 1961, opinions on his early and later poetry has become more collective, as if from the perspective of a completed task: “Fearing’s true appeal as a revolutionary poet was his ability to combine realistic description and STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 169 political comment in the form of a readable poem that lost nothing of its quality as a poem while it gained in propaganda value. He died […] still very much the same kind of poet.” 25 In an obituary on Fearing intended for the National Guardian , M. L. Rosenthal mused, “Fearing remains one of our purest stylists in a brilliant age of poetry.” 26 Approaching Fearing’s poetry through a Digital Humanities perspective will, I believe, encourage a closer examination on the style and content of Fearing’s later poetry. Criticisms such as the aforementioned have served as my impetus to examine the stylistic development across Fearing’s poetry. The reviewers I am citing have commented on both Fearing’s content and style; justifi - ably, then, there is value in determining if this stylistic shift produced new (and untapped) functional avenues of expression that, as of yet, have been unparsed by Fearing scholars and may expose new elements in his content. In some respects, my project parallels the stylometric corpus work performed on Henry James’s oeuvre by David L. Hoover. In Hoover’s approach to James’s work, he observes that “an author’s various styles can be similar enough to appear alike when compared with the styles of other authors and yet different enough to be distinguished from each other.” 27 Following Hoover, I examine the clusters within Fearing’s poetry to determine if his use of repetition diminished over the years. There is need to delve below the modernist images and dark moods to the function words that, as recent stylometry research has demonstrated, are important factors in defi ning personal writing style. 28

CURRENT STUDY AND PROCEDURE Acquiring the data set required digitizing the Kenneth Fearing: Complete Poems. Scanning the text, then performing optical character recognition, generated the digital data. Once scanned, the entire corpus was proofed against the printed text to ensure that any digitization-related errors were removed and/or corrected. Each poem was separated into individual fi les, while coin- ciding publication dates incorporated into the fi le names for organization. Each corresponding fi le was placed into a folder for each book marked with the publication year. The machine learning analysis was performed with WordSmith, from Lexical Analysis Software Ltd. Since the working hypothesis is that Fearing’s poetry does demonstrate a meaningful stylistic change across the date range, the various tests have been performed on each of the six published texts. Before examining the values of each specifi ed word or phrase occurrence within a poem, the number of occurrences was normalized; 29 170 W.E. ARNOLD

normalization entails summing the number of occurrences in each poem and dividing this value by the total number of words in the poem. Likewise, for the calculated number of data occurrences in a book, the total number of designated words in the corresponding book was divided by its sum total of words. This procedure aids data accuracy so that a word appearing repeatedly in one particular poem does not skew the data for that time period. The data set size averages around 6,000 tokens per poetry collection. Microsoft Excel was used to tabulate the data output from WordSmith, and concordances were generated for select words. These word lists allow for determining how many occurrences of a particular word-phrase appears in a text as well as the total times the word is employed across the corpus. My initial interest derives from the use of pronouns in Fearing’s texts, and these results may aid in determining if there is a demonstrative subject shift in Fearing’s thematic focus. Attention to pronouns is additionally encouraged through the research of James Pennebaker, of The University of Texas at Austin. In one study, Pennebaker’s work with Shannon Wiltsey Stirman examines pronoun usage in suicidal and nonsuicidal poets (i.e., poets who did not commit suicide). They note that the “writings of suicidal poets contained more words pertaining to the individual self and few words pertaining to the collective than did those of nonsuicidal poets.” 30 For this study, Pennebaker and Stirman used a control group of nonsuicidal poets in order to make their observations on the suicidal group of poets. The brief biographical information on Fearing does not seem to suggest that he was suicidal; his well-known and extreme alcoholism, however, encourages the comparison with Pennebaker and Stirman’s work. Their suggested hypothesis for suicidal poets did not hold true for Fearing as the use of fi rst-person singular pronouns (I , me , my ) were low. Indeed, examining the data diachronically, their control group had means of early, 2.5 percent, middle, 1.6 percent and late, 2.5 percent; if we group Fearing’s work into similar categories, we fi nd lower percentages: early (AA & P ), 0.37 percent, middle (DR and CP ), 0.67 percent, late (AP and CI ), 0.77 percent (Table 7.1 ). The general synopsis by Pennebaker and Stirman’s data provides only a tentative basis for determining suicidal tendencies in writing styles. Of particular interest for thematic transformation, as well a shift in perspective, is Fearing’s gradual increase in the plural pronouns. With the exception of “we” in Angel Arms , the data suggests a movement toward camaraderie in his later work, especially in the poems of Coney Island . Looking at pronouns from another angle, Pennebaker and King (1999) STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 171

Table 7.1 Means of fi rst-person singular and fi rst-person plural across Kenneth Fearing’s poetry

Angel Poems Dead Collected Afternoon Coney arms (%) reckoning poems (%) pawnbroker (%) Island (%) (%) (%)

“I” 0.75 0.06 0.35 1.02 0.98 0.54 “me” 0.17 0.03 0.09 0.09 0.05 0.09 “my” 0.33 0.16 0.00 0.12 0.32 0.09 “we” 0.82 0.35 0.57 0.43 0.48 0.78 “us” 0.05 0.18 0.08 0.17 0.12 0.22 “our” 0.05 0.06 0.08 0.07 0.05 0.26

Table 7.2 Distribution of “you” and “your” across Fearing’s poetry

Angel Poems Dead Collected Afternoon Coney arms (%) reckoning poems (%) pawnbroker (%) Island (%) (%) (%)

“you” 0.38 2.15 2.22 2.11 1.04 0.76 “your” 0.31 0.51 0.30 0.40 0.15 0.20 have found that fi rst-person pronoun usage is negatively correlated with fi rst-person plural pronouns when writing about emotional issues. 31 Performing a pronoun search for you and possessive pronoun your highlights how Angel Arms contains a measurably reduced presence of you . As Table 7.2 shows, across the fi rst four volumes the you / your pronoun usage measure higher mean values before showing a marked decrease in the last two publications. Across Fearing’s entire poetic output, the token you has an occurrence frequency of 1.44 percent. This particular pronoun in Fearing’s vocabulary is marked with a distinct level of keyness, being present in eighty-eight of the 142 poems. Words that appear in a textual context with a qualitative frequency are considered to have a certain keyness. In the poems under analysis in this study, you has an occurrence of 539 tokens at a rate of 1.48 percent across the corpus. Comparing the key words of Fearing’s six works against the Brown reference corpus shows that Fearing’s use of you is prominent by almost double the keyness. Certain uses of you occur in Fearing’s dialogue either between characters or for other poetic purposes and appear in quotation marks. These you values within quotation marks (twenty-seven total) have been removed from the data set to 172 W.E. ARNOLD avoid possible misrepresentation of the pronoun. The occurrence rate of the you function word and the variance across Fearing’s corpus suggest a thematic shift during his middle period of writing. Angel Arms , Afternoon Pawnbroker and Coney Island have signifi cantly fewer occurrences of you , as Table 7.2 demonstrates. The variance in the use of the you pronoun brings up a two-part analysis question. If, as suggested by Fearing scholars Rita Barnard and Nathaniel Mills, Fearing uses shock tactics—in the Walter Benjamin context 32 —to reach his readers, how does the presence or lack of presence of the you pronoun infl uence the poem’s impact on the reader? Essentially, does the role of pronoun usage in Fearing’s poetry represent a shift in style and therefore alter the impact for readers? Is this change related to the critics’ reaction in the 1940s, claiming that Fearing had lost his edge as a poet? To divert briefl y, Mills argues that “Fearing crafts the unsettling aesthetic impact of shock as a stimulus toward a moral and political conviction on the part of his reader.” 33 But like many who have examined Fearing’s poetry, Mills places primary emphasis on Fearing’s poetry up until roughly 1938, ten years before Fearing’s last collection of new poetry. Examining these stylistic changes brings up the question of perspective, as scientifi c studies have shown that “using the pronoun you to describe an action […] leads readers to mentally simulate the action from the performer’s perspective, whereas the pronouns I and he promote an observer’s perspective.” 34 From this stance, then, we could hypothesize that one of the varying stylistic alternations that Fearing’s poetry undergoes is a reader perspective shift directly related to observer/performer. The pronoun data implies that Fearing’s poetry does change. Following Pennebaker’s analysis, the state of negative emotion becomes apparent through pronouns in the expression of anger. In The Secret Life of Pronouns , he notes that “[w]hen angry, people focus on others and rarely on themselves. In addition to using high rates of second-person (you ) and third-person (he , she , they ) pronouns, angry people talk and think in the present tense.” 35 Bearing in mind that you escalates through the fi rst three Fearing publications, calculating the means for the third-person pronouns suggests that anger may be more thematic throughout the work of the latter period. There is, however, confl icting data with Dead Reckoning , which shows a dramatic decrease in the employment of he and she . 36 After the publication of Angel Arms , there is a continual decrease of she across his work, while the other two pronouns in Table 7.3 remain STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 173

Table 7.3 Means of third-person pronouns across Fearing’s poetry

Angel Poems Dead Collected Afternoon Coney arms reckoning poems pawnbroker Island

“he” 0.5098 0.2464 0.1092 0.5507 0.4768 0.6965 “she” 0.3660 0.1095 0.0624 0.0344 0.0159 0.0183 “they” 0.5490 0.5613 0.2496 0.5851 0.6357 0.4765 relatively consistent, except for Dead Reckoning , which shows an obvious decrease in all three pronouns. These variances seem to be a thematic- based decrease that needs more exploration at the poem level. One biographical element of interest between the publication of Angel Arms and Poems , as Robert M. Ryley highlighted in private correspondence with Fearing’s friend, Carl Rakosi, is that from 1930 to 1934, Fearing published just one article—apparently, no poems were published. 37 To elaborate on the lexical dispersion across Fearing’s poetry collections, specifi c Machine Learning features in WordSmith generate detailed consistency reports. A detailed consistency report measures whether Fearing’s publications exhibit lexical overlapping between one work and the next or to what extent vocabulary differentiation appears between the beginning and the end of his career. To calculate consistency reports, WordSmith compares both the content and function words of one book against all the other books to determine which two texts are most similar or dissimilar. The report assigns a relation percentage useful for emphasiz- ing any lexical commonalities. Table 7.4 reveals a larger overlap between Fearing’s later works; these commonalities, however, may be related more to function words rather than content words. The most noticeable lexical convergence occurs with Dead Reckoning and Collected Poems (38.53 percent), highlighting what I consider the 1938–40 transition period. That Collected Poems maintains a higher consistency relationship with the next two publications (CP with AP , 37.4 percent; CP with CI , 36.87 percent) than with the fi rst two works (CP with AA , 33.41 percent; CP with P , 34.07 percent) implies that the stylistic shift after this period was continued. The fi rst two publications share a high level of consistency and signifi es that these works are more closely related to each other than with the later poems (AA and CI overlap by 32.15 percent). The data for Coney Island supports the hypothesis of a separation from Fearing’s writing before 1938 period, since the consistency levels 174 W.E. ARNOLD

Table 7.4 Consistency percentages across Fearing’s poetry

39.00% 38.00% 37.00% 36.00% 35.00% 34.00% 33.00% 32.00% 31.00% Dead Collected Afternoon Coney Angel Arms Poems Reckoning Poems Pawnbroker Island (1929) (1935) (1938) (1940) (1943) (1948) Angel Arms (1929) 0.00% 36.31% 35.54% 33.41% 33.08% 32.15% Poems (1935) 36.31% 0.00% 37.47% 34.07% 35.05% 33.16% Dead Reckoning (1938) 35.54% 37.47% 0.00% 38.53% 37.16% 37.26% Collected Poems (1940) 33.41% 34.07% 38.53% 0.00% 37.40% 36.87% Afternoon Pawnbroker (1943) 33.08% 35.05% 37.16% 37.40% 0.00% 37.11% Coney Island (1948) 32.15% 33.16% 37.26% 36.87% 37.11% 0.00% among these last four works are comparable. While the percentage spread is arguably minimal (roughly 5 percent difference across all publications), the divergences highlighted here become more meaningful when we look at the n-gram distribution.

TYPE/TOKEN RATIO AND N-GRAM REPETITION The type/token ratio (TTR) is a method of looking at the number of distinct words within a writing sample. Each unique word is counted only once as a discrete type; a token is a single word, and regardless of frequency every word used is counted as a token. The TTR is calculated by summing the individual types (always the smaller number) and then dividing this value by the sum total of tokens. For this project, the TTR aids in determining uniqueness throughout Fearing’s publications as well as comparing his work with other poetry collections. The tally of unique types is generated through a wordlist, calculated by WordSmith. For each book, the sum of the types was divided by the sum of the tokens. The outcome (see Table 7.5 ) reveals a relatively homogenous rate of type usage with only the 1938 Dead Reckoning exhibiting a lower level of uniqueness within the poetry. Fearing’s average TTR across all six books is 30.22 percent. If Fearing does over-employ repetition in his poetry, the outcome would arguably STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 175

Table 7.5 Type/token percentage across Fearing’s poetry

Angel Poems Dead Collected Afternoon Coney arms (%) reckoning poems (%) pawnbroker Island (%) (%) (%) (%)

Type/token 30.29 31.27 27.96 29.68 30.91 31.23 contain a noticeably lower percentage in the TTR, one in which the types decreased due to repetition. Expanding the analysis, Fearing’s TTR is only slightly higher when compared with poets of his period. I examined three poets who were either infl uential on Fearing or his contemporary. In order to ensure a proper analysis, and since these other works were longer in word count, the control groups were broken into 5,000–5,500 word blocks and then calculated for the TTR. Ryley suggests that Edgar Arlington Robinson was a model for some of Fearing’s poetry of the 1920s. 38 The TTR for Robinson’s The Man Against the Sky (1916) reveals 24.97 percent (5133 tokens). For Robinson’s The Three Taverns (1920), the TTR is 21.46 percent (5426 tokens). Another infl uence on Fearing was Carl Sandburg, his Selected Poems of Carl Sandburg (1926) reveals a TTR of 29.54 percent (5296 average tokens). William Carlos Williams, a personal acquaintance with Fearing, 39 shares a similar TTR in Sour Grapes (1921), with 29.67 percent (6728 tokens). From the perspective of unique words, Fearing’s poetry demonstrates a slightly higher percentage, suggesting that his compositions, while containing some repetition, is quite lexically diverse. Fearing’s TTR gives hint to only minor and expected shifts between volumes; it is when we examine the n-gram repetition, however, that a noteworthy change across Fearing’s work appears. N-grams represent a sequence of words appearing in succession. For this study, the n-gram clusters were set at a minimum of 3-grams and a maximum of 8-grams (in Fearing’s work the largest number of repeated words in a single sentence is seven). The n-grams were calculated with WordSmith by creating an index of each book of poems. An index generates a digital marker for each word position in order to locate clusters of words. Once the index is created, the n-grams can be determined by using the Clusters option. Punctuation was used as a stop marker in order to demarcate more precise clausal repetitions. Upon generating the n-gram list for each book, it was then necessary to eliminate duplicate entries. For instance, when calculating the n-grams in 176 W.E. ARNOLD

Table 7.6 N-gram distribution across Fearing’s poetry

10.00% 9.16% 8.29% 8.00% 6.43% 6.00% 3.81% 4.00% 3.07% 2.11% 2.00%

0.00% Angel Arms Poems Dead Collected Afternoon Coney Island Reckoning Poems Pawnbroker

Angel Arms , the phrase “on the fl oor” and “on the fl oor of” are both listed with a frequency of three occurrences, therefore, the shorter of the two was removed due to redundancy in the list. Additional precautions were taken for other forms of repetition. In Dead Reckoning , the phrase “do you remember” occurs seven times, while the phrase “do you remember that” occurs three times; accordingly, the shorter phrase was reduced to a count of four to remove the duplication. After the n-grams were determined, the total sequence value was calculated by multiplying the word count of each phrase by the frequency of occurrence. These values were totaled and then divided by the entire word count of the book for normalization. Table 7.6 shows the distribution of the n-grams through the six books. The fi ndings shown in Table 7.6 are important because they clearly demarcate a turning point in Fearing’s poetic construction: Dead Reckoning is dominated by Fearing’s repetitive technique. A factor for the n-gram decrease potentially arises from sentential length variability, as this compositional element may directly infl uence the stylistic construct that encouraged Fearing’s use of repetition. Table 7.7 provides the average sentence lengths (determined by a full stop) across the six books and again emphasizes a break in composition style during 1938–40. The sentence is the largest form of organization in grammar, and though there is debate over its “precise theoretical status,” 40 it is an essential element in style (as in the long sentences of Faulkner). What is noticeable in the poems is that Fearing varies his punctuation style. In the poems, he enacts what I call “sentential manipulation” by using other forms of punctuation: the semi-colon, the question mark or the dash, sometimes even STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 177

Table 7.7 Mean sentence length across Fearing’s poetry

60 58.03

50 43.32 40

30 24.09 20.93 16.9 20 13.59 10

0 Angel Arms Poems Dead Collected Afternoon Coney Island Reckoning Poems Pawnbroker

Table 7.8 Punctuation use across Fearing’s poetry

Angel Poems Dead Collected Afternoon Coney arms reckoning poems pawnbroker Island

“.” 309 242 86 333 230 84 “?” 26 62 83 98 67 14 “—” 10 24 33 7 31 95 “?—” 0 13 2 0 0 5 “;” 67 133 95 10 40 42 Total poems 27 21 30 20 22 22

appearing together: “?—”. Focus on these forms of punctuation reveals a part of his stylistic method (Table 7.8 ). Again, DR and CP (years 1938–40) mark the transition period for Fearing, as both texts incorporate the highest number of question marks, while CP shows a marked difference with other forms of punctuation. There is a noticeable relation within DR concerning the n-gram distribution, sentence lengths,and punctuation. A direct correlation between repetition and sentence length is not clear; examining the use of punctuation, however, elucidates a distinguishable range of distribution throughout the beginning or the end of his publications. A quick look over the later poems reveals that Fearing does employ a shorter 2-gram usage. For instance in “Decision,” published in Coney Island , Fearing writes “Do not be dismayed,” and then a few lines later, “Do not forget.” 41 The shorter n-gram presence encourages closer poem-level evaluation as 178 W.E. ARNOLD many of Fearing’s repetitions occur at 2-gram and single word frequency. In “What If Mr. Jesse James Should Some Day Die?,” Fearing uses repetition in the following manner: “O dauntless khaki soldier, O steadfast pauper, O experienced vagrant, O picturesque mechanic, O happy hired man”; 42 even the 2-gram search will not catch this obvious repetition. Even so, Fearing’s use of repetition at the smaller n-gram level does suggest a departure from his earlier, longer 3- to 7-gram tendency.

DISCUSSION This research started with the hypothesis that Fearing’s poetic output does undergo stylistic change over the nineteen-year period of his publications of poetry. The data certainly indicates that there is a period of distinctive change between the years 1938 and 1940. Additionally, Angel Arms has a few peculiar traits that make it stand apart from the following book, Poems ; 43 I am thinking of the pronouns she , I and in particular, you . The consistency reports diachronically illustrate that each sequential book is closely related in word usage with either the preceding or following publication. Dead Reckoning and Collected Poems are the most correlated with each other, despite the sentence length changes. These two works are also important in their dominance of n-gram repetition; it is possible that this overlap occurs due to the relatively close period of composition and publication. Sentence length is a serious area for future analysis as it may indicate that Fearing’s poems have become more complex in their content and construction. 44 The pronoun data highlights distinct implications on both Fearing’s poetic content as well as his stylistic changes (see Tables 7.1 , 7.2 and 7.3 ). In particular, you and she are worth briefl y noting. Poems , Dead Reckoning and Collected Poems all show you as being greater than two percent of the token count. Conversely, the most striking change is the uninterrupted decreasing presence of she in the poems. This decrease may be the result of Fearing providing names for more female characters rather than using the pronoun; Pennebaker’s work with negative emotions does encourage a thematic analysis of these particular pronouns. Also, the fi rst person pronoun, I , when compared with you , unveils a conspicuous development in two later Fearing books. Collected Poems and Afternoon Pawnbroker both have roughly one percent of their word count consisting of I . In this instance, Collected Poems overlaps with the greatest percentage of you and I combined, which again corresponds to the 1938–40 stylistic shift. STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 179

There surfaces a tendency in Angel Arms and Fearing’s last work, Coney Island to have potential converging trends. Us and our have minimal values; whereas, the predominance of we in CI hints at a revived collective presence in the poems. They also share lower values of you and your . Consistency data in Table 7.4 also emphasizes that these two works are the least uniform with each other. Data such as the n-gram distribution as well as sentence length also lends itself to suggest that these two works do not parallel each other in compositional style. When the data started to reveal a signifi cant transition around the years 1938–40, it was tempting to correlate it with a shift in Fearing’s authorial interest, perhaps marked by the publication of his fi rst novel, The Hospital (1939). 45 This correlation, however, proves too weak, as Fearing had already written three unpublished novels during the 1930s. 46 Ryley’s seminal introduction to the complete poems represents only a portion of his work on Fearing. In an unpublished chapter, intended for a book-length publication on Fearing, Ryley examines a very crucial circumstance concerning the printing of Dead Reckoning . Due to technical diffi culty, Herbert R. Cahn of Random House requested that Fearing aid the compositor with the lineation for the new collection of poems, and, as Ryley notes, “probably for the fi rst and only time in his career, [Fearing] directly supervised the setting of type for his poems.” 47 Another substantial characteristic to which Ryley draws attention arises in the poetic construction: “poems in Dead Reckoning are silently grouped by pattern of lineation.” Specifi cally, there are four patterns that Ryley observes; the fi nal, simpler pattern was adapted by Fearing in all of his subsequent poetry publications. The stylistic impact of this pattern reveals itself in the detailed consistency report (Table 7.4 ). Such biographical material—which surfaced two years after the data for this project was calculated—lends vital support to the implications of the statistical analysis. Other biographical details for this period, such as Fearing purposefully distancing himself from the communist party in 1939, 48 may have also contributed to a stylistic shift; these biographical perspectives, therefore, reinforce rather than detract from the statistical analysis of the poetry.

CONCLUSION Performing this data mining opens up new perspectives for future analysis of Fearing’s poetry. For instance, as noted above, the dramatic decrease of the pronoun she after Angel Arms may serve as one basis for thematic 180 W.E. ARNOLD analysis when we consider Fearing’s troubled relationship with Margery Latimer. Of main interest is the increase and then drastic decrease of the n-gram repetition. Fearing’s awareness of the reviews clearly impacted his mood and likely his poetic output, as they pointed to his “verbal monot- ony.” 49 The continued regression of the n-gram repetition after 1940 has not been examined properly, but as this study demonstrates, there are grounds for further poem-level examination. Finally, the current study does not attempt to examine the infl uence of repetition through poetic analysis; rather, it seeks to bring it to the attention of those interested in Fearing and to highlight that there is not a consistent methodology across his work. What is apparent, however, is that the hypothesis under which this work began has been established as calculable: Fearing’s poetic style does change across his six publications.

NOTES 1. An early draft of this study was presented at the English Literary Society of Japan , 2014 conference. I would like to thank Dr. Claiborne Rice, The University of Louisiana at Lafayette, for his advice with the preliminary data calculations. 2. Wald, Alan M. American Night: The Literary Left in the Era of the Cold War (University of North Carolina Press, 2012), xiv. 3. In 1926, Fearing fi rst began publishing the poems that would eventually comprise Angel Arms . Additionally, his poetic output spanned over thirty years. 4. “Check List of New Works.” The American Mercury (July 1929), xviii. 5. Dahlberg, Edward. “Kenneth Fearing: A Poet for Workers.” New Masses (May 21, 1935), 24. 6. Untermeyer, Louis. “Dollar Poets.” The Saturday Review (Nov. 23, 1929), 450. 7. Walton, Eda Lou. “Fearing, Poet of a Transformed World.” New York Herald Tribune (July 7, 1935), F5. 8. Halliday, M. A. K., and Ruqaiya Hasan. Cohesion in English (New York: Routledge, 2014), 5. 9. Tannen, Deborah. Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse . 2nd edn (Cambridge: Cambridge University Press, 2007), 32. 10. In 1938, Fearing shared some of his opinions on rhyme through an interview in the Daily Worker : “‘Great poetry […], such as is to be found in the Bible and Odyssey, or in the works of Shakespeare and Whitman, does not rhyme. For rhyme is a manifestation that the poet is unable to let STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 181

his thoughts and feelings stand on their own legs. […]. One can write a propaganda poem without thyme and it will be just as popular.” See: Tilkin, Anita. “The Poet of Irony.” Daily Worker (Dec. 20, 1938), 7. 11. Milner, Thomas Howard. “The Early Poetry of Kenneth Fearing.” (MA thesis University of Florida, Gainesville, 1962), 34. 12. Finnegan, Ruth H. Oral Poetry: Its Nature, Signifi cance and Social Context . (New York: Cambridge University Press, 1977), 90. 13. Gregory, Horace. “A Contrast in Satires.” Poetry 49.5 (1937), 284. 14. Zaturenska, Marya. The Diaries of Marya Zaturenska, 1938–1944 . Ed. Mary Beth Hinton. (Syracuse: Syracuse University Press, 2002), 27. 15. Zaturenska, The Diaries , 68. 16. Stephan, Ruth. “Fearing and the Art of Communication.” Poetry 63.3 (1943), 163, 165. 17. Gregory, Horace, and Marya Zaturenska. A History of American Poetry, 1900–1940 . (New York: Harcourt, 1946), 467. 18. Nathan, Leonard. “Three American Poets.” Poetry 90.5 (1957), 327. The bemoaning stance of this reviewer towards Fearing’s apparent redundancy emphasizes the danger of repetition. Later reviews that Fearing received lament what they consider to be the fact that he is no longer demonstrating any revolutionary messages through his poetry. Debra Fried, in her article on repetition and epitaphs, extends an observation of Socrates’ in order to demonstrate how “writing is silent, and writing is repetitious; writing refuses to speak, and yet keeps on saying the same thing: refusal to say anything different is tantamount to a refusal to speak. Repetition thus becomes a form of silence.” See: “Repetition, Refrain, and Epitaph.” ELH 53.3 (1986), 620. 19. Kees, Weldon. “Fearing’s Collected Poems.” Poetry 57.4 (1941), 269–70. 20. Kahn, Sy. “Kenneth Fearing and the Twentieth Century Blues.” The Thirties: Fiction, Poetry, Drama . Ed. Warren French (Deland: Edwards, 1967), 139. 21. Wald, Alan M. Exiles from a Future Time: The Forging of the Mid-Twentieth- Century Literary Left . (Chapel Hill: University of North Carolina Press, 2002), 321. 22. As Rita Barnard observes in her critical work on Fearing and Nathanael West, Fearing’s use of repetition is justifi ed. Commenting on Fearing’s 1956 “Reading, Writing, and the Rackets,” Barnard argues that the Fearing of 1956 would jokingly suggest “that ‘only singing commercials,’ the ‘authentic classics of our age,’ merit repetition.” See: The Great Depression and the Culture of Abundance: Kenneth Fearing, Nathanael West, and Mass Culture in the 1930s . (Cambridge: Cambridge University Press, 1995), 98. If Fearing’s tendency for repetition stems from the production culture he observed in New York City, it is not surprising that 182 W.E. ARNOLD

it became incorporated into his poetry to some degree, as the newspapers, magazines, and commercials all play part in the “mechanical repetition of the same culture product.” See Horkheimer and Adorno. Dialectic of Enlightenment . Ed. Gunzelin Schmid Noerr. Trans. Edmund Jephcott (Stanford: Stanford University Press, [1944], 2002), 133. 23. Halliday, Mark. “Damned Good Poet: Kenneth Fearing.” Michigan Quarterly Review 40.2 (2001), 385. 24. Milner, “The Early,” 27. 25. Novak, Estelle Gershgoren. “The ‘Dynamo’ School of Poets.” Contemporary Literature 11.4 (1970), 534. 26. Rosenthal, M. L. “Kenneth Fearing.” N.d. TS. Kenneth Fearing Collection. Department of Special Collections, Memorial Library, University of Wisconsin, Madison, WI. Aug. 3, 2012. By courtesy of the Department of Special Collections, Memorial Library, University of Wisconsin-Madison. 27. Hoover, David L. “Corpus Stylistics, Stylometry, and the Styles of Henry James.” Style 41.2 (2007), 176. 28. Luyckx, Kim, and Walter Daelemans. “The Effect of Author Set Size and Data Size in Authorship Attribution.” Literary and Linguistic Computing 26.1 (2011), 45. 29. Baron, Alistair, Paul Rayson, and Dawn Archer. “Word Frequency and Key Word Statistics in Corpus Linguistics.” Anglistik 20.1 (2009). 30. Pennebaker, James W., and Shannon Wiltsey Stirman. “Word Use in the Poetry of Suicidal and Nonsuicidal Poets.” Psychosomatic Medicine 63.4 (2001), 517. 31. Pennebaker, James W., and Laura A. King. “Linguistic Styles: Language Use as an Individual Difference.” Journal of Personality and Social Psychology 77.6 (1999). 32. Benjamin, Walter. Illuminations . Ed. Hannah Arendt. Trans. Harry Zohn. (New York: Schocken, 1968), 167. 33. Mills, Nathaniel. “The Dialectic of Electricity: Kenneth Fearing, Walter Benjamin, and a Marxist Aesthetic.” Journal of Modern Literature 30.2 (2007), 21. 34. Brunyé, Tad T. et al. “Better You Than I: Perspectives and Emotion Simulation During Narrative Comprehension.” Journal of Cognitive Psychology 23.5 (2011), 660. Brunyé et al. cite the following articles: Brunyé et al. 2009; Ruby and Decety 2001. 35. Pennebaker, James W. The Secret Life of Pronouns: What Our Words Say About Us (New York: Bloomsbury, 2011), 107. 36. It is tempting to look at Fearing’s personal relationship with Margery Latimer during this period as a possible focus on female characters. Other considerations do apply, as Kaufman and Sexton observe, “professional writers typically are not writing their deepest thoughts and feelings about STYLISTIC PERSPECTIVE ACROSS KENNETH FEARING’S POETRY... 183

the most traumatic experience of their life.” James C. Kaufman and Janel D. Sexton, “Why Doesn’t the Writing Cure Help Poets?” Review of General Psychology 10.3 (2006): 276. 37. “Letter from Robert Ryley to Carl Rakosi.” 26 May 1984. Carl Rakosi Collection. Department of Special Collections, Memorial Library, University of Wisconsin, Madison, WI. Aug. 3, 2012. By courtesy of the Department of Special Collections, Memorial Library, University of Wisconsin-Madison. This “dry” period, Ryley presumes, is what led Albert Halper, with his fi ctitious representation of Fearing in Union Square (1933), to refer to Fearing as the “ex-poet and ex-communist, pot-boiler writer for the cheap sex-story magazines and former student of world affairs.” See Union Square (New York: Viking, 1933). 38. Ryley, Robert M. “Introduction.” Kenneth Fearing Complete Poems (Orono, ME: The National Poetry Foundation, 1994), xxix. 39. While it is unclear whether Williams ever met Fearing in person, in a personal letter dated 1949, he wrote that “Kenneth Fearing’s composition is for me just another revelation of his greatness. He is one of my greatest admirations.” The Selected Letters of William Carlos Williams . Ed. John C. Thirlwall (New York: New Directions, 1957), 278. 40. Simpson, Paul. Stylistics: A Resource Book for Students (London: Routledge), 59. 41. Fearing, Kenneth. Kenneth Fearing: Complete Poems . Ed. Robert M. Ryley (Orono, ME: National Poetry Foundation, 1994), 262. 42. Fearing, Kenneth Fearing , 111. 43. There are some noticeable stylistic changes between Angel Arms and Poems . Cameron Bardrick specifi cally notes, “the poetic style of the collection Poems is characterized by an increasingly expanded prosaic line, more complex punctuation including the frequent use of semi-colons, less emphasis on the individual line, and more reliance on indentation and the typographical spacing of stanzas that come to resemble prose paragraphs.” See: “Social Protest and Poetic Decorum in the Great Depression: A Reading of Kenneth Fearing, Horace Gregory, and Muriel Rukeyser.” Diss. Columbia University, New York, 1993. University Microfi lms. 44. Pennebaker, The Secret , 295. 45. In December, 1939, while discussing with Donald Klopfer, at Random House, the logistics of his next collection of poetry (destined to become the Collected Poems anthology), Fearing, having inquired concerning the sales of The Hospital , was probably pleased to know his fi rst novel had thus far sold over 6,000 copies. “Letter from Donald Klopfer to Kenneth Fearing,” December 18, 1939. Random House Collection, Rare Book and Manuscript Library, Columbia University in the City of New York. 46. Ryley, “Introduction,” xv. 184 W.E. ARNOLD

47. Ryley, Robery M. “The Bleak Hour, the Polished Gem: Finding Fearing in Fearing’s Poems.” N.d. Draft chapter of Kenneth Fearing biography, in possession of the author. By courtesy of Alison Ryley. Ryley goes on to provide details concerning Fearing’s composition of the poems, not the least of which was an impending deadline, driving Fearing to compose quickly a bulk of the poems for Dead Reckoning within a three-month period. 48. Wald, American Night , 36. 49. Gregory, “A Contrast,” 284. CHAPTER 8

In the End Was the Word: A Computational Approach to T. S. Eliot’s Poetic Diction

Adam James Bradley

Five words in T. S. Eliot’s poetic works are cited by the Oxford English Dictionary 1 as fi rst documented usages: polyphiloprogenitive, inoperancy, juvescence, laquearia and piaculative . Although the neologism is a small mechanism in the workings of poetic diction, each of these words lends a perspective on the innovations of twentieth-century verse. The word “polyphiloprogenitive,” capitalized and lineated on its own, is laid out as an announcement at the start of “Mr. Eliot’s Sunday Morning Service” (1920):

POLYPHILOPROGENTIVE The sapient sutlers of the Lord Drift across the window-panes. In the beginning was the word.

The allusion to John 1:1 is appropriate to the setting of the poem, but lineated in this way, it also asks us to contemplate the poem’s own strange opening. Polyphiloprogentive is both a sign of its own defi nition (as all

I am indebted to David-Antoine Williams, il miglior fabbro , for his always thoughtful guidance and continuing mentorship, especially throughout the duration of this project.

A. J. Bradley () University of Waterloo, Waterloo, ON , Canada email: [email protected]

© The Editor(s) (if applicable) and The Author(s) 2016 185 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_8 186 A.J. BRADLEY words are) and also the orthographic embodiment of its own history, a semantic Möbius strip which periodically returns to its own beginning, and to the question of beginnings. Buried in its etymological formation is one of the main concerns of Eliot’s theory of diction—tradition. In “The Three Provincialities,” he writes:

Such knowledge facilitates his task of giving to the word a new life and to the language a new idiom. The essential of tradition is in this; in getting as much as possible of the whole weight of the history of the language behind his word. 2

The idiom he speaks of is rooted in the giving of “new life” to old words. In the case of polyphiloprogenitive, we can see this in action. Eliot’s prefi x denotes the senses “many” and “much,” and can be understood to mean “a lot of.” This prefi x is joined to the word “philoprogenitive” (of or relating to love or care of offspring; showing love for one’s offspring), its usage fi rst recorded in the London Times in 1842. Showing love for one’s offspring in the sense of Eliot’s “new life,” the word evokes the role of the artist as creator, but, when paired with the biblical allusion that concludes the stanza, the word creates a tension between the idea of an all-knowing god and the power of the poet. The word, found originally in the Wycliffe Bible c. 1384, is formed from “progenetive,” which was fi rst recorded in 1769. It roots from progenitor , “A person from whom another is descended; an ancestor, a forefather.” As we unpack the word “polyphiloprogenitive,” it becomes apparent that its genesis and its use are both indicative of the tradition Eliot espouses—the tradition that artistry is held in those words that our words are descended from. To press the investigation further, the word “progenetive,” itself a compound, is formed from the classical Latin pro ̄genit- , past participial stem of prōgignere (to beget, bear, bring forth). The word “polyphiloprogenitive” contains within itself an entire pathway, a set of stepping-stones through the language and a bringing forth of the tradition from whence it came. The word that opens up “Mr. Eliot’s Sunday Morning Service” becomes itself a sermon on the history and power of words that form the tradition within English verse. Piaculative , also from “Mr. Eliot’s Sunday Morning Service,” is constructed from the English piaculum, (1. A sin, crime, offence; 2. An expiatory offering; a sacrifi ce), which roots from the Latin pia ̄culum (expiatory offering, act of atonement, action which calls for expiation, or sin). The two existing English senses of the word are marked rare and obs . IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 187

(obsolete) in the OED , and Eliot endeavors to reinvigorate the etymon by ignoring the existing adjective piacular (making or requiring expiation or atonement, in use from the early seventeenth century) in favor of the never-used piaculative . The suffi x -ative stands in for the idea of something or someone possessing a particular attribute, and Eliot uses it to charge a handful of pennies with metaphoric weight:

The Sable presbyters approach The avenue of penitence; The young are red and pustular Clutching piaculative pence.

The act of the young “[c]lutching” these coins suggests both childish timidity and a reluctance to walk the “avenue of penitence” or, perhaps more appropriately, buy their atonement. The contrast of the red and pustular young with the church elders dressed in sable black, a heraldic symbol of mourning, the children themselves are both mourning their sin and the money that will atone for that sin. The red faces are both a sign of embarrassment and also a representation that refers back to the youthful pustules of adolescence. In this way Eliot imbues the image of the young with an innocence that holds itself in a tensive force of these two lines, themselves being clutched to the previous two lines by the semi-colon. Considering that all of this is held in opposition to the established age and ability of the church elders, a simple shift to the adjectival suffi x -ative allows Eliot to create these dichotomies between the old, the young and the economic power of sin. Note that the children themselves are not piaculative, but it is the “pence,” giving function to the previously inanimate object. Although using the form piacular would enact the same function, it falls one beat short of piaculative and would disrupt the eight beat lines of the stanza. As is the case for useful poetic neologisms, Eliot conjures “piaculative” out of both a formal and metaphoric necessity. His allusion both calls to tradition and begins his own tradition anew.

T.S. ELIOT AND TRADITION In no small part due to the infl uence of Eliot’s early critical prose, the question of tradition pervades the discourse of twentieth-century poetics. When Eliot coins new words, he is enveloped in the history of the language; in the cases of “polyphiloprogenitive” and “piaculative,” he has created words that embody the very tradition that he pursues. It is 188 A.J. BRADLEY this focus on a particular tradition that sets Eliot apart from his contemporaries, especially those poets labeled as “Georgians.” 3 The modernists, presenting themselves as experimentalists, portrayed the Georgians as dab- bling in the leftovers of Romantic sensitivity and Victorian mores. Robert Graves, represented by eighteen poems in the set of anthologies known as “Georgian poetry” published by Harold Monroe out of his Poetry bookshop in London and edited by Edward Marsh, spoke of “the foul tidal basin of modernism”; 4 as Cuda and Schuchard write in their introduction to the Collected Prose , vol. 2, Eliot was of the view that “[t]he rustic, pastoral themes of Wordsworth had become, by the early 20th century, the parochial, self-satisfi ed verse of the Georgian poets.” 5 Harriet Monroe echoes this critique of stale Georgian verse in her review in Poetry :

The “Georgians” live in the 20th century, no doubt, but their subjects, ide- als and methods follow the old standards of English song … almost nothing in the book reminds us of the age we live in. 6

While the Georgians are criticized for their attachment to the recent generations of English poets, modernists are praised for their return to traditions much older. It appears as though there are two types of traditions at play here: those of the immediate predecessors, who for Eliot have become stale and unproductive in their use of idiom; and, those poets of what we can call the long tradition, whose words have the potential to be reinvigorated within modern verse. But, there were also those trying to rid themselves of tradition altogether. At the Poetry Banquet held in his honor in Chicago on March 1, 1914, William Butler Yeats addressed the room about the Rhymers' Club and their attempts to rid themselves of Victorian rhetoric and diction:

We were weary of all this. We wanted to get rid not only of rhetoric but of poetic diction. We tried to strip away everything that was artifi cial, to get a style like speech, as simple as the simplest prose, like a cry of the heart. 7

In these lines we hear the echo of William Wordsworth’s call for “a selection of the language really spoken by men,” made in his Preface to Lyrical Ballads . 8 Eliot, too, speaks of his diction in connection with Wordsworth’s oft-quoted lines:

I myself can remember a time when some question of “poetic diction” was in the air; when Ezra Pound issued his statement that “poetry ought to IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 189

be as well written as prose” 9 ; and when he and I and our colleagues were mentioned by a writer in The Morning Post as “literary bolsheviks” and by Mr. Arthur Waugh (with a point that has always escaped me) as “drunken helots”. But I think that we believed that we were affi rming forgotten standards, rather than setting up new idols. Wordsworth, when he said that his purpose was “to imitate, and as far as possible, to adopt, the very language of men”, was only saying in other words what Dryden had said”. 10 11

While Eliot sought to write within the bounds of a new poetic diction, the New Critics, working at roughly the same time, reasoned that poetry was a subset of language itself; they were fi xated on the idea that verse was a function of its own medium. For I. A. Richards, this differentiating attribute lay in ambiguity, a fi gure his student William Empson would most fully elaborate and taxonomize. 12 As Elder Olson writes in Modern Philology , “[a]ppar- ently he [Empson] reasons that, since poetry is language highly charged with meaning, the poetic word must invariably stagger under the full weight of its dictionary signifi cances.” Olson points out that “the instrument by which he detects the possible meanings of words is the Oxford English dictionary; although it is seldom mentioned by name, its presence everywhere is neither invisible nor subtle.” 13 Empson would have used the 1933 OED 1, which is wholly contained in the OED 2. The analysis that follows originates from within the tradition of Richards and Empson, although now with the help of the machine, modernist literary critics have the ability to access the “staggering weight” of words in ways previously impossible.

NEW CRITICISM AND TECHNOLOGY Olson’s objections to Empson’s methods provide an important caution to those who would augment the New Criticism with the help of technology. Olson charges Empson with employing a “mechanical method…capable of all the mindless brutality of a machine," 14 a critique that will be familiar to many engaged in digital humanities enquiries. Olson writes:

The theories of Richards and Empson illustrate a tendency, very prevalent among critics who rate diction as important, to rate it as entirely too important. In the order of our coming to know the poem, it is true, the words are all-important; without them we could not know the poem. But when we grasp the structure we see that in the poetic order they are the least important element; they are governed by everything else in the poem. 15

It is our job to re-evaluate the methods of the past within the new medium, to test and reshape past theories and interpretations, to participate and 190 A.J. BRADLEY try to better understand the tradition. Were diction rated too highly, we would not fi nd anything of consequence in an investigation of it. Olson’s caveats notwithstanding, the question of diction has nonetheless been omnipresent in the modernist discourse of poetics. Graves writes that “powerful and restrained language” meant “nouns and verbs outnumbering the adjectives.” 16 Ezra Pound, in his Imagist essay/manifesto A Few Don’ts by an Imagiste (1913) rallied against using superfl uous words, especially adjectives. Eliot himself conjectured about adjectives in poems, including in “Verse Pleasant and Unpleasant,” his review (1918) of Marsh’s Georgian Poetry :

“Fountains,” with the exception of a few adjectives, is a success. “Promenades” is not tight enough …. [T]he “rich” is superfl uous. “Prospect Road” tends to dissolve into its constituent adjectives and substantives, and “gigantic” should not be followed by “immense” in the next line … . “London” shows Mr. Sitwell in risk of becoming descriptive. 17

For decades after the Modernist project had withered, criticism saw Eliot and Pound as the exemplars of modernity. Eliot took it upon himself to defi ne what was “modern,” and the critics were apt to follow. C. K. Stead writes that critics have been disposed to view the Georgians “through spectacles provided … by the later, more vigorous movement led by Pound and Eliot.” 18 But Georgian and Modernist motivations tended to be quite similar at the outset. Myron Simon writes that

[b]oth Georgian and Imagist recoiled from Victorian doctrinizing, from turgid and ornate poetic diction, and from enervated sensualism. The popular poetry of this time was mutually unacceptable to Marsh and Pound. Thus, they found themselves briefl y in accord as to the poetic habits and mannerisms that were to be discarded. 19

Modern poetics for the Georgians “meant poetry which concentrated on the removal of archaic diction and pompous language from its text” and “valued plain language and subject matter which refl ected the mundane and pastoral in contrast to the more complex language and philosophical speculations of the modernists.” 20 According to Sir Henry Newbolt, by the turn of the new century, Robert Bridges was urging “the great need of modern poetry for a fresher diction and a broader freedom.” 21 Ultimately, then, it seems fruitful to extend the vision of Richards and Empson and investigate Eliot’s work in relation to Georgian poetics, for they were both quite concerned with freeing themselves from the shackles of Victorian IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 191 usage, or as Eliot put it, we need to “stimulate the worn nerves and release the arthritic limbs of our diction.” 22

TWENTIETH-CENTURY DICTION After my short review of tradition and poetic diction in the fi rst half of the twentieth century, I am left with questions concerning the validity, as it pertains to diction, of the claims levied against the Georgians. Why was Eliot thought of as an exemplar of the time, and how did the differences in his and the Georgian’s use of words affect our defi nitions of modernism? Due to the size of the enquiry, these questions are diffi cult to answer without augmenting our critical practices. Taking account of all the word use in large corpora can take years, but with digital texts and a computer we are able to ask questions in regards to poetic diction that were previously so labor intensive we would fi rst need to develop an entire concordance to even begin. We can discern three claims from the existing criticism that lend themselves to quantitative verifi cation. These are:

1. The Georgians express themselves in an idiom characteristic of the Romantics and Victorians (Eliot, Monroe); 2. The modernists look to older traditions to inform their verse (Eliot, Pound); 3. Poetic diction is of subsidiary importance (Olson).

The fi rst step in any enquiry such as this is to organize and pre-process one’s materials. All of the scripts for this project were written in Python and were bespoke for each stage of investigation. Digital copies of the corpora in question, which luckily do exist, were sourced and all of the headings, page numbers and line markings were removed. 23 Because Eliot and his contemporaries were so critical of the Georgians—by the 1920s they were calling them “week-end poets”—I chose Eliot, the great beacon of modernist poetry, and juxtaposed his work with the whole selection of poets taken from Marsh’s Georgian Poetry anthologies. All fi ve Georgian poetry volumes were used for comparison (1912, 1915, 1917, 1919, 1922). These compilations were fi rst published by Harold Monroe, out of the now famous Poetry Bookshop in London and edited by Edward Marsh, who in his preface to the fi rst edition wrote, “[t]his collection, drawn entirely from the publications of the past two years [1911–12], may if it is fortunate help the lovers of poetry to realize that we are at the beginning of another ‘Georgian period’ which may take rank in due time 192 A.J. BRADLEY with the several great poetic ages of the past.” 24 There is a great disparity with Marsh’s expectation and the critical reception. The expectations set by Marsh for his poets were never quite realized, and the individual fame of artists like Graves overshadowed the movement. This tension tends to be described by critics in the difference, newness and modern expression of Eliot and his cohort. The digital investigations that follow are employed to identify whether this proposed tension is rooted within the texts themselves or were simply critical posturing. For Eliot, I used a PDF version of the complete poems. Because the idea of modernity was so prevalent in the critical discourse that surrounds both of these selections, I fi rst wanted to see how different, if at all, the collective vocabulary of the two selections were. I made a concordance of each corpus, tokenizing and lemmatizing 25 each set of texts. Based on the probabilistic nature of the algorithms, I was concerned that the consistency of the lemmatizers currently in use are not effi cient enough to produce results accurate enough for literary critical conclusions. To try to negate some of this variance, I included the original words alongside their lemmatized versions in the concordance. This was a critical decision to which I lent a lot of thought. I landed on following my instinct that catching more of the vocabulary would be a more accurate representation of word use across the corpora, though either approach was going to be imperfect. I then wrote a script that would parse the metadata from the OED 2 for each word entry found in each concordance. What does tradition, as Eliot lays it out, mean in terms of diction? To try to unpack this idea of tradition, I began by taking all of the words used in the selected corpora and organizing them by the fi rst known date of usage in the OED . This way it could be tested as to whether the “tradition” has to do with the relative ages of words used. Although this date may or may not be the actual date of coinage, it is the fi rst recorded usage in print that was available to the compilers of the dictionary. I categorized the output by century because my original hypothesis was that the diction of modernity would be found in the difference between the usage of twentieth-century words that Eliot and the Georgians were writing and those of the past. At fi rst I calculated their vocabulary in terms of their individual totals to yield a percentage (Fig. 8.1 ). The similarity of the results was unexpected and forced me to reex- amine my original hypothesis. The OED 2 contains 615,100 word forms defi ned and/or illustrated, so I quickly dismissed the notion that it was a defect of the dictionary—although it should be noted that due to the IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 193

Fig. 8.1 Percentage of total words used by century of fi rst usage by Eliot and the Georgians history of our language development and the availability of documentary evidence some centuries (e.g., the twelfth century) are noticeably sparse. Was this simply a function of the English language, shared by all who use it? And, in terms of my original question in relation to twentieth-century usage, how close were Eliot and the Georgians to each other in the use of words that came into English in the twentieth century? To arrive at a comparable measure of difference, Z-scores were calculated for each century. This metric demonstrates the standard deviation from the mean of the sample. It allows for an “apples to apples” comparison of unlike data. When I compared the Georgians and Eliot, the resulting percentages were still very close (Fig. 8.2 ), although Eliot noticeably over-uses words introduced in the eleventh century, while the Georgians use sixteenth-century words more frequently. That the relative ages of vocabularies were so similar across corpora is unexpected, given the extent to which literary criticism has insisted on difference. The eleventh century, where one of the discrepancies is situated, was a watershed during which English vocabulary doubled. This epoch is the source of nearly all our most commons words and within a poetic analysis could indicate an unnatural vocabulary by its underuse. The common idiom that Eliot and Pound were searching for would certainly be found in the eleventh century, and the charges against the Georgians of being antiquated in their 194 A.J. BRADLEY

Fig. 8.2 Georgian poets’ versus T. S. Eliot’s Z-scores for total usage by century diction can be explained by their lack of use in older but more common words. To investigate further, I decided to test the Victorians against both the Georgians and Eliot, as one of the main criticisms of Georgian poetry has been that they are leftover Victorians. I therefore sourced a PDF copy of the Penguin Book of Victorian Verse and tested its vocabulary against the other two (Fig. 8.3 ). All three sets of lines are virtually identical, with again Eliot using more eleventh-century words and the Victorians and Georgians using more sixteenth-century words. What appears to be shown is that in terms of vocabulary, with very few exceptions, the Victorians, the Georgians and Eliot use the same distribution of words from each century (regardless of total usage) when writing poetry. This may simply be a function of normal vocabularies—that what we are seeing is a representation of how everyone uses the language—but it may also contain within it evidence that all poetry may tend towards this type of distribution. To fashion a control group to explore this admittedly bold speculation, I compiled the same data using only original words from the texts and added in the Brown News Corpus (Fig. 8.4 ), which I chose because it was of a recognized different genre. Compiled in 1962, it was also chronologically the closest IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 195

Fig. 8.3 Comparison with Victorian Z-scores for total usage by century

Fig. 8.4 Comparison with Brown News Corpus Z-scores 196 A.J. BRADLEY corpus I could source to the decades when Eliot and the Georgians were writing. 26 The data shows that when considering genre, a glaring difference in usage emerges, based on the ages of words. If we accept the New Critical idea that poetry is a subset of language as a whole, then this is to be expected. The difference that it does show is that the poets rely much more heavily on the words coined in the eleventh, twelfth, thirteenth and fourteenth centuries, whereas the newspaper articles use many more sixteenth-, seventeenth- and nineteenth-century words (perhaps because those words were culturally important by the time they were writing). In terms of vocabulary, two points are clear. The Victorians, Georgians and Eliot (with few exceptions) all use a set of vocabulary that is virtually identical across the history of our language, and when this set of poets are balanced with a corpus from a separate genre, it becomes apparent how similar their vocabulary distributions really are. These distributions are not affected by the omission of lemmatized versions of these words. From these results, we are left with one outlier: the underrepresentation of sixteenth-century words in Eliot’s poems. To investigate this question, I employed a similar methodology as before, only instead of generating a concordance of words for each corpus, I simply accounted for all words. 27 The totals words used were as follows: Victorians: 152,092; Georgians: 134,378; Eliot: 30,648; Brown Corpus: 89,658. I then calculated the Z-scores for each corpus by the relative ages of the words used (Fig. 8.5 ). What is interesting here are that the discrepancies between the poets virtually disappears, meaning that Eliot uses a greater distribution of individual words from the eleventh century than the Georgians, but they use theirs more frequently. This suggests then that the difference between the two samples is situated in their vocabulary and not their usage. Many of the most common words English language speakers use are early remnants from the language, words such as “an,” “and,” “as,” “at,” “be,” “but,” “for,” “from,” “had,” “have,” “he,” “her,” “his,” “I,” “in,” “it,” “of,” “on,” “that,” “the,” “this,” “we,” “were,” “that” and “who,” which are all from the eleventh century and can account for the dispro- portionate spike. One possible explanation involves the average number of denotative meanings per word. The higher percentage use of eleventh- century words could explain this discrepancy based on the fact that many of those words are prepositions and conjunctions, which tend to have above average number of senses per entry in the OED . 28 This phenomenon would speak to the Imagist mandate towards direct language, which IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 197

Fig. 8.5 Full table of Z-scores for total usage by century

Henry Newbolt labeled as trying to “break through the crust of poeti- cisms which enclosed the Edwardians and Georgians.” 29 What is counter- intuitive about this is that the modernity that may be enclosed in their vocabulary and usage is not simply using new words: it is the reinvigora- tion of old words, to which Eliot’s greater use of eleventh-century ver- biage attests. Admittedly, corpora-wide analyses are quite limited in their interpretive possibilities, but it is within those limits that the potential to witness small phenomena of language arises. My original question about the Georgians and Eliot (whether “modern” language is defi ned by newer words) is still unresolved. This is where the use of machines must coincide with the use of critical inquiry. I originally wanted to know about twentieth-century words, and because those subsets are so small, I generated a word list for Eliot and the Georgians usage from that time and went to the OED by hand—that is, I consulted each relevant OED entry—to investigate further. I found that the Georgians used fi fty-eight words coined in the twentieth century, and Eliot used thirty-six. Words such as “weatherwise,” “Quebec,” “quiring,” “rattled,” “sputtered,” “sweeney,” “tiens,” “waldo” and “waves”—which all have twentieth-century variants in usage, meaning that these words were appropriated into slang or used in new ways 198 A.J. BRADLEY without the form of the word changing at all—were removed by hand from the datasets. 30 This left seven words used by the Georgians coined in the twentieth century: “Bloomsbury,” “gah,” “rhodes,” “rilled”, 31 “sac- ramented,” “verboten,” “vespering.” After applying the same process to Eliot, twelve words coined in the twentieth century remained that were not reinvigorated variants of older words, fi ve 32 of which were his own coinage. These words include “gotta,” “grimpen,” “inoperancy,” “juvescence,” “laquearia,” “ltd,” “piaculative,” “polyphiloprogenitive,” “rolls- royce,” “citroen” and “taxi.” The usages of twentieth-century words for Eliot and the Georgians are identical in relation to their overall diction, further confi rmation that Eliot’s modernity was not found in recently coined words. Instead, it is in the underrepresentation of sixteenth-century words that we can start to review the idea of modern vocabulary. Accordingly, in “Tradition and the Individual Talent,” Eliot begins discussing the realign- ment of English poetic tradition when he says that modern poets “must be very conscious of the main current, which does not at all fl ow invariably through the most distinguished reputations.” 33 This line of enquiry extends to Eliot’s Essay “The Metaphysical Poets,” in which he writes of metaphysical poets, “the meaning is clear, the language simple and elegant. It is to be observed that the language of these poets is as a rule simple and pure; in the verse of George Herbert this simplicity is carried as far as it can go—a simplicity emulated without success by numerous modern poets.” 34 This emulation, when approached from a whole-corpus perspective, shows a strong connection between Eliot and Herbert. Their usage appears to be in line with the poetic vocabulary generally used in English, but for Herbert and Eliot, there is a dramatic shift found in the absence of sixteenth-century usage. Eliot writes, “the language went on and in some respects improved; the best verse of Collins, Gray, Johnson, and even Goldsmith satisfi es some of our fastidious demands better than that of Donne or Marvell or King. But while the language became more refi ned, the feeling became more crude.” 35 And what of Donne? Eliot on many occasions championed the verse of John Donne, writing that “[a] poet like Donne, or like Baudelaire or Laforgue, may almost be considered the inventor of an attitude, a system of feeling or of morals.” 36 I procured both Herbert and Donne’s complete poetry and ran the same vocabulary and usage scripts on them, and what resulted was quite extraordinary. Not only did Eliot speak of realigning the poetic tradition, but intentional or not: he also emulated the sixteenth-century vocabulary usage of Herbert IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 199 and Donne. It can be seen (Fig. 8.6 ) that Eliot and Donne use more eleventh-century words, and those two (with the addition of Herbert) fall below the Victorians and the Georgians in sixteenth-century usage. One might expect this result, given the lag of centuries; there is less time for words to catch on. For Eliot, it appears that one way of approaching modernity was actually by returning to the past and realigning the vocabulary with the tradition that he saw as “modern." This modernity ran through Marlowe, Ben Johnson, Herbert, Donne, Marvell and Dryden. By approaching questions of modernity through a whole-corpus approach, trends otherwise only intuited can be shown to have real effects in comparison. Eliot wrote, “my business is, I believe, to endeavor to determine what is meant by ‘modern’ poetry, and to trace, among the variety of currents and eddies, what is the line of true poetry, as distinguished from mere novelties.” 37 For Eliot, this process was partially enacted by limiting his own vocabulary and aligning himself with the English tradition he saw as the foundation of being modern. The difference between the Georgians and the Modernists may be one of subtlety. Their usage and vocabulary are, with respect to the qualities measured here, identical in other than the eleventh and sixteenth centuries. Eliot made claims that he was realigning English poetics many times, and empirically, when it comes

Fig. 8.6 Z-score vocabulary across centuries 200 A.J. BRADLEY to diction, it appears that this is the case. But these inquiries are only the beginning; much work is needed to decipher how, and in what ways (if any), this vocabulary played out in the actual lines of extant poetry. The machine can aid us in generating new hypotheses or confi rming previously intuited lines of reasoning, but it cannot do the work of literary criticism. We must heed Olson’s warnings of becoming too machine-like but understand that the processing of texts and the interpretation of outputs are two separate parts of a larger process. When it comes to the pre-processing and processing of texts to investigate hypotheses, being machine-like is an asset. But, for the literary critical work, we must employ an augmented criticism, one rooted in humanistic enquiries that maintain humanistic goals. Just as Eliot tried to do in his poetry, we must turn our sight back on the tradition of criticism while carving out our own.

NOTES 1. For this project a combination of data was used from the database of the OED 2 and the Online OED 3. 2. T. S. Eliot, The Three Provincialities . The Complete Prose of T.S. Eliot; The Critical Edition, vol. 2. 2 (Baltimore: Johns Hopkins University Press, 2014), 392. 3. This is the name that has come to describe the litany of poets that were included in the set of anthologies known as “Georgian poetry” published by Harold Monroe out of his Poetry bookshop in London and edited by Edward Marsh. 4. Robert Graves, Poetic Craft and Principle. (Faraday Close: Littlehampton Book Services Ltd., 1967), 14. 5. Anthony Cuda and Ronald Schuchard, eds. The Complete Prose of T. S. Eliot: The Critical Edition , vol. 2 (Baltimore: Johns Hopkins University Press, 2014), xxii. 6. Harriet Monroe, “King George’s Poets ,” Poetry , 16 (May 1920). pp. 108–9. 7. Jewel Spears Brooker, ed., The Complete Prose of T. S. Eliot: The Critical Edition , vol. 1 (Baltimore: Johns Hopkins University Press, 2014), 611. 8. Samuel Taylor Coleridge and William Wordsworth., Lyrical Ballads 1798 and 1800 , eds. Michael Gamer & Dahlia Porter (New York: Broadview, 2008). 9. This is from Pound’s essay A Few Don’ts of an Imagiste . 10. Eliot is referring here to Dryden’s intro to his didactic poem Religio Laici, in which he writes: “the expressions of a poem designed purely for instruction ought to be plain and natural, yet majestic.” IN THE END WAS THE WORD: A COMPUTATIONAL APPROACH TO T. S. ELIOT’S... 201

11. T. S. Eliot, The Use of Poetry and the Use of Criticism: Studies in the Relation of Criticism to Poetry in England . (London: Faber, 1933), 71. 12. William Empson, “Seven Types of Ambiguity.” 13. Elder Olson, William Empson: Contemporary Criticism and Poetic Diction, Modern Philology, 47:4 (May 1950), pp. 222–52. (Chicago: The University of Chicago Press), 224. 14. Ibid., 225. 15. Ibid., 230. 16. Robert Graves, Poetic Craft and Principle (Faraday Close: Littlehampton Book Services Ltd., 1967). 17. T. S. Eliot, Verse Pleasant and Unpleasant . The Complete Prose of T. S. Eliot: The Critical Edition , vol. 1 (Baltimore: Johns Hopkins University Press, 2014), 681. 18. C. K. Stead, The New Poetics (New York: Ams Pr Inc., 1979), 81. 19. Myrmon Symon, The Georgian Poetic. The Bulletin of the Midwest Modern Language Association,Vol. 2, Papers of the Midwest Modern Language Association, Number 1. Poetic Theory/Poetic Practice (1969), 121. 20. Patrick J. M. Quinn, ed. Mark Dady. Reader’s Guide to Literature in English . “Georgian Poetry” (London: Fitzroy Dearborn Publishers, 1996), 322. 21. Sir Henry John Newbolt, My World as in My Time (London: Faber and Faber,1932), 194. 22. T. S. Eliot, Prose and Verse . The Complete Prose of T. S. Eliot; The Critical Edition Vol. 2., Ed. Anthony Cuda & Ronald Schuchard (Baltimore: Johns Hopkins University Press, 2014), 324. 23. A saving grace in any of these efforts is to fi nd professionally published PDF versions of fi les. OCR is a poor choice for scanned books, but published PDFs have easily recognizable text that can be manipulated algorithmically with almost no trouble. 24. Edward Marsh, ed. Georgian Poetry 1911–12. (London: Poetry Bookshop, 1912). 25. I used the porter2 stemmer. Not perfect but adequate and I have had success using it. Any time that stemming is being used we must realize that the process is probabilistic and hand editing of the results may be necessary. 26. Finding reliable corpora is simply part of the task of this type of approach. Within the digital humanities digitizing texts is a constant project and often the choices available are not ideal. Problems arise with proprietary data concerns and often even if a corpora exists it may be inaccessible. The decision made here was based on availability and consistency. 202 A.J. BRADLEY

27. Accounting for all words counts total usage. If the stop-word “it” was used 100 times, the concordance method accounts for the words existence once and accounting for all words counts all instances. The former gives us a metric related to vocabulary and the latter produces a measure of usage. 28. The word “that” has the most number of senses and sub-senses in the OED 2, 512–83 of which are marked obsolete. 29. Sir Henty John Newbolt, My World as in My Time (London: Faber and Faber,1932), 194. 30. The decision to remove these words was based on the fact that to include them was to analyze every single usage in all corpora, which is simply not within the scope of these initial investigations. 31. Rilled, actually coined in 1899 according to OED 2, is included because any usage after coinage would technically fall in the twentieth century. 32. inoperancy, juvescence, laqueria, piaculative, polyphiloprogenitive. 33. T. S. Eliot, Tradition and the Individual Talent . The Complete Prose of T. S. Eliot; The Critical Edition Vol. 2, ed. Anthony Cuda & Ronald Schuchard. (Baltimore: Johns Hopkins University Press, 2014), 107. 34. T. S. Eliot, The Metaphysical Poets . The Complete Prose of T. S. Eliot; The Critical Edition Vol. 2, ed. Anthony Cuda & Ronald Schuchard (Baltimore: Johns Hopkins University Press, 2014), 378. 35. Ibid., 381. 36. T. S. Eliot, Andrew Marvell . The Complete Prose of T. S. Eliot; The Critical Edition Vol. 2, ed. Anthony Cuda & Ronald Schuchard. (Baltimore: Johns Hopkins University Press, 2014.), 309. 37. T. S. Eliot, Modern Tendencies in Poetry . The Complete Prose of T. S. Eliot; The Critical Edition Vol. 2, ed. Anthony Cuda & Ronald Schuchard (Baltimore: Johns Hopkins University Press, 2014), 212. CHAPTER 9

A Macro-Etymological Analysis of James Joyce’s A Portrait of the Artist as a Young Man

Jonathan Reeve

The English language is a palimpsest, bearing traces of the languages it has contacted. French, Latin, Ancient Greek and Irish are among the languages that have contributed words to English, and these ancestor languages comprise modes of expression whose every word recalls the contexts of their acquisition. When a writer chooses the word “chew” over “masticate,” or “enchantment” over “spell,” what does that decision indicate? How can we measure these stylistic vectors? This study uses a computational analysis of the etymologies of words in James Joyce’s novel A Portrait of the Artist as a Young Man in order to identify etymological registers, generic resonances and levels of formality. In particular, this study will attempt to measure the maturing language of this novel’s protagonist Stephen Dedalus through his use of Latinate words, and to identify ways in which this novel’s macro-etymological signals refl ect its structural elements. The works of James Joyce are ideal for macro-etymological analysis. Joyce was famously multilingual, and many see his novels as a crescendo

J. Reeve () Columbia University, New York, NY, USA e-mail: [email protected]

© The Editor(s) (if applicable) and The Author(s) 2016 203 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_9 204 J. REEVE of linguistic experiments. In the words of Laurent Milesi, “Joyce’s oeuvre is best seen as constantly trying to inform an evolutive linguistic poetics.” 1 Finnegans Wake , the culmination of his career in literary experimentation, is arguably unparalleled in its paranomasia and polysemy; in its composition, Joyce employed word roots from forty languages. 2 But this impulse was present in Joyce’s early works as well; his words are deliberately chosen to suggest their ancestors and cognates. They are serio-comic puns made to extend along etymological axes to new meanings in other languages. Joyce himself was keenly interested in etymology. In his early critical essay “The Study of Languages,” he argues that “in the history of words there is much that indicates the history of men, and in comparing the speech of to-day with that of years ago, we have a useful illustration of external infl uences on the very words of a race.” 3 Joyce’s interest in etymology was that of the application of word history to English usage. He argues for the study of Latin, of which “a careful and well-directed study must be very advantageous,” because it “acquaints us with a language, which has a strong element in English, and thus makes us know the derivations of many words, which we then apply more correctly and which have therefore a truer meaning for us” (16). This itself may be an etymological pun, since the word etymology is derived from the Greek etumon for “true.” It follows that, by studying the etymologies of Joyce’s words, we might discover more of the diversity of what Joyce considered to be “truer.” In Stephen Hero , an early version of Portrait , Stephen Dedalus is described as having “read Skeat’s Etymological Dictionary by the hour.” 4 Although purely autobiographical readings of the two novels are problematic, we may safely assume that Joyce had also read and loved this work. In an etymological reading of the Dubliners story “Ivy Day in the Committee Room,” Michael Brian argues that Joyce had a such a “detailed and profound knowledge” of Skeat’s dictionary, and that it had such an infl uence on this story that “one could say [it] is written in Skeatish.” 5 Stephen Whittaker takes it as so obvious that Joyce was intimately familiar with Skeat that he is more interested in the question of whether Joyce worked from the third or fourth edition of the dictionary. 6 In Portrait , Stephen routinely muses about words, considering their sounds, shapes and beauty. “Suck,” Stephen considers “a queer word,” 7 but “wine” he thinks “a beautiful word” (39). Seeing the word “fœtus” carved into a desk “startle[s] his blood” (75) but upon hearing Cranly say “ mulier cantat ,” he remarks on the “soft beauty of the Latin word” (205). It is this logophilia that justifi es, in part, the following quantitative A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 205

methodology, even at the risk of decontextualizing individual words. “One diffi culty in esthetic discussion,” Stephen seemingly cautions us, “is to know whether words are being used according to the literary tradition or according to the tradition of the marketplace” (157). This is one of the diffi culties of computational literary criticism, as well: the so-called “bag of words” model of digital text analysis cannot suffi ciently account for context. However, Joyce’s attention to words and their histories valorizes an investigation such as this. Joyce was keenly aware of not only the aesthetic, but also the political dimensions of word choice. Marjorie Howes argues that Joyce “consistently embedded the complexities of colonialism and nationalism in particular words,” and cites his use of ivory , as a spiritual metaphor (Mary is a “tower of ivory” [Portrait 29]), a proto-sexual image (Eileen’s hands were like ivory [ibid.]), and a colonial commodity. 8 Stephen daydreams about this word, and imagines it prisming: “The word now shone in his brain, clearer and brighter than any ivory sawn from the mottled tusks of elephants. Ivory, ivoire, avorio, ebur .” 9 The splitting of the word into its cognates approximately traces its etymology, from English to (Norman) French to Italian (Vulgar Latin) and Latin. In fact, each of these four forms for “ivory” are given both in Skeat’s dictionary and in the OED in precisely this order, although this is not their direct lineage. This word history, therefore, traces a path that locates language among nations, and fi nds “the history of men” in “the history of words.” 10 This vector is also Joyce’s own biographical path of exile, from Ireland to France and fi nally to Rome, where he fi rst began to rework Stephen Hero into Portrait . The narrative style of Portrait is another of its properties that makes it appropriate for macroanalysis. Whether called Erlebte Rede , or, in Flaubert’s term, le style indirect libre , it is style in which the boundaries between the narrator’s language and the characters are blurred. When Wyndham Lewis disparaged Joyce’s phrase “Uncle Charles repaired to the outhouse,” complaining that “people repair to places in works of fi ction of the humblest order,” Hugh Kenner responded by explaining that “‘repaired’ wears invisible quotation marks. It would be Uncle Charles’s own word should he chance to say what he was doing.” 11 Kenner thus dubbed this Joycean narrative technique the “The Uncle Charles Principle,” which he defi nes by explaining that “[Joyce’s] words are in such delicate equilibrium, like the components of a sensitive piece of apparatus, that they detect the gravitational fi eld of the nearest person” (16). For Kenner, this style is primarily observed on the level of the individual word. “[Joyce] is not,” 206 J. REEVE he writes, “like Beckett, an Eiffel nor a Calder of the sentence. The single word—‘repaired’; ‘salubrious’—is his normal means to his characteristic effects” (20). This might be because, as Joyce was aware, the histories of each word made them richly polysemous. This property of Portrait is one that makes macroanalysis meaningful: the histories of the individual words aren’t simply functional aspects of the language, but crucial stylistic and ontological units saturated with traces of their origins. Since Uncle Charles himself makes only a brief appearance in Portrait , a more signifi cant effect of the Uncle Charles Principle may be observed in the language of Stephen Dedalus, whether expressed directly or through the narrator. Stephen’s language, and therefore largely the language of the novel as a whole, begins with juvenile songs and ends with mature prose. The following experiment is designed to quantify that development, by analyzing each of the chapters of the novel individually. The initial hypothesis is that the Macro-Etymological Analyzer will show an increase in proportions of words of Latinate origin throughout the course of the novel. This hypothesis is confi rmed, but not without surprises.

THE EXPERIMENT The Macro-Etymological Analyzer is a web app 12 written using a LAMP stack—Linux, Apache, MySQL and PHP. It ingests a text, tokenizes it, and looks up each word in the Etymological Wordnet, a relational database created from Wiktionary data by the computer scientist Gerard de Melo. The program fi nds the fi rst language ancestor of each word and categorizes it according to language family. Since words of French origin and words of Latin origin often share roots—many English words come from Latin through French or Anglo-Norman—these are grouped together into the category “Latinate,” along with words of Italian or Spanish origin. Words descended from Old or Middle English, German or Dutch are categorized as “Germanic;” words of ancient and modern Greek origin are denoted “Hellenic;” and words of Irish or Scottish origin are “Celtic.” The program then determines the proportions of words of each category. 13 A Portrait of the Artist as a Young Man contains 90 percent words of Germanic origin, 5 percent words of Latinate origin, and less than 0.1 percent each of Hellenic, Slavic, Iranian, Afroasiatic and Celtic. A further 4 percent of the words in the text were not found in the dictionary, many of them proper names. These data alone are not very interesting, however, since we have no control group with which to compare them. We must therefore begin by calibrating the program. A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 207

CALIBRATION To fi nd signifi cance in these etymological signals, the Macro-Etymological Analyzer was trained on genres extracted from the Brown University Standard Corpus of Present-Day American English, a much-studied linguistic corpus of approximately one million words, created in the 1950s. 14 The corpus is broken into genre categories such as “science fi ction,” “belles lettres,” “humor” and “news.” Each of these categories was extracted using the Python NLTK and analyzed. Figure 9.1 shows the occurrence of Latinate words in categories of the Brown Corpus. The genres are divided fairly cleanly between fi ction and non-fi ction, with the fi ction genres “adventure” and “romance” on the low end of the spectrum, and the non- fi ction genres “learned” and “government” on the high end. Strikingly, the genres “Lore” and “Religion,” which are arguably of ambiguous fi c- tionality, fall in the middle. “Science Fiction,” which is probably the most non-fi ctional of the fi ction genres, lies in the same quadrant, and exhibits the highest proportion of Latinate words of a fi ctional genre. Based on this calibration, we might say that high proportions of Latinate words (hereafter “L scores”) in Portrait would have a good chance of exhibiting styles similar to learned text, offi cial documents, or non-fi ction.

Brown Corpus

Proportions of Latinate Words

0 romance humor editorial lore hobbies mystery fiction adventure reviews learned religion government science fiction news belles lettres

Fig. 9.1 Latinate words in the Brown News Corpus 208 J. REEVE

Brown Corpus

Proportions of Hellenic Words 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 romance humor editorial lore hobbies mystery fiction adventure reviews learned religion government science fiction news belles lettres

Fig. 9.2 Hellenic words in the Brown News Corpus

Among proportions of Hellenic words, as shown in Fig. 9.2 , the picture is similar, but with a few key differences. Here, “religion” has a higher rank, and “government” a lower. Since Hellenic words represent such a tiny percentage of any given text, however—a total of sixty-six words for Portrait —we cannot treat measurements of this category as equally statistically signifi cant. The same is even more true for proportions of words of Celtic origin, since only a single word was detected in that category. Germanic etymologies were inversely correlated with Latinate etymologies, so these values are already roughly represented by L scores. Each of these categories deserves an in-depth discussion.

LANGUAGES

Latinate The calibration experiments performed above suggest that high proportions of Latinate words are correlated with non-fi ction and formal or authoritarian language. In part, this can be explained by the history of the introductions of Latinate words to English. Directly following the Norman Conquest of 1066, French became the language of aristocracy, and where French words entered English, it was often in this domain. A classic example is that names of animals—cow , pig and deer , for instance— are almost all of Old English inheritance, while the names of those meats at the table—beef , pork and venison —are of French. 15 The English-speaking A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 209

Chapters of A Portrait of the Artist as a Young Man

Proportions of Latinate Words 14

0 Ch. 1 Ch. 2 Ch. 3 Ch. 4 Ch. 5

Fig. 9.3 L scores for chapters of Joyce’s Portrait of the Artist as a Young Man lower classes would be more likely to be in contact with the animals themselves, while the French-speaking upper classes would be likelier to be concerned with the commodity. As previously discussed, the hypothesis for the analysis of Portrait was that there would be an increase in the L scores across chapters in the novel. Figure 9.3 shows that this hypothesis is partially confi rmed. There is a rise in the proportions of Latinate words over Chaps. 2 , 3 and 4 , which would seem to correlate with the maturation of Stephen’s thought and speech. The L score plateaus or drops in Chaps. 5 and 6 , however. How might this be interpreted? To answer this question, it is necessary to conduct a more granular analysis. Figure 9.4 shows the text divided into sections—segments of chapters that Joyce separates with rows of asterisks. 16 The L scores for these divisions exhibit much less of a simple progression from low to high. Where the climax of the chapter-based analysis seemed to be in Chap. 4 , the climax here appears to be Chap. 5 , Part 1. With the exception of Chap. 3 , the longest and only fi ve-section chapter, the highest L scores for each chapter come in the fi rst section. The fi nal sections of each chapter are among the lowest in L scores. Seen broadly, there is a pattern here suggestive of what Riquelme calls a “structural rhythm”—a repeating sawtooth shape. 17 A number of critics have noticed this cyclical structure. Sidney Bolt describes it thus: 210 J. REEVE

At the beginning of each chapter Stephen is presented as the subject of a distressing tension, which develops to a crisis leading to a resolution. At the beginning of the next chapter, however, this resolution is seen to have produced a new tension, and the process is continued in a new form. This wave-like, pulsating movement is characteristic of every scene. 18

Thomas Connolly calls this form a play between spiritual and corporeal forces. “Each [of these forces] nullifi es the other,” he argues, “and a nexus results until the aesthetic perception of the beautiful breaks the knot and kinesis yields to stasis.” 19 Diane Fortuna describes these cycles in terms of labyrinth imagery and the Dedalus myth and adds that “aside from the initial subsection of Portrait , each of the subsequent 18 divisions of the novel presents at least one image of rolling, cyclical, or circling motion.” 20 Bolt, Connolly and Fortuna’s observations, although ostensibly applicable to large textual movements (such as plot), could nonetheless be applied as approximate descriptions of the rolling, cyclical etymological trends shown in Fig. 9.4 . It is possible here that the Latinate register mirrors Portrait’s sinuosities. Another description of this phenomenon is David Hayman’s reading of this structural oscillation as one between epiphanies and anti-epiphanies. The epiphanic moment is “a lyrical and wish-fulfi lling moment during

Sections of A Portrait of the Artist as a Young Man

Proportions of Latinate Words

0 1-1 1-2 1-3 1-4 2-1 2-2 2-3 2-4 2-5 3-1 3-2 3-3 4-1 4-2 4-3 5-1 5-2 5-3 5-4

Fig. 9.4 L scores for sections of Joyce’s Portrait of the Artist as a Young Man A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 211 which the illusory is made to appear as immediate and valid;” it is “both art and event.” These moments then engender an “anti-aesthetic impulse to action.” 21 While the epiphany is a “vision” or “illusion,” it is followed by an anti-epiphany that “show[s] Stephen to be increasingly involved with the world” (174). Riquelme calls this oscillation “a stylistic double helix,” and adds that “Joyce employs the two epiphanic modes of stark realism—‘the vulgarity of speech or of gesture’—and visionary fantasy … as delimiting extremes in his character.” 22 These properties—lofty visions and earthly pragmatics—map roughly to properties of L and G registers, respectively. A closer reading of these sections might be more useful than a simple mapping of criticism to macro-etymological data, however. For that, we must make the analysis even more granular. The section with the highest L score is 4.1, which Riquelme titles “Spiritual Discipline.” Grant Redford, for one, claims that this is the climactic section of the novel. 23 Within this section, the highest L score can be found in the second quarter. Here is a single sentence excerpted from that subsection:

The imagery through which the nature and kinship of the Three Persons of the Trinity were darkly shadowed forth in the books of devotion which he read—the Father contemplating from all eternity as in a mirror His Divine Perfections and thereby begetting eternally the Eternal Son and the Holy Spirit proceeding out of Father and Son from all eternity—were easier of acceptance by his mind by reason of their august incomprehensibility than was the simple fact that God had loved his soul from all eternity, for ages before he had been born into the world, for ages before the world itself had existed. 24

This passage is verbose, fl orid and multi-syllabic; its subject matter is religious, authoritarian and deathly serious. Compare that with the passage with the lowest L score, at the end of section 1.3, “Christmas Dinner,” when the argument about Parnell becomes heated. Mr. Casey’s livid yet comic remark here neatly illustrates the Germanic register used in this section: “She stuck her ugly old face up at me when she said it and I had my mouth full of tobacco juice. I bent down to her and Phth! says I to her like that”(30). With the notable exception of “tobacco,” which is ultimately descended from an indigenous Haitian language, most of these words are monosyllabic and of Germanic origin. The rhythm here is faster, and the tone lighter. There is a certain playfulness evident in the onomatopoeia Phth! , a kind of neologism which we shall see is characteristic of Germanic Joyceanisms. 212 J. REEVE

Germanic While the proportions of words of Germanic origin are, roughly speaking, inversely proportional to those of Latinate origin, they warrant discussion. Words of Germanic origin in Portrait are frequently monosyllabic and evoke raw, unfi ltered speech that is often undecorated with euphemism or social formality. When Cantwell says, “He’d give you a toe in the rump for yourself,” Stephen thinks, “[T]hat was not a nice expression.” 25 What if Cantwell had used the French-derived synonym “derrière,” or the Latin- derived “posterior?” That might not still be a nice place to have a toe, but the expression would be more polite. For Stephen, “rump” might not be “a nice expression” because it bears the resonances of the Germanic register. Early in the novel, young Stephen overhears someone use the word “suck” and thinks, “suck was a queer word … the sound was ugly.” 26 This passage, and indeed, this “queer word” has been much discussed, most notably in Derek Attridge’s study. 27 This may also have been what H. G. Wells had in mind when he accused Joyce of having a “cloacal obsession:” “He would bring back into the general picture of life aspects which modern drainage and modern decorum have taken out of ordinary discourse and conversation. Coarse, unfamiliar words are scattered about the book unpleasantly.” 28 If we remember that Stephen compares the sound of “suck” to that of “dirty water” going down the drain, modern drainage is literally that which creates this “coarse, unfamiliar” word sound. Wells’s critique highlights the reason passages like this one were so “coarse” for readers contemporary with Joyce—the sounds and registers of their words. In fact, Portrait was rejected by early English publishers on this basis. In a reader’s report for the publishers Duckworth & Company, Edward Garnett calls the novel “too discursive, formless, unrestrained,” because “ugly things, ugly words, are too prominent.” 29 Are words like “suck” ugly because they belong to the Germanic register, and carry those associations, not only to Stephen, but to the publishers as well? Stephen’s thoughts about the word “suck” also begin a onomatopoetic theme that is chiefl y associated with words of Germanic origin. Stephen later explains the word “kiss” in onomatopoetic terms—when he thinks of his mother’s kiss, he recalls that “her lips … made a tiny little noise: kiss.” 30 Mr. Casey’s Phth! falls into this same category. Jeri Johnson notices the preponderance of these words, and argues that “if Stephen could be said to have a theory of language at this point, it would be the bow-wow or onomatopoeic theory: the word for the thing imitates its actual acoustic equivalent in reality: ‘suck’ has its name because things that ‘suck’ make A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 213

‘sucky sounds.’” 31 Young Stephen’s “onomatopoeic theory” is the young poet’s way of learning to translate his auditory experience into words, and vice versa; Joyce highlights this developmental stage of Stephen’s through the use of Germanic onomatopoeia. In addition to their sonic associations, Germanic-derived words in Joyce have strong visual connotations. “The word was beautiful: wine,” Stephen thinks. “It made you think of dark purple because the grapes were dark purple that grew in Greece.” 32 Although wine has a distant ancestor in Latin (vı num̄ ), its immediate parents are Middle and Old English. Stephen’s associations are, fi rst, of a certain dark purple color and, second, of Greece, which recalls the Homeric cliché that appears thrice in Joyce’s Ulysses : “ epi oinopa ponton ”—“the wine-dark sea.” 33 This is important to keep in mind, since, on the same page, a Latin lesson begins, in which Father Arnall “asked Jack Lawton to decline the noun mare ,” or sea. Jack fails to decline the noun, and “could not go on with the plural,” an implicit choice of the Germanic sea over the Latin mare . The fi rst plural of mare is maria , which is also the Latin name for Mary. Does Lawton’s failure to produce “Mary” in front of Father Arnall prefi gure Stephen’s eventual rejection of the Sodality of the Blessed Virgin, of which he was prefect? This would be a far-fetched hypothesis on the subject of any other author, but given Joyce’s famous love of puzzles, it is entirely plausible, and it takes place along an etymological axis. When Stephen prepares to confess his sins in section 3.3, we see another passage with a low L score, notable for its alliteration: “His blood began to murmur in his veins, murmuring like a sinful city summoned from its sleep to hear its doom. Little fl akes of fi re fell and powdery ashes fell softly, alighting on the houses of men. They stirred, waking from sleep, troubled by the heated air.” 34 The double alliterative structure here—a string of S– words interrupted by a string of F– words—recalls the verse style distinctive of Old English poems such as Beowulf . Germanic associations are thus present in both the poetics of this passage and the history of its words: since most of these words are of Germanic origin, this passage resonates with the immediacy necessary to evoke Stephen’s guilt and anxiety.

Hellenic Portrait ’s words of ancient Greek origin deserve a brief discussion. Greek words are some of the more diffi cult to quantify, as many of the Greek loanwords in English come to us through Latin and a few (such as “alchemy”) 214 J. REEVE through Arabic. When classical Greek works began to be rediscovered in 1453, after Greek scholars fl ed Turkish-occupied Constantinople, they brought along a number of associated loanwords. 35 This could explain why many Greek loanwords seem at home in Aristotle or Plato: drama, comedy and pathos recall the Poetics , while phenomenon, noumenon and democracy seem appropriate to a Socratic dialogue. As the analysis of the Brown Corpus hints, religious words, too, are heavily Hellenic: angel , evangelist , hagiography , bible and so on, are all descended from Greek. We might fi nd, therefore, that an aesthetic treatise of the kind Stephen presents in 5.1, or a religious sermon like Father Arnall’s in 3.2, might contain a higher proportion of words of Hellenic origin. Figure 9.5 shows that those two sections have, respectively, the fi rst and third highest H scores of any section. Father Arnall’s sermon in section 3.2 features the emotionally-charged Hellenic words agony (which appears, in this section alone, an amazing eight times), demon and zealous, along with the more tame words baptism, poetry and eon . In section 5.1, those words more befi t their setting in a physics classroom—physics , energy , and kinetic , along with didactic . These categories of religious and learned language are consistent with the analysis of the Brown corpus Fig. 9.2 . 36 The section with the second-highest H score is 4.3. Interestingly, this is the section where Stephen’s classmates taunt him in Greek: “Stephanos

Sections of A Portrait of the Artist as a Young Man

Proportions of Hellenic Words

0.25

0.2

0.15

0.1

0.05

0 1-2 1-3 1-4 2-1 2-2 2-3 2-4 2-5 3-1 3-2 3-3 4-1 4-2 4-3 5-1 5-2 5-3 5-4

Fig. 9.5 H scores for sections of Joyce’s Portrait of the Artist as a Young Man A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 215

Dedalos! Bous Stephanoumenos! Bous Stephaneforos!” 37 These words are a polyglot pun on his name and Greek words for a sacrifi cial cow adorned with a wreath. 38 However, this is not what the Macro-Etymological Analyzer is detecting; since the program doesn’t recognize words of languages other than English, it treats Bous and Stephanoumenos as errors. 39 The words of Hellenic origin in this section, then, are other English words: ecstasy and antogonism, for instance. That Joyce is using more than the usual number of Hellenic words here fi ts with the Daedalian myth, for on this same page, we see the epiphanic culmination of this metaphor, in the imagery of fl ight: His heart trembled; his breath came faster and a wild spirit passed over his limbs as though he were soaring sunward. His heart trembled in an ecstasy of fear and his soul was in fl ight. His soul was soaring in an air beyond the world and the body he knew was purifi ed in a breath and deliv- ered of incertitude and made radiant and commingled with the element of the spirit. An ecstasy of fl ight made radiant his eyes and wild his breath and tremulous and wild and radiant his windswept limbs. 40 Moments later, there is an Icarian anti-ephiphany that risks bathos, as Stephen’s thought is interrupted by the voices of a schoolmate playing in the water: “Oh, Cripes, I’m drownded!” (ibid.). It could certainly be argued that the Hellenic words represented here are more useful to close reading than they are to distant reading. Since there are so few Hellenic words, they are statistically insignifi cant. However, in literary analysis, the signifi cance of a single word could form the basis of a critical argument, while it may remain statistically uninterest- ing. Thus, even if the variance in H scores does not deviate very far from that in a random sample, the analysis might still be useful in helping to identify word categories such as those described above.

Celtic Although the Macro-Etymological Analyzer identifi ed only one word descended from the Irish language, sugan , the language has a deep effect on the styles of the novel. In fact, O’Hehir’s Gaelic Lexicon identifi es sixteen words of Irish descent in Portrait . 41 Some of O’Hehir’s words, like cool (from cúl , goal) are homographs with unrelated English words, a fact that might help to explain why they cannot be found by the program. 42 Others, like “smugging,” are of dubious Irish etymology. O’Hehir supposes that this word may be derived from smug or smuga , meaning “snot, 216 J. REEVE nose drip,” or “slime” (ibid.). In classic Joycean fashion, this word is not glossed. It appears early in the novel, when Athy relates that the Simon Moonan and Tusker Boyle are caught “smugging” in the restroom. This is such a somber revelation that the rest of the boys are silenced by the thought, but Stephen does not understand; “What did that mean about the smugging?” he thinks. 43 Johannes Hedberg hypothesizes that smugging can be traced to the Old English word smūgan , by way of the Middle English verb smuȜen , 44 but Alarik Rynell contends that Hedberg’s etymology is erroneous, and that it is more likely decended from Old English smugge , “a small secret place.” 45 Rynell uses the English phonesthemes of smugging to argue that “ smugging must indeed have seemed an appropriate colloquialism for masturbating .” This is also Attridge’s theory. 46 Most others assume that it’s a euphemism for homosexual play (see Howes 47 255, for instance), although Fargnoli claims it is “entirely made up and has no established meaning.” 48 The OED gives “to caress, fondle,” citing another of Joyce’s uses of the word in Ulysses , as well as the early nineteenth century poet Scottish poet Ebenezer Picken. That the only two citations for this sense are from Celtic writers lends some credence to the theory of Celtic etymology, but that all of these theories assume some kind of taboo schoolyard sexual act is more suggestive of the Germanic origin of the word. This is supported not only on the basis of G score of Brown Corpus romance texts, but also given the number of other similar four-letter sexual words in the Germanic register. More important to this discussion than the Celtic words themselves is the political function of the Irish language, especially as it existed in Ireland on the eve of independence. The revival of the Irish language was intimately associated with the nationalist movement, from which Joyce, as a self-imposed exile, had distanced himself both physically and intellectu- ally, but with which he nonetheless felt some affi nity. Although neither Stephen nor Joyce himself knew much Irish—Stephen stops taking Gaelic League classes after the fi rst lesson—they seem uncomfortable with the dual political and linguistic dominance of Britain over Ireland. A passage that illustrates this unease is Stephen’s conversation with the English dean over the word tundish . Stephen thinks: The language in which we are speaking is his before it is mine. How different are the words home, Christ, ale, master , on his lips and on mine! I cannot speak or write these words without unrest of spirit. His language, so familiar and so foreign, will always be for me an acquired speech. I have A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 217 not made or accepted his words. My voice holds them at bay. My soul frets in the shadow of his language. 49 Here, “his language” could be read as both the Dean’s British English dialect and English more generally. Since Irish is the ancestral language of Ireland, English is an “acquired speech” in this historical sense. More immediately, the tonal differences in their speech distinguish their two Englishes. Anthony Burgess has a notable phonetic interpretation of this passage, suggesting that Stephen likely chooses these four words because they are pronounced differently in British and Hibernian dialects—dip- thongs instead of long open vowels, fi nal schwas instead of retrofl ex Rs . 50 The words are ontologically different, as well. “Home” refers to different cities for the two men, “Christ” is very different for the Catholic and the Protestant, and as a student, Stephen’s “mastery” is that of a subject he is taught, while the Dean’s is that over people, that of a colonist and a schoolmaster. These colonial undercurrents are useful to a discussion of etymology in Joyce because they help to reveal choices of etymological modes as domains of nations, with histories and political uses. Joyce’s decision to have Stephen’s uncle Mat Davin use Irish-derived words like camann (from the Irish camán , the stick used in hurling) enforces the earlier description of him as a “young peasant” who “worshipped the sorrowful legend of Ireland.” 51 Stephen refers to this same object with an Anglo- Saxon word when he scoffs at the most recent Irish uprising, calling it “a rebellion with hurleysticks.” 52 Johnson explains that this is “a ‘sneerer’s’ comment on the failed Fenian Rising of 1867, training having taken place not with guns but with camann .” 53 Joyce’s use of camann and hurleysticks is not interchangeable, but chosen to evoke histories, politics and tones that each word carries. Words of Irish-language descent, therefore, join those of other national connotations to reveal the political dimensions of etymological word choice.

UNKNOWN WORDS Camann was one of about 4 percent of the words of Portrait that the Macro-Etymological Analyzer failed to fi nd. These words proved to be revealing about Joyce’s style, especially concerning etymological associations. Many of the unknown words are proper names, given that proper names were purposely removed from the database, as they would skew the results unnecessarily. 54 Other unknown words, however, are Joyce’s 218 J. REEVE

inventions. Some of these are true neologisms, while others are portmanteau words or unhyphenated compound words. While much critical attention has been paid to the neologisms in Finnegans Wake , since they are undoubtedly its distinctive property, not much has been discussed regarding Portrait , even though at least one word, pandybat has its fi rst OED citation in the novel. Regarding Joyce’s work as a whole, Katie Wales identifi es two neologistic strategies: “con- versions” and “compounds.” “Conversion,” Wales relates, “extends the semantic range of existing words by changing the grammatical function.” 55 “Compounds” refers to portmanteau words, which are littered throughout the novel. These two categories are extended by Joseph Prescott’s conception of Joyce’s “renovation” and “innovation.” 56 By “renovation,” Prescott claims that Joyce “imposes on words of common currency a fresh lustre, usually the brilliance of their fi rst years.” Among the examples Prescott gives for “renovation” is a passage from Ulysses where Joyce uses the word “crazy” in its etymological sense of “fractured” (309). In illustration of his category of “innovation,” Prescott calls Joycean neologism “dynamic onomatopoeia,” citing the “crescendo” of cat noises in Ulysses (311). When onomatopoeia is “triumphant” in Joyce, he argues, it constitutes the “anastomosis of style and subject.” Given the etymological associations with onomatopoeia established earlier, this fusion of style and subject could be said to take place along etymological vectors, as well, further foregrounding the importance of these vectors in Joyce’s style. To identify a list of Joycean neologisms beyond the errors introduced by the Macro-Etymological Analyzer’s treatment of proper names, the text was run through a command-line spell-checking program, and the results sorted, with this chain of Linux commands:

The result was a list of words the command aspell determined were “mis- spelled.” After manual curation to remove words in Latin, proper names, and real but obscure words, this became a list of Joycean terms. Most of these words are compound words, and often formed from two Germanic words, like suddenwoven or rainladen . These words can then be grouped into several themes. First, there are color words, like ambered , bloodred , greenwhite and redeyed , along with hueless and nocolored . Next, there are kinship terms, including fosterbrother , fosterchild , greatgrandfather , A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 219

halfbrother and granduncle . Another category features agrarian or pastoral terms like cowdung , cowhairs , goatish , milkcar , boghole and bogwa- ter . Finally, there is theme related to dirt, fi lth and the street: sootcoated , thumbblackened and greasestrewn . All of these categories are associated with the Germanic register. It should perhaps not be surprising that so many of Joyce’s neologisms and portmanteau words are of Germanic origin, since word compounding in this style is a feature of many modern Germanic languages, most notably modern German. In fact, many of these words, if separated into their constituent words (great grandfather ) and translated into German, prove to be one German word (Urgroßvater ).

CONCLUSIONS Joyce achieves many of the narrative effects of A Portrait of the Artist as a Young Man through the use of etymological registers. Just as the language of his narration, according to the Uncle Charles Principle, follows the thoughts of his characters, his oscillations between Germanic and Latinate linguistic modes mimic oscillations between epiphanic and anti-ephiph- anic scenes. Macro-etymological analysis, therefore, demonstrates that it might be well-suited to become part of suite of analytic tools that can participate in the detection of structural patterns of a novel. Along with word frequency analysis, principal component analysis, and segmentized type/ token ratio calculation, macro-etymological analysis might form a part of a greater textual analytic system that can inform and improve computational literary criticism.

NOTES 1. James Joyce and the Difference of Language (Cambridge, UK: Cambridge University Press, 2003), 1. 2. These languages are listed at the end of Joyce’s manuscript for the Wake. Katie Wales, The Language of James Joyce (New York: St. Martin’s Press, 1992), 31. 3. “The Study of Languages,” in Occasional, Critical, and Political Writing (Oxford, UK: Oxford University Press, 2008), 15. 4. Stephen Hero (New York: New Directions, 1944), 32. 5. “‘A Very Fine Piece of Writing’: An Etymological, Dantean, and Gnostic Reading of Joyce’s ‘Ivy Day in the Committee Room’,” in ReJoycing: New 220 J. REEVE

Readings of Dubliners (Lexington, KY: University Press of Kentucky, 1998), 220. 6. “Joyce and Skeat,” James Joyce Quarterly 24, no. 2 (1987): 178. 7. Joyce, A Portrait of the Artist as a Young Man , 8. 8. Marjorie Howes, “Joyce, Colonialism, and Nationalism,” in The Cambridge Companion to James Joyce , ed. Derek Attridge, 2nd edn (Cambridge, UK: Cambridge University Press, 2006), 255. 9. Joyce, A Portrait of the Artist as a Young Man , 150. 10. “The Study of Languages,” 15. 11. “The ‘Portrait’ in Perspective,” The Kenyon Review 10, no. 3 (1948): 17. 12. jonreeve.com/etym 13. In the current version, these are proportions of the total tokens, but a future version of this program will calculate proportions of the types. 14. An important early study of this corpus is Henry Kucerǎ and W. Nelson Francis, Computational Analysis of Present-Day American English (Providence: Brown University Press, 1967), Computational Analysis of Present-day American English ; a more recent study is Geoffrey Leech, Change in Contemporary English: A Grammatical Study (Cambridge University Press, 2009), Change in Contemporary English: A Grammatical Study . 15. Albert Croll Baugh and Thomas Cable, A History of the English Language (Abingdon: Routledge, 2010), 171. 16. John Paul Riquelme names and discusses each of these parts in “The Parts and the Structural Rhythm of a Portrait,” in A Portrait of the Artist as a Young Man (New York: W. W. Norton & Company, 2007), 307. 17. Ibid. 18. A Preface to James Joyce , 2nd edn (London: Longman, 1981), 63. 19. “Kinesis and Stasis: Structural Rhythm in Joyce’s Portrait,” University Review 3, no. 10 (1966): 22. 20. “The Art of the Labyrinth,” in Critical Essays on James Joyce’s a Portrait of the Artist as a Young Man (New York: G. K. Hall & Co., 1998), 197. 21. “A Portrait of the Artist as a Young Man and L’Éducation Sentimentale: The Structural Affi nities,” Orbis Litterarum 19 (1964): 164–5. 22. “The Parts and the Structural Rhythm of a Portrait,” 119, 104. 23. “The Role of Structure in Joyce’s Portrait,” in Joyce’s Portrait: Criticisms and Critiques , ed. Thomas E Connolly (Appleton, 1962), 108. 24. A Portrait of the Artist as a Young Man , 124–5. 25. Ibid, 7. 26. Ibid, 8. 27. “‘Suck Was a Queer Word’: Language, Sex, and the Remainer in a Portrait of the Artist as a Young Man,” in Joyce Effects (Cambridge, UK: Cambridge University Press, 2000), 59. A MACRO-ETYMOLOGICAL ANALYSIS OF JAMES JOYCE’S A PORTRAIT... 221

28. Wells, quoted in Robert Deming, James Joyce, the Critical Heritage (New York: Barnes & Noble, 1970), 86. 29. Ibid. 30. A Portrait of the Artist as a Young Man , 10–11. 31. “Introduction,” in A Portrait of the Artist as a Young Man (Oxford University Press, 2000), vii–xxxvii. 32. Joyce, A Portrait of the Artist as a Young Man , 39. 33. This phrase is discussed at length in William Gladstone’s Studies on Homer and the Homeric Age , where Gladstone argues on the basis of color words in Homer that the ancient Greeks lacked the ability to perceive colors like blue. 34. Ibid., 130. 35. Thomas Pyles and John Algeo, The Origins and Development of the English Language (Boston, MA: Wadsworth Cengage Learning, 2010). 36. The pattern of Hellenic words in Figure 9.5 also closely resembles patterns of religious images identifi ed in a 1980 computational study of Portrait by John B. Smith “A Computational Analysis of Imagery in James Joyce’s a Portrait of the Artist as a Young Man,” in Information Processing 71: Proceedings of IFIP Congress, 1971 , 1971, 283, http://www.cs.unc.edu/Research/jbsAr- chive/docs/JoyceStudy-IFIP71/IFIP71Chapter.pdf . In this study, Smith counts “images” that belong to certain taxonomies like “fi re” and “water,” and plots them according to their location in the novel. The category of “religion” aligns very roughly with the Hellenic plot in Figure 9.5 . 37. Joyce, A Portrait of the Artist as a Young Man , 141. 38. Brendan O’Hehir, A Classical Lexicon for Finnegans Wake: A Glossary of the Greek and Latin in the Major Works of Joyce Incl. Finnegans Wake, the Poems, Dubliners, Stephen Hero, a Portrait of the Artist as a Young Man (Berkeley [u.a.]: University of California Press, 1977), 528. 39. Although the program doesn’t detect other languages, it might be interesting to read macro-etymological results alongside proportions of foreign languages present in Portrait . Would high L scores correlate with inci- dence of Latin speech? 40. Joyce, A Portrait of the Artist as a Young Man , 141. 41. A Gaelic Lexicon for Finnegans Wake, and Glossary for Joyce’s Other Works (Berkeley: University of California Press, 1967), 335–6. 42. It is problematic that the Macro-Etymological Analyzer, and for that matter, most tools for computational text analysis, cannot differentiate between homonyms. Word sense disambiguation would certainly mitigate this issue. According to Gerard De Melo, this is a featured planned for a future version of the Etymological Wordnet, and when that is released, it will be incorporated into a future version of the Macro-Etymological Analyzer, as well. 222 J. REEVE

43. Joyce, A Portrait of the Artist as a Young Man , 35. 44. “Smugging. an Investigation of a Joycean Word,” Moderna Sprak 66 (1972): 25. 45. “On the Etymology of James Joyce’s Smugging,” Moderna Sprak 66 (1972): 367. 46. “‘Suck Was a Queer Word’,” 63. 47. “Joyce, Colonialism, and Nationalism.” 48. James Joyce A–Z (Oxford, UK: Oxford University Press, 1995), 207. 49. Joyce, A Portrait of the Artist as a Young Man , 159. 50. Here Comes Everybody (New York: Random House, 1987), 28. 51. Joyce, A Portrait of the Artist as a Young Man , 151–2. 52. Ibid., 169. 53. “Introduction,” 274. 54. Portrait would show unusually high proportions of Hellenic words, for instance, in every section where the word Stephen would appear. Analyses of Christian bibles showed similar results every time the word Jesus was mentioned, irrespective of the author’s choice to use words in the Hellenic register. 55. The Language of James Joyce , 115. 56. “James Joyce: A Study in Words,” Publications of the Modern Language Association 54, no. 1 (1939): 308, http://www.jstor.org/ stable/458640 . CHAPTER 10

Body Language: Toward an Affective Formalism of Ulysses

Kurt Cavender , Jamey E. Graham , Robert P. Fox Jr. , Richard Flynn , and Kenyon Cavender

In 2011, Critical Inquiry published a devastating critique by Ruth Leys of the then-prevailing, cognitivist account of affect—an account that posited affect as an unconscious, unrepresentable physiological response with the power to determine cognition. Since the publication of Leys’s article, textual critics including Fredric Jameson and fi lm scholar Eugenie Brinkema have called for a formalism of affect to specify how affect may be consciously represented by artists. 1 In this study, we use digital methods to explore and develop some predictions by Jameson regarding a formalism of affect in James Joyce’s Ulysses . Our results not only add evidence to Jameson’s compelling theory of literary history but also illuminate Joyce’s

K. Cavender () • R. Flynn Brandeis University, Waltham, MA , USA email: [email protected] J. E. Graham Le Moyne College, Syracuse, NY, USA R. P. Fox Jr. Tufts University, Medford, MA , USA K. Cavender Independent Scholar

© The Editor(s) (if applicable) and The Author(s) 2016 223 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_10 224 K. CAVENDER ET AL. literary techniques, touching upon such crucial stylistic concerns as corpo- reality, stream-of-consciousness and the representation of empathy. In The Antinomies of Realism (2013), Jameson identifi es novelistic representations of affect and in so doing proposes a new theory of the novel. According to Jameson, the modernist novel, far from a mere alternative to or reaction against the realist novel, is what follows from the completed history of the realist project understood as a historical dialectic of story and affect. As Jameson explains, a story (récit ) tells events in the past tense, gives names to the emotions felt by its characters and is loaded with implications of meaning and inevitability. Beginning around 1840, however, feelings that previously have had no name, meaning or inevitability come to be privileged as vehicles of the “real.” Nineteenth-century realism unfolds dialectically as story adapts to include these feelings—affects— only thereby to open a space for the discovery of new feelings that elude contemporary forms of story. Jameson claims that Ulysses occupies the liminal position between nineteenth-century realism and post-realist modernism. In Ulysses , story has at last assimilated affect, all forms of realist story are present as fragments of consciousness, and for this very reason, “[S] uch categories may no longer be the best ones to convey everything which is unique” about the novel as a form. 2 Like other theorists of affective form, such as Eugenie Brinkema, Jameson assumes that a turn to form demands a (re)turn to the close reading of discrete passages consciously selected by the critic. It is here, however, that we believe Jameson’s analysis meets its limitations; for if, as Jameson suggests, what is at stake is that affective forms in Ulysses are all- inclusive, omnipresent and fragmented, any critical approach that begins with close reading of isolated passages will struggle to negotiate between the confounding diversity of the text’s totality and the presumably revela- tory potential of whichever passage it selects as a starting point. Jameson’s reliance on close reading constrains his treatment of Ulysses in two ways. First, the novel is reduced to a collection of memorable features, which become exempla standing in for realism’s aufhebung by modernism. Molly stands in for the preservation-in-cancellation of the realist novel of adultery. Stephen similarly represents the Bildungsroman. The scene in the Sirens episode in which Bloom belches during lunch exemplifi es realist commit- ment to traditional personhood, which is superseded at the same time by impersonal subjectivity. The prominence of dialogue throughout the novel demonstrates the relocation of voice from the individual person to the individual body. The Nighttown scene of the Circe episode typifi es the libera- BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 225 tion of multiple storylines from a unifi ed plot. 3 In these equations, Ulysses functions as a storehouse of images available to illustrate theoretical claims developed by Jameson elsewhere, in his sustained interpretations of earlier and later novels. But to treat Ulysses primarily as a resource for theoretical arguments elsewhere minimizes the degree to which a sustained, holistic or self-present reading of Ulysses is made diffi cult and perhaps even precluded by the intangibility of dialectical movement, which can only be understood retrospectively. Inherent in this way of proceeding is the risk that Ulysses will become a mere repository of comparisons and symbols, a transcenden- tal index that lays bare the mechanisms of literary historical development elsewhere even as its own internal dialectic is ever more obscured. A second limitation of Jameson’s approach is that his method of close reading works against a key element of his dialectic, namely the dichotomy between affect and named emotion. Consistent with Leys’s views, Jameson argues that the opposition on which conventional affect theory is based—of value-laden “emotions” to psychophysiological “affects” or “feelings”—is in truth an ancient and self-contradicting distinction. He proposes the alternative opposition of named emotion to any feeling that has so far escaped reifi cation in language, which unnamed feeling he calls “affect.” Freed from the conceptual and cultural freight of established usage, affect “seems to have no context, but to fl oat above experience without causes and without the structural relationship to its cognate entities which the named emotions have with one another.” 4 Story rapaciously converts affect into named emotion, forcing the realist novel to seek out ever more hidden sources of unnamed feeling. To defi ne affect thus, however, poses a problem for the student of affective form: how are we supposed to isolate affect within a literary work if affect by defi nition resists straightforward representation in language? In response to “the resistance of affect to language,” 5 Jameson produces close readings of passages that he claims are affective in the novels of Eliot, Zola, Tolstoy Woolf, and others, trusting interpretation to overcome the absence of conventional systems of meaning. Yet he reluctantly concedes that this risks consigning affect to the eye of the beholder, going as far as to say of one passage that its affective dimension “is little more than a presumption, the reader must somehow introduce it from the outside.” 6 The appearance of affect in closely read passages therefore cannot be taken as a guarantee that there exist objective linguistic markers of affect, in Ulysses or in any novel. To correct for the foregoing limitations in Jameson’s approach, we propose two objective markers of Jamesonian affect, markers that will support 226 K. CAVENDER ET AL. verifi cation by distant reading. Because affect, for Jameson, is the opposite of named emotion, one logical, negative marker of affect within a text should be the relative absence of words that name emotions. A search for the absence, not the presence, of a vocabulary of sentiments avoids both the emotion-oriented categorizations of conventional affect theory and the idiosyncratic evidential choices characteristic of affect theory after Leys. In addition to this negative marker, we propose a positive marker of affect: words that pertain to the body. Rooting psychic phenomena in the body is the aim uniting virtually all research on affect. 7 To this end, Jameson gestures toward an affective bodily vocabulary that includes the names of organs, appendages and “sense data” independent of any web of meaning. 8 We take it as our hypothesis that the more a textual segment dwells on embodiment while at the same time suppressing named emotions, the more likely it will be a vehicle of affect. In the following pages, we test this hypothesis using Affectcrawler, a program we designed specifi cally for the task, in order to locate, measure and analyze affect in Ulysses . In doing so, we hope to demonstrate that affect studies, for which Ulysses has long been an important text, has much to gain from the tools of computational analysis, many of which are already being put to use by a growing and diverse body of digital scholarship on Ulysses .

METHODOLOGY Recent digital scholarship on Ulysses can be grouped into three broad categories. 9 Digital editions, such as Project Gutenburg’s free download- able PDF of the 1922 text, Amanda Visconti’s interactive and collectively annotated Infi nite Ulysses (based on the Modernist Versions Project digital edition) or Robert Berry’s webcomic adaptation Ulysses Seen , all seek to make Joyce’s text available across multiple media forms and platforms and to transform the reading experience into a hybridized, collaborative effort. 10 Alternatively, mapping efforts proliferate, such as Alex Christie and Katie Tanigawa’s Dislocating Ulysses project, which overlays “the reader’s geotemporal experience” onto a 3-D model of 1925 Dublin, and Steve Cole’s Ulysses Meets Twitter 2011 project, which restages the novel as a series of tweets (@11lysses). These efforts foreshadowed the Modernist Versions Project’s recent Year of Ulysses , as well as JoyceWays , a mobile app developed by a team of modernists and digital humanists from Boston College that combines four hours of audio explanations with a GPS guided tour of Dublin, transcode the text’s richly imagined fi ctional world into a historical Dublin whose geographical, architectural and anthropological BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 227 traces remain accessible to literary tourism. 11 Finally, the comparably fewer instances of computational analysis of Ulysses , such as Wayne McKinna and Alexis Antonia’s lexical analyses of interior monologue and style, make use of various text-mining tools to abstract formal or semantic patterns, which then become the object of critical attention. 12 The fi rst two categories make use of digital tools and platforms to enhance the reader’s experience of Ulysses, with the consequence that interpretive work is largely shifted into the background; the intervention consists of designing new reader interfaces and creatively curating supplemental materials. Although such projects can contribute to energetic new readings of the novel in which, as Mark Marino has said of efforts by Louis Armand and others to hypertextualize Ulysses , “the reader will co-create the work,” too often textual analysis begins and ends with this “imaginative and ontological leap … from modernist work to hypertext.” 13 The third category—in which we would place our reading of Ulysses through Affectcrawler—uses machine algorithms to recognize patterns in single texts or massive textual corpora. These patterns can be abstracted as quantifi able data, which are then subjected to a focused, humanistic interpretive effort. The best of such efforts anticipate Matthew Jockers’ recent call for a synthesis of computational and humanistic analysis that “promises a new, enhanced, and better understanding of the literary record” by shifting attention away “from looking at the individual occurrences of a feature in context to looking at the trends and patterns of that feature aggregated over an entire corpus.” 14 This raises the question, however, of why programs already developed in recent years to analyze textual sentiment are, in their current forms, inadequate as mechanisms for analyzing literary affect. What makes a tool specifi c to literary language necessary? Machine sentiment analysis (SA) most often takes the form of programs that measure a data set against predetermined lexica of linguistic sentiment, such as the Dictionary of Affect in Language (DAL), the Word Net Affect dictionary (WNA) or the General Inquirer lexicon (GI). Although distinct from each other in the way they structure lexical relations and theorize emotion, the DAL and the WNA share the basic assumption that emotion, affect, hedonic sensation and other registers of attitude toward a stimulus are complementary and coexisting dimensions of experience. 15 That is to say, sentiment analysis based on these lexica necessarily imagines affect as a multidimensional, homogeneous space defi ned by orthogonal axes and then attempts to locate the affective qualities of a text by locating individual points within 228 K. CAVENDER ET AL. this space. Since we, following Jameson and Leys, suppose an irreducible antinomy between affective experience and emotional categorization, we would need to introduce into any such space a dimension of negation, an axis of lexical relation along which affect is transformed into its opposite. To address this problem, we situate two lexical categories, Named Emotions (E) and Affective Intensifi ers (I), as mutually exclusive and mutually negating. The fi rst category comprises familiar emotional experiences reifi ed into stable concepts: Love, Sadness, Pity, Hatred and so on. The second category comprises liminal or uncanny impressions or intensities that can transform familiar bodily sensation into an experience charged with disruptive or transgressive meaning—in a word, affect: Darkness, Queerness, Glowing, Bursting. Simply put, we theorize literary affect as the transformation of mundane embodied existence—represented by a third lexical category (B)—under pressure from such affective intensifi ers, without reference to the emotional language that might contain or defi ne the moment’s dissonant intensity. Thus, situating these two categories as oppositional allows us to introduce a split into the homogeneous multidimensional space imagined by conventional SA approaches, and in so doing to formalize affect as an experience outside of and irreducible to familiar emotional frameworks. A second general problem with existing SA lexical corpora is that they are limited in scope and purpose. Research suggests that while “there does appear to be a language of sentiment distinct from general language,” the nature and range of affective expression in a text varies signifi cantly according to the kind of text. 16 Consequently, tools for public sentiment analysis, designed for use in commerce and the social sciences, cannot be assumed to transfer easily into literary applications. As computational efforts by Reyes, et al. to formalize literary phenomena such as irony have shown, the literary dimension of affect is often located in the tension between the surface meaning of a text and multiple “non-literal meanings” that express “the opposite of what a shallow interpretation might normally conclude.” 17 As a result DAL and GI, which derive their lexica from various lists of most commonly used English words, 18 are insuffi ciently specifi c. We therefore argue any lexicon supporting an affective analysis of Ulysses must derive from an archive comprising one work—Ulysses —and must take into account the many neologisms, layered meanings and peculiar lexical relations specifi c to that text. In producing the lexicon used by Affectcrawler to read Ulysses , we begin with the 2,000 most commonly appearing words in Ulysses , which BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 229 we tag by hand as belonging to one or more of four categories: named emotion (E), embodied sensation (B), affective intensifi er (I) or null (X). The B and I lists are allowed to overlap, while the E and I lists are not. The process of compiling the lists reveals some unique challenges of theorizing affect as a literary phenomenon. For example, there exist words that would be null in a general context but that may intensify affect in Ulysses , such as words relevant to Catholic-Protestant tensions (“trinity”; “priest”) or Irish-British tensions (“Dublin”; “empire”). Some words for colors or clothing appear in contexts where physical sensation is being described (“yellow”; “coat”), whereas others do not, indicating that their bodily referent is displaced into a metaphorical, potentially affective domain (“golden”; “petticoat”). Incidentally, the likelihood that a word will appear in both the B and the I categories increases as its frequency in the text decreases, suggesting that such metaphors are constrained by context. Allowing our category assignments to overlap and to depart from common usage sensitizes Affectcrawler to the literary registers of Joyce’s language, the formations of non-literal or apparently contradictory meanings that could be lost to an approach derived from general language use. Moments of semantic ambivalence appear often as affective intensifi ers, linguistic catalysts that transform embodied experience into a moment of powerful sentiment; the affective dimension of such language appears as a secondary or subtextual current beneath a conventional sensory meaning. By isolating passages in the novel densely populated with I and B terms but minimizing E language, Affectcrawler is able to distinguish between moments of textual sentiment, grounded in named emotional states and moments of affective intensity, which register a greater range of meanings beyond what they name. We do not attribute any immediate interpretive value to the computational results, but rather rely on computational analysis to draw attention to semantic patterns that persist deep below the surface of the text. Affectcrawler creates a textual map of these patterns by locating passages within the text according to four variable parameters: minimum frequency of B and I words, maximum frequency of E words, and minimum passage length. Once these variables are determined, Affectcrawler proceeds through the text in single-word increments, identifying and isolating passages in which each minimum- length sequence meets the word frequency conditions. As a consequence, search parameters of <1 (E), >3 (B), and >3 (I), and specifying a minimum segment length of eighteen words, might return some passages as long as fi fty- three words (one such result appears in the Oxen of the Sun. 230 K. CAVENDER ET AL.

episode). Remarking these isolated passages in the full text allows us to map the contours of the text’s affective intensities, identifying passages that exhibit strong evidence of performing affective work. By searching for moments in the text that demonstrate dense clusters of embodied sensation and affective intensifi ers but that do not make sense of these experiences in terms of named emotions, we locate peaks of the affective intensity that, according to Jameson, characterize modernist realism. While the goal of both Jameson’s analysis of affect and our own is similar—the identifi cation of formal elements that work together to create locations of affective intensity within a literary work—the two methodologies approach the text from opposite directions. Jameson’s analysis begins by isolating and then closely reading a passage he deems to be particularly pregnant with affective intensity, before going on to speculate that the passage refl ects a larger pattern of affect to be found in the text as a whole. In contrast to this “bottom-up” approach, our “top-down” approach begins with the text in its entirety, on which, in the words of Jockers, “the computer assists in the identifi cation and compilation of evidence” before we “in turn, interpret and explain that derivative data.” 19 Using this blended approach, which combines both empirical and textual analysis, we (aided by Affectcrawler) begin by transforming the Ulysses text algorithmically into isolated segments of affective intensity and by analyzing the characteristics common to such segments. We then proceed to interpret passages with a signifi cant presence of such segments in order to consider how affect is created and heightened at such moments.

RESULTS AND INTERPRETATION Running the text of Ulysses through Affectcrawler—which, as noted, highlights passages with greater than three B (embodied sensation) and I (affective intensifi ers) words for each E (defi ned emotion) word— 20 we found that a segment length of eighteen words generated the greatest number of fi ndings with a total of 219 highlighted passages. After assigning the tagged segments to the eighteen episodes into which critics customarily divide Ulysses (Telemachus, Nestor, Proteus, etc.), we then derived three results from the data: the density of affective intensity across episodes, the specifi c passages which presented a heightened concentration of fi ndings, and the quantity and diversity of embodied sensation words in segments centering on Leopold Bloom and Stephen Dedalus. To compensate for variations in page length between episodes, we calculated “affect density” by dividing the total number of fi ndings in each BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 231 episode by that episode’s page count. To confi rm that the results generated were not particular to eighteen-word segments, we performed the same operation using fi fteen, twenty, twenty-fi ve and thirty word counts. While the average number of tagged segments per page for each episode did vary based on segment word length (Table 10.1 ), the overall distribution of episode densities was largely independent of segment length (Fig. 10.1 ). The graph highlights the fi ve episodes of Ulysses with the highest levels of affect density: Calypso, Lestrygonians, Lotus Eaters, Sirens and Nausicaa. Signifi cantly, all are episodes in which Leopold Bloom is the central character, as opposed to Stephen Dedalus or Molly Bloom. A closer look at the affect density results for all eighteen episodes further underscores the correlation of affective intensity with Bloom: the average density of the eight episodes that center on Bloom (the aforementioned plus Hades, Aeolus and Cyclops) is 0.58, signifi cantly higher than the 0.40 average density of the episodes that follow Stephen (Telemachus, Nestor, Proteus, and Scylla and Charybdis) and the 0.24 average density of episodes in which both Bloom and Stephen appear (Wandering Rocks, Oxen of the Sun, Circe, Eumaeus and Ithaca). Penelope, Molly’s monologue, which ends the novel, has a density of 0.44.

Table 10.1 Sentence length by episode in Ulysses

Episode 30 words 25 words 20 words 18 words 15 words

Telemachus 0.13 0.13 0.43 0.43 0.04 Nestor 0 0 0.42 0.42 0 Proteus 0.08 0.08 0.58 0.58 0 Calypso 0.29 0.21 0.93 0.86 0.07 Lotus Eaters 0.14 0.14 0.86 0.86 0 Hades 0.07 0.04 0.48 0.52 0 Aeolus 0.06 0.06 0.28 0.28 0 Lestrygonians 0.2 0.13 0.57 0.57 0.1 Scylla and Charybdis 0.03 0.03 0.16 0.16 0.03 Wandering Rocks 0.19 0.15 0.33 0.33 0.04 Sirens 0.26 0.2 0.66 0.53 0.03 Cyclops 0.04 0.04 0.23 0.23 0.06 Nausicaa 0.4 0.36 0.76 0.73 0.2 Oxen of the Sun 0.1 0.07 0.27 0.23 0.07 Circe 0.05 0.03 0.2 0.21 0 Eumaeus 0.02 0.05 0.19 0.19 0 Ithaca 0.06 0.08 0.22 0.22 0.03 Penelope 0.12 0.05 0.33 0.44 0 232 K. CAVENDER ET AL.

Fig. 10.1 Affective density by episode in Ulysses

After observing the affect density of segments in episodes associated with specifi c characters, we identifi ed local passages presenting a heightened concentration of affective segments, which we defi ne as three or more fi ndings per page or fi ve or more fi ndings in fi ve pages. The tagged results show far more passages of heightened concentration—and thus more localized affective intensity—in episodes centering on Bloom (ten instances) as opposed to Stephen (two instances). Thus, whether viewed from the perspective of broad “affect density” or more localized “heightened concentration,” our results indicate a strong correlation of affective intensity with Bloom and a comparably weak one with Stephen. That Bloom would be more closely associated with the body while Stephen would be more cerebral comes as no surprise to the reader familiar with Ulysses , 21 the difference in their temperaments being succinctly captured during the Nighttown scene of Eumaeus, when Stephen “thought to think of Ibsen” while Bloom “inhaled with internal satisfaction the smell of James Rourke’s city bakery” (614). Although this confi rmation of conventional readings of Bloom’s and Stephen’s characters is helpful in checking our digital methods, we believe that the true signifi cance of our BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 233 results lies instead in what they can tell us about the literary operation of affect in Bloom- and Stephen-centered episodes. More specifi cally, in the next section, we will argue that the superabundance of embodied sensations in key Bloom episodes does not merely signal affect but uses affect to represent empathy and a changing point of view. Bloom thinks and emotes with his body ; he empathizes with others and adopts their subjective perspectives all through the affective displacement of his bodily sensations. Stephen, meanwhile, thinks and emotes in the more traditional terms of conscious thought and named emotion. Consequently, he is restricted to his own point of view and is less capable of empathy than Bloom. Soon after leaving Davy Byrne’s pub, weary from the cacophony of sights, sounds and smells that assault his senses there, Bloom runs into a “blind stripling” tapping his way down the sidewalk. An alien creature, unresponsive, almost without expression, the boy is an enigma. As he helps the boy cross the street, he sees what the stripling cannot: a van, the bright red hair of its driver, a horse, a pretty girl. The boy, for his part, remains opaque, almost silent. Bloom perceives the boy’s face as a “wall face” (180)—a hard wall off of which Bloom’s thoughts rebound in the absence of any meaningful feedback from the stripling. Bloom’s initial thoughts are, signifi cantly, emotional rather than affective as he tries to ascertain what the boy is trying to do. Even as Bloom strives to “not do the condescending,” his initial perceptions are those of a superior onlooker pitying the diminished, handicapped body, the “[p]oor young fellow” with a “thin elbow” and a “limp seeing hand to guide it forward” (181). 22 While Bloom begins to imagine what it would be like to inhabit the boy’s body, he does so from his own point of view as a sighted person, a point of view which automatically introduces a dimension of emotion. Bloom thinks of staining one’s coat, making a mess of one’s food, and needing to be spoon-fed, experiences which a sighted person might well consider humiliating or disgusting. Crucially, however, Bloom does not remain in that world of emotional experience. At this moment, the text adopts a linguistic register of affective synesthesia to represent Bloom’s efforts to imagine the blind stripling’s embodied existence. Through the synesthetic substitution of touch, hearing, smell and taste for sight, Bloom experiences anew the van in the street, the cobblestones streets of Dublin, the reading of books, the seasons and even the love of women with alternative senses—as it were, with a new, blind body. In so doing Bloom constructs a point of view for the stripling and situates himself within it. This operation is fi gured through a familiar trope of walking in another’s shoes: 234 K. CAVENDER ET AL.

Mr Bloom walked behind the eyeless feet, a fl atcut suit of herringbone tweed. Poor young fellow! How on earth did he know that van was there? Must have felt it. See things in their foreheads perhaps. Kind of sense of volume. Weight. Would he feel it if something was removed? Feel a gap. Queer idea of Dublin he must have, tapping his way round by the stones. Could he walk in a beeline if he hadn’t that cane? … Look at all the things they can learn to do. Read with their fi ngers. Tune pianos … Of course the other senses are more … Sense of smell must be stronger too. Then the spring, the summer: smells. Tastes … And with a woman, for instance … Must be strange not to see her. Kind of a form in his mind’s eye. The voice temperature when he touches her with fi ngers must almost see the lines, the curves. His hands on her hair, for instance. Say it was black for instance. Good. We call it black. Then passing over her white skin. Different feel perhaps. Feeling of white. (181–2)

Here, the notion of walking in another’s shoes is taken literally, not as a fi gure of emotional imagination, but as bodily imagination. Bloom sub- stitutes tactile spatial relations for the qualitative distinctions he would normally visualize. The van is “felt,” volumetrically; its absence would feel like a “gap.” Straight lines are felt out by the blind boy’s cane. Text on the page is felt. The notion that the blind touch, hear, smell and taste more profoundly (“Of course the other senses are more …”) appeals to the sensual Bloom, and so he embarks on a hedonistic fantasy, smelling and tasting spring and summer before coming to his favorite sensory memory: sex. Imagining sex as the blind boy, Bloom’s senses blend together (“voice temperature”) until he sees white skin and black hair solely by feel. The scene ends with Bloom similarly “seeing” the colors and textures of his own fl esh, the “fi ne fi ne straw” of his “hair combed back above his ears” and the “whiteyellow” of the “slack fold of his belly” (182). Bloom thus comes full circle: having begun the scene in a state of frustrating epistemological limitation vis-à-vis the other, Bloom uses synesthesia to access the blind stripling’s elusive point of view, thereby achieving not only empathy without pity but also a fresh perspective on himself. That Bloom’s encounter with the blind stripling causes him to experience a moment of synesthesia is also important insofar as it challenges the customary understanding of the nature of perception on which the realism of Ulysses is based. Over the course of the scene, Bloom goes from accessing the world empirically—through what his own senses, particularly sight, tell him is there—to a more phenomenal approach, 23 “seeing” what he believes ought to be there by means of his non-optical senses. In the BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 235 process, Bloom steps outside of himself, exchanging his own point of view for the imagined, internalized point of view of the stripling. Indeed, the Dublin which Bloom has heretofore experienced alone becomes a space defi ned by Bloom’s expectations of what the blind stripling must perceive. The Dublin that Bloom inhabits, then, is not simply the city as he experiences it through his senses, but as it has been remapped by investing this bodily experience with an affective charge. The affective intensity of the blind stripling scene appears even more strongly in contrast with a similar scene in the Proteus episode in which, wandering on Sandymount Strand, Stephen contemplates the nature of perception, in his words “ineluctable modality of the visible” (37). 24 Wondering what it would be like not to be able to “read” the world around him, not to have “thought through [his] eyes” (37), Stephen imagines what it would be like to “[s]hut [his] eyes and see” (37). But whereas for Bloom “blindness” was a place of discovery, “blindness” for Stephen is a place of risk: although Stephen tries to reassure himself that he is “getting along nicely in the dark,” the experience leaves him anxious, afraid that he will fall “over a cliff” and even more afraid that, when he opens his eyes, all will have “vanished since,” leaving him “for ever in the black adiaphane” (37). Unlike Bloom, whose experience affi rms the possibility of “seeing” through other senses, Stephen never experiences synesthesia. If Bloom “walks behind the eyeless feet” of the blind stripling, thus gaining vicarious access to the boy’s experiences, the blind for Stephen will always be “them” (7). While Bloom immerses himself in the blind stripling’s body, Stephen insists on trying to draw blind experience into himself. Focused on his own rational processes rather than on collapsing distinctions between his senses, the “wrack and shells” that he “crush[es]” under his boots as he walks across Sandymount strand will never be anything more than something “to hear” (37). What becomes clear, then, in the contrast between Bloom’s and Stephen’s experiments with imagined blindness is that the entryway into Joycean affect is through the body, not through reason. And because Joycean affect is embodied rather than rational, affective imagination is not limited to rational bodies. It allows for an affective resonance with animals as well as humans. That Bloom’s bodily experience allows for a heightening of affective intensity across species is apparent from his fi rst appearance at the beginning of Calypso. Bloom’s breakfast prepa- rations are interrupted by the sudden appearance of the family cat, who enters “walk[ing] stiffl y round a leg of the table with tail on high” (55). The cat’s initial “Mkgnao” followed by a “Mrkgnao” (55) (in response 236 K. CAVENDER ET AL. to Bloom’s question of “milk for the pussens” and a “Gurrhr” when she settles into his lap) suggests, through the use of the different “words” (not merely sounds), that the cat can “understand … better” what Bloom says and does, while the cat’s language is impenetrable to Bloom. The distance between cat and man in this scene is similar to the epistemological gulf that divides the blind stripling from Bloom in Lestrygonians. In both cases, Bloom’s communication with his companion is limited by a problem of embodiment, and in both cases affective compensation allows Bloom to project himself into the bodily experience of his companion. While at fi rst Bloom refers to the cat and her kind as “them” (much like Stephen did with the blind “them” in the Proteus scene), he soon shifts to wondering “what [he] look[s] like to her” (55), suggesting an interest in bridging the human/animal divide. Clearly Bloom’s engagement with the cat is not about anthropomorphizing the cat’s experience but rather about zoomorphizing Bloom’s experience. 25 In the opening lines of Calypso, Bloom is described as having a lower carnivore’s taste for “thick giblet soup, nutty gizzards, [and] a stuffed roast heart” along with “grilled mut- ton kidneys … a fi ne tang of faintly scented urine” (55): this penchant for offal is what provides the basis for Bloom’s shared bodily experience with the cat. Like his later observations of the blind stripling, Bloom seeks here to experience the world from the perspective of the other. For Bloom, to watch “the bristles shining wirily in the weak light as she tipped three times and licked lightly” and then “wonder if it is true [that] if you clip them they can’t mouse after” (56) is to question the limitations of the senses in experiencing the world; the cat’s inability to move in the dark with clipped whiskers is analogous to the stripling’s inability to move in blindness. Affective imagination is the mechanism through which the text zoo- morphizes Bloom. If synesthesia is what allows Bloom to share the blind stripling’s embodied experience, then hyperesthesia through the imagination of extra sensory organs is what allows Bloom to share the cat’s experience. Bloom’s refl ections on how the roughness of the cat’s tongue allows her “to lap better, all porous holes” (56) is motivated by curiosity about the feeling, not by a spirit of rational, empirical inquiry. Similarly, Bloom’s conjectures about the cat’s whiskers are characterized by a certain concern with whiskers as extensions of bodily experience: “Wonder is it true if you clip them they can’t mouse after. Why? They shine in the dark, perhaps, the tips. Or kind of feelers in the dark, perhaps.” Here, Bloom imagines not just the utility of whiskers, but about the sensual experience, the feeling , of having whiskers as a part of his body. Affective imagination thus becomes a capacity to virtually modify the body. BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 237

CONCLUSION In conclusion, Affectcrawler proved a helpful tool for charting affective intensity in Ulysses . We recognize, however, that there are limits to the patterns of affect that Affectcrawler—or, for that matter, any program that works with word frequencies—can detect. For example, basing an algorithm on embodied sensation and named emotion words supposes that affect resides more in the subject experiencing affective intensity than in the object inspiring such resonance. Yet as Jameson’s reading of the “sym- phony of cheeses” episode from Zola’s La Ventre de Paris (1873) suggests, affect’s “liberation from meaning” can sometimes be found in an “impersonal” present, as when the smells wafting from a cheesemonger’s stall at Les Halles prove more than just a metaphor for the biting gossip of two passing shoppers. 26 While we can imagine additions to our method that might detect such variations, it seems likely that there are more forms of affect in heaven and earth than are dreamt of in our algorithm. Were we to run this experiment again, we would certainly consider developing more comprehensive lists of the B, I and E words on which Affectcrawler operates. It would improve the accuracy of our results to form a lexicon of more than 2,000 words (this fi gure being less than 15 percent of Ulysses ’ total of approximately 30,000 unique words). Many valid B and I words are surely not caught using such a low ratio of the lexicon, and even in the passages we analyze in this paper, the proportion of B and I words may be higher than our method determined. Further, we need to do more work in order to identify both compound words (the present lists include only single words) and words with particular valences within this particular text. That said, we had found it reasonable to put an upper limit on the number of listed words, as classifying words was the most time-consuming process as well as the one most subject to human error. This leads to another consideration: should we create a standardized process for classifying words? The word lists we generated were a product of intuition and consensus: each of the 2,000 words was classifi ed by two of the authors, and where disagreements occurred we debated each case and came to a consensus agreement about how to classify each word. Would the results have been compromised had we set up clear objective standards ahead of time by which to resolve these cases? What would be lost in such a process is any leeway for an individual author’s idiosyncra- sies. While we make certain allowances for what might be affective in Joyce , a standardized process would not. This, however, seems like a reasonable tradeoff for having a process that we could apply to texts more broadly. 238 K. CAVENDER ET AL.

We can envision further applications for Affectcrawler, one of the most immediate being to test Jameson’s theory of affect more broadly. Part of Jameson’s thesis in The Antimonies of Realism is that between 1840 and 1920 affect becomes increasingly pervasive in the novel. It is feasible to test this theory using Affectcrawler by taking a number of novels from each year or decade and testing whether the density of I and B words in the relative proportion to E words increases over time or not. There are, to be sure, diffi culties to overcome in such an investigation. For instance, how does one devise a lexicon to test novels over time, when the semantic value of words does not remain constant over time? It would be too time- consuming to develop a separate lexicon (as we do for Ulysses ) for each novel considered. Could we instead make one for each decade? Or perhaps restrict ourselves to a set of words whose meanings are relatively stable over time? Another phenomenon such a project would have to address is the tendency of affective feelings to become codifi ed over time, thus turning affect into named emotions. Jameson gives a compelling account of just such a transformation with the nineteenth-century terms “ennui” and “anxiety,” which “articulated undiscovered states of being which, while perhaps not newly emergent, were at least dormant if not unconscious in everyday human existence” (30). Once these feelings are named and begin to participate in the conceptual organization of conscious life, they are no longer properly affective. We would need to adjust our algorithm to account for such historical changes. While we have yet to resolve all these challenges, we did perform some preliminary testing on George Eliot’s Middlemarch (1871–72) in order to compare the Victorian novel par excellence with the modernist novel par excellence. This bore promising results as to what a more in-depth investigation of affect in the nineteenth-century novel might reveal. We compiled B, E and I lists for the 2,000 most commonly occurring words in Middlemarch and then ran Affectcrawler on the entire book using the same parameters as we did for Ulysses . The fi rst fi nding preceded our use of Affectcrawler: while compiling our B, E and I lists, we found that far fewer words fell into any of those categories than was the case for Ulysses ; this relative dearth of concrete perceptual language already provides some support for Jameson’s historical claim. Running Affectcrawler on the text, in turn, identifi ed far fewer affective passages in Middlemarch than in Ulysses : at the optimal segment length (twenty words), Affectcrawler identifi ed fi fty-four highly affective passages, less than one quarter of the number of passages identifi ed in Ulysses (219), a book of roughly the same page length. Undoubtedly, some of the difference can be explained by the fact that far more of the words we tagged for Middlemarch fell into the “null” BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 239 category, meaning our B, E and I lists for Middlemarch were shorter than they were for Ulysses; nonetheless, the lists were not four times shorter. We suspect that Affectcrawler identifi ed fewer passages because Middlemarch focuses more on conventional emotions at a conscious, conceptual level rather than at the level of physical sensation. That being said, it is surprising that the optimal segment length in Middlemarch (twenty) is roughly the same as for Ulysses (eighteen); we expected that the unit of affect would be smaller, supposing that affect would be shorter in duration in a “less affective” text. On the one hand, it may be that the type of affect Affectcrawler is programmed to detect is best represented in eighteen- twenty word segments, regardless of period or genre. On the other hand, it could be that the type of affect found in Middlemarch is more similar to affect in Ulysses than we might have thought, and that literary representations of affect are surprisingly stable over time. To probe more deeply into these fi ndings could prove important for our understanding not only of the literary innovations of modernism but of literary history more broadly. In one of the more than sixty fragments of the Aeolus episode which chart the paths taken by characters journeying through the Irish capital, the narrator of Ulysses observes: “Dublin. I have much, much to learn” (144). That the much to learn lies in one’s ability to shift between the fragment and the path, be it in Dublin or Joyce’s novel, presents an apt metaphor for the top-down approach we employ in our reading of affective intensity in Ulysses. Affectcrawler traces the coordinates of the path of affect that Jameson traces through the nineteenth- and early twentieth- century novel, calling attention to landmarks of affective intensity laying along the way. We hope that the algorithm we have developed may prove helpful to other readers charting their own paths.

NOTES 1. Eugenie Brinkema. The Forms of the Affects . Duke University Press, 2014. Frederic Jameson, The Antinomies of Realism , Verso, 2013. Ruth Leys, “The Turn to Affect: A Critique.” Critical Inquiry 37 (Spring 2011): 434–72. 2. Jameson, Antinomies of Realism , 216. 3. From ibid.: Molly and the novel of adultery, 150–1; Stephen and the Bildungsroman , 150–1; Bloom’s lunch, 24; dialogue, 98; Nighttown, 207–8. 4. Jameson, Antinomies of Realism , 35–6. 5. Ibid., 31. 240 K. CAVENDER ET AL.

6. Ibid., 37. 7. Contemporary affect theory operates on the assumption that “affects must be noncognitive, corporeal processes or states” (Leys, “The Turn to Affect,” 437). 8. Jameson, Antinomies of Realism , 35–6. 9. The projects mentioned here are not meant to offer a comprehensive index of innovative or important digital work with or on Ulysses , but only to demonstrate the conceptual and methodological diversity of such projects. Jessica Pressman’s Digital Modernism: Making It New in New Media (Oxford University Press, 2014) offers another such catalog, and shows how these projects often challenge our expectations regarding what constitutes meaningful literary information. 10. Ulysses by James Joyce . Project Gutenberg Aug. 11, 2014. Amanda Visconti. Infi nite Ulysses . Maryland Institute for Technology in the Humanities. 2012. Robert Berry. Ulysses Seen . Throwaway Horse LLC. 2009. 11. Christie, Alex and Katie Tanigawa. “Scholarly Relevance and Impact.” Dislocating Ulysses . University of Victoria, 2013. Steve Cole. liberateU- lysses . 2011. It’s All About YoU(lysses) . Modernist Versions Project. 2012. Joseph Nugent. JoyceWays . Joyceapps Limited. 2012. 12. C. W. F. McKinna and A. Antonia. “‘A Few Simple Words’ of Interior Monologue in Ulysses : Reconfi guring the Evidence.” Literary and Linguistic Computing 11.2 (1996): 55–66; and “The Statistical Analysis of Style: Refl ections on Form, Meaning, and Ideology in the ‘Nausicaa’ Episode of Ulysses .” Literary and Linguistic Computing 16.4 (2001): 353–73. 13. Mark C. Marino. “Ulysses on Web 2.0: Towards a Hypermedia Parallax Engine.” James Joyce Quarterly 44.3 (Spring 2007): 476. 14. Matthew L. Jockers. Macroanalysis: Digital Methods and Literary History . Champaign: University of Illinois Press, 2013. 25–6. 15. Ann, Devitt and Khurshid Ahmad. “Is There A Language of Sentiment Analysis? An Analysis of Lexical Resources for Sentiment Analysis.” Language Resources and Evaluation 47.2 (Spring 2013): 483–5. 16. Ibid., 508–9. 17. Reyes, et al., “A Multidimensional Approach for Detecting Irony on Twitter.” Language Resources and Evaluation 47.1 (Winter 2013), 260. 18. Devitt, et al., “Is There A Language of Sentiment Analysis? An Analysis of Lexical Resources for Sentiment Analysis,” 482–3. 19. Jockers, Macroanalysis, 30. 20. Because Affectcrawler proceeds from word to word (employing the test- able sequence = word n+17, where the value of n corresponds to the index or enumeration of the word in question and advances through the text in single-word increments), such extended passages represent a block of text in which any 18 word sequence consists of fewer than 1 E word and greater BODY LANGUAGE: TOWARD AN AFFECTIVE FORMALISM OF ULYSSES 241

than 3 each of B and I words. Output takes the form of a separate .txt fi le listing passages that exhibit the desired density of affective language, often sustained across an extended range. 21. As Siân E. White notes, Stephen, when not “repulsed by touch and the body in general,” only “consider[s] touch in his characteristically philosophical and removed manner,” in contrast to Bloom who “is the most bodily of the three main characters.” White goes on to note how the out- wardly lustful Molly’s bodily connections are, at least as presented in the text itself, quite limited, her “ physical exchanges tak[ing] place almost entirely off the page, in her memory, her imagination, or the implied visits from Blazes Boyland.” Siân E. White, “‘O, despise not my youth!’: Senses, Sympathy, and an Intimate Aesthetics in Ulysses. ” Texas Studies in Literature and Language 51.4 (Winter 2009): 503–36. 508, 526 n. 26. 22. As Siân E. White also notes, this image of the hand that “‘limply sees’ … invok[es] pity … because it implies that the stripling’s body is weak or diminished,” thus “bear[ing] the weight of sympathy … as well as pity and condescension in Bloom’s act.” White, “Senses, Sympathy, and Intimate Aesthetic,” 507. 23. For a discussion of how the “Joycean encounters are phenomenological rather than empirical,” see Gerald L. Bruns, “What’s in a Mirror: James Joyce’s Phenomenology of Perception.” James Joyce Quarterly 49. 3–4 (Spring-Summer 2012): 573– 588, 573. 24. As Erwin R. Steinberg also observes in “‘Lestrygonians,’ A Pale ‘Proteus’?” Modern Fiction Studies 15.1 (Spring 1969) 73–86. Steinberg traces Stephen’s refl ections on the senses back to Artistotle’s De Anime (73). For Steinberg, Stephen’s response to the senses is preferable to Bloom’s: “Stephen’s thoughts here, stemming from the academic, are abstract and often unworldly—but sophisticated; Bloom’s thoughts, stemming from the mundane, are practical but often unaware and misinformed and generally naïve” (74). We wish to argue the opposite: that Stephen’s cerebral approach blocks his experience of connection to others and thus empathy. 25. In “The Cat’s Meow: Ulysses, Animals, and the Veterinary Gaze,” James Joyce Quarterly 46. 3–4 (Spring-Summer 2009) (529–43) David Rando argues that in the latter nineteenth century there evolved what he calls the “veterinary gaze,” shaped by both the “revolution in perception [which] took place with respect to the bodies of animals” and the “language of rationality.” Rando argues of Calypso that “Bloom has absorbed and been shaped by resulting changes in the perception of animals, and there are many instances in the text in which he refl ects the cultural shift by looking at animals through veterinary eyes” (530). We feel it is important to point out that Bloom does not always look at animals through veterinary eyes in Calypso: as we try to make clear, he also sees himself through animal eyes. 26. Jameson, Antimonies of Realism , 59–66. CHAPTER 11

“We Twiddle … and Turn into Machines”: Mina Loy, HTML and the Machining of Information

Andrew Pilsch

In addition to other accolades showered upon her by modernist scholars, Mina Loy—avant-garde poet, sculptor, lover of prominent Futurists— is typographically one of the most innovative and important fi gures in modernism. In this essay, I discuss these typographic innovations in three phases: fi rst as an anticipatory, critical fi gure of the emergence of an informational machine post-WWII; second, in the context of that informational machine as it shaped HTML (HyperText Markup Language, the basic markup language that structures all content on the World Wide Web); and third, as a fi gure for thinking through avant-garde practices of reading and writing in the present. I further tie these three phases of Loy’s critical, machinic poetics to the process of digitizing her work for circulation on the Internet in order to show how Loy’s brand of modernist praxis continues to inform our ways of thinking about digitization of text and the digital transmission of information. I run an online archive of Mina Loy’s poetry 1 —primarily editions of the Futurist canon identifi ed by Janet Lyon in Manifestoes: Provocations of the

A. Pilsch () Texas A&M University, College Station, TX, USA e-mail: [email protected]

Modern 2 —that strives to maintain, as accurately as possible, the typographic properties of Loy’s poetry. The narrative of how I (a rhetorician by training) became interested in Loy is suggestive of why this is important. I fi rst encountered Loy’s poetry when I asked a colleague to cover a class in a literature survey I was teaching. She suggested teaching Loy’s “Parturition,” and I agreed. That semester, I was teaching without an assigned anthology, so I linked to a version of the poem hosted on a free poetry website and thought little more of the assignment. When I returned and asked my colleague about class, she informed me that it had been a disaster because the copy of the poem I had assigned misrepresented the spacing. The students could not make sense of “Parturition,” including its complex deformation of subjectivity, because a major feature of the poem was missing. In “Parturition,” as I discuss more in depth below, the consciousness of the speaker, experiencing the pains of childbirth, begins to shift outside the confi nes of her body. To indicate this slippage, Loy breaks up the lines of her poems into textual fragments that bounce around the echo chamber of her narrator’s suffering and ultimately fi nd peace through a connection with a cosmic humanity. 3 An example of this fragmentation and its connection to modernism’s non-Cartesian understanding of human limits can be seen in stanza three of the poem:

Locate an irritation without It is within Within It is without4

The oscillation between “within” and “without” is central to the oscillation between self and other that leads to a cosmic perspective of one- ness in the “cosmic initiation” that the narrator fi nds in giving birth, but equally important (perhaps even more so) is the spacing that indicates this oscillation. 5 By fragmenting the line as she does, Loy indicates this movement beyond the self. Without the spacing, the poem is rendered signifi - cantly less radical, curtailing the performative quality of its typesetting:

Locate an irritation without It is within Within It is without “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 245

Minus the spacing, this selection is appears to be nothing more than a banal New Age platitude; however, offset as it is in print, the poem has a much greater force than a slogan one might fi nd on a poster in a yoga studio. Thus, losing the typesetting in the online circulation of Loy’s work is a massive disservice to an important modernist poet. As Loy’s critique of Futurism became more central to my own work on the origins of transhumanism and the rhetoric of evolutionary futurism, this issue with Loy’s poetry online increasingly came to bother me. My dissatisfaction with how this poorly set poem had shaped my students’ understanding drove me to dig deeper into what was lost by the sloppy digital anthologizing: the Futurist model of consciousness modeled by typography and typesetting. As such, I set out to produce an archive of Loy’s poetry that would maintain proper the proper typographic spacing. However, as I discovered in assembling this archive, the nature of HTML resists—even prevents —the easy introduction of this kind of typographic experimentation, despite, as Alan Liu has argued in The Laws of Cool , the fact that much of what counts as cool in web design is explicitly informed by modernist typography. 6 From this insight, I have come to realize that Loy’s typography is an important counterpoint to the dominant (and masculinist) culture of modernist typography. As her poetry was lost before its later recovery, so has the critique of this gendered typographic method. More specifi cally, Loy’s spaces, though important to the particular meanings embedded in each of her poems, are equally important to the narrative these poems tells about the changing nature of the machine as the focus on laboring bodies gives ways to informational machines and prefi gures the rise of computation. In digitizing Mina Loy online, I discovered that her typographic practices resist the informational rationality and linear, stable reading subject embedded in both print textual and online hypertextual practices—despite the revolutionary rhetoric with which the latter is greeted. I conclude that, in addition to recouping Loy as an important feminist critic of Italian Futurism, we must also work to understand her as an important critic of the machining of information and the rationality it implies.

LOY’S SPACES: MACHINE, INFORMATION, DISCOURSE The portion of Mina Loy’s poetry I have digitized for my archive is from a period in which Loy increasingly breaks from, as Janet Lyon, argues, the implied “republican motherhood” of Italian Futurism, in which it is the 246 A. PILSCH duty of Futurist women, as Valentine de Saint-Point argues in “Manifesto of Futurist Women,” to raise and nurture genius in their male children. As Lyon argues, Loy instead articulates a position that is critical equally of this servile function and of the toxic conception of male genius that creates it. Her poems are heterogeneous and deliberately construct gender as a networked operation in which multiple perspectives fuse to create a cosmic understanding of humanity. In Lyon’s reading of Songs to Joannes , this cosmic humanity emerges from the many viewpoints enfolded into the poem and the seeming exclusion of Joannes from this cosmic perspective. 7 As Lyon draws out, in the course of the songs:

we are shown the fragments of a discontinuous sexual and intellectual relationship; but we are also shown alternative fragments, equally discontinuous, of a relationship that might have been. In short, we are made to see, almost immediately, that “what is” is only one of several sets of charged fragments, the presence of which undermines any defi nitive claims to comprehensive representation. 8

This multiverse approach to poetics, not only revealing multiple “real” perspectives but also encompassing the virtual or the hypothetical, breaks up the notion of the singular genius into a tissue of subjectivities and positions, both human and non-human (as with the mother cat in “Parturition”). This heterogeneous approach to subjectivity that, as Lyon argues, militates against the aggressively masculine positions of singular subjecthood evinced in the work of the Futurists Loy was associating with during this period. In Poem IX of Songs to Joannes , for instance, Loy suggests that the narrator and Joannes’s love opened their eyes to “A cosmos/ Of coloured voices,” but one that Joannes cannot appear to see. 9 Describing the pair’s lovemaking, Loy writes in Poem XI “Dear one at your mercy/ Our Universe/ Is only/ A colorless onion” while refl ecting on Joannes’s “dis- heartening odour.” 10 She contrasts this smelly physicality to the metaphysical or cosmic potentials of their love, and this oscillation is at the core of the feminist argument Loy makes in these poems: the physical gross- ness and seemingly boorish behavior of Joannes occludes his access to this larger sphere of cosmic becoming. And this same behavior is a result of, not in contrast to, the cult of male genius that animates Futurism. One could imagine a humorous parody of the futurist genius in which his “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 247 smelling of onions contrasted with the provocations of his writing, commenting on the foibles of the intellect in a way similar to the pretentious, posturing male artiste narrating Katherine Mansfi eld’s “Je ne parle pas français.” However, what Loy does in Songs is so much more devastating and ultimately tragic: by documenting the world that Joannes has no access to, we see how truly impoverished his limitations make him and, by extension, mainstream Italian Futurism. Loy’s typopgraphic decisions in these poems further underscore this fragmenting, fractal approach to the space of reality and Joannes’s blindness to it. In Poem XIII of the cycle, the desires of the ego for individual boundaries and the desires of eros for continual blending beyond borders of the self collide spectacularly, but this specifi c struggle is also enforced by the spacing of the lines:

Come to me There is something I have got to tell you and I cannot tell Something taking shape Something that has a new name A new dimension A new use A new illusion

In the fi rst two lines, which divide into two fragments each, we see a collision of two possible conversations that argue against one another, attempting to capture the complexity of a thought that cannot be accurately expressed using the syntax of standard English. From this internal controversy, the poem spills over into a fi ght between the two lovers:

It is ambient And it is in your eyes Something shiny Something only for you Something that I must not see It is in my ears Something very resonant Something that you must not hear Something only for me11

Both “you” and “me” in this poem have experiences that the desire to keep separate, despite the continual push toward a cosmic togetherness Loy fi nds suggested in experiences of both love and sex. Thus, as Poem XIII continues, “Let us be very jealous/ Very suspicious/ Very conservative/ Very cruel,” the fi ghting between the two fi gures is constituted 248 A. PILSCH as a contest between to “inviolate egos.” 12 The spaces in these stanzas continue to fragment consciousnesses, working upon both characters in contact with another and with the “ambient,” thus implying an oscillation not only between the two people but their own internal desires and the demands of the setting of the fi ght itself. Thus the space inside the Loy’s lines not only suggest dialogue but the modernist fragmentation of subjectivity into a polyphony of competing internal and external shards celebrated by her peers in Futurism. Thus, one of the most challenging aspects of Loy’s poetry from this period is attempting to sort out the various speakers and where the individual fragments, often within a single line, are coming from. For Loy’s cosmic vision; however, it ultimately may not matter. Poem XIII concludes on a note of unity:

we might tumble together Depersonalized Identical Into the terrifi c Nirvana Me you—you—me13

Loy suggests that in the end, the experience of the totality which she labels Nirvana here (and calls “god” a stanza above) is the ultimate goal of the poems. Dizzying in its polyvocality, Songs to Joannes uses a fragmented typography to document the tensions surrounding but the ultimately positive affordances of adopting a post-egoic, cosmic perspective. This experience of ego-death is in sharp contrast to the handling of this topic by more orthodox Futurists, such as F. T. Marinetti. In “The Foundation and Manifesto of Futurism,” point 11 of the manifesto begins by suggesting that Futurism “shall sing of the great multitudes who are roused up by work, by pleasure, or by rebellion.” 14 Marinetti goes on to document “the many-hued, many-voiced tides of revolution in our modern capitals” with a now familiar array of futurist tropes (“violent electric moons” hover over “arsenals and shipyards” while “railway stations” are “voraciously devouring smoke-belching serpents”). 15 This evocation of multitude is not the same as the polyvocality discussed by Loy. If nothing else, the militaristic imagery Marinetti uses is not a vector of understanding; as Marinetti intones, “art … can be nothing but violence, cruelty, and injustice.” 16 Where Marinetti’s version of Futurism seeks to sing of a violent, seething multitude always on the brink of explosion, Loy offers a “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 249 vision of identifi cation and experiential peace, a differing response to the provocations of the avant-garde modernist stance of subjectivity. In Poem XXV, Loy strongly differentiates her vision of a post-ego polyvocal cosmos with the violent multitudes of Marinetti when she connects the idea of cosmic unity to a becoming machine. In this poem, an unidentifi ed group approaches the sun dawning over the Arno river: “We twiddle to it/ Round and round/ Faster/ And turn into machines.” 17 In this machinic transformation, “the sun/ Subsides in shining/ Melts some of us/ Into abysmal pigeon-holes,” but in this transformation, there are those who are not thus destroyed: “Some few of us/ Grow to the level of cool plains/ Cutting our foot-hold/ With steel eyes.” 18 Loy’s usage of the word “plain” here implies an expansion, a spreading-out— a geography of becoming—and this specifi c becoming is a particularly odd notion of the machine: grown to expand across the geography, free- fl oating, dotted “with steel eyes.” Similarly, in Poem XXVI (“We sidle up/To Nature”), 19 these machines that “we” become are infi nite and seemingly shapeless. This idea of the infi nite, plain-sized machine contrasts to earlier machinic imagery in Songs to Joannes . For instance, in Poem II, Joannes is introduced as fi rst a “skin-sack” and then as “Something the shape of a man/ To the casual vulgarity of the merely observant/ More of a clock- work mechanism/ Running down against time.” 20 By describing Joannes as a clock-work mechanism in the shape of a man, Loy is calling specifi cally to his membership in the Futurists and to their specifi c usage of machinic imagery, as the notion of clock-work is more familiar to the modernist imaginary and the understanding of “machine” extant at that time. Hal Foster has argued that this modernist machinic imaginary is primarily haunted by a “double logic of the prosthesis” in which an advocated return to a natural body or the celebration of a monstrous machinic body are both determined by the shattered bodies of WWI veterans and mangled industrial workers. 21 As Foster writes, tracing the function of this double logic through the fascist, machinic aesthetic of F. T. Marinetti and Wyndham Lewis:

However grounded in the damaged body of the worker-soldier, this fantasy remains psychic, internal to the subject, in particular to masculine subjects prone to fascism like Marinetti and Lewis. For this subject the body in pieces represents the energies that it cannot bind, that it sees as fragmentary, fl uid, 250 A. PILSCH

feminine. With Marinetti the relation to this unbound body is often ecstatic; he likes to imagine his body exploded in this way. 22

In Foster’s formulation, this fascist aesthetic articulates “the machine as a castrative trauma and as a phallic shield against such trauma” and one that male artists must militate against. 23 Under this reading of the modernist avant garde, this crisis of masculinity is at the heart of modernist aesthetics and is, in both Lewis and Marinetti, turned into Fascism. This formulation of Marinetti is specifi cally interesting for thinking through a machinic understanding of Loy’s poetics. Foster argues above that Marinetti celebrates the supposedly “feminine” position of a fragmented and fl uid identity; however, Loy specifi cally rejects Futurism’s deluded bluster about male genius because of its lack of true engagement with this fragmentation. Moreover, as we saw above, Marinetti’s celebration of this supposedly feminine position, in the “Manifesto,” is specifi cally routed through quint- essentially male tropes of violence, explosions, death and warfare. Thus, by combining Foster’s point about this feminine position with the discussion of Songs to Joannes , we can see Loy’s collection of poems as a rebuke of this common modernist understanding of the machine. As Foster argues, works derived from the modernist spirit of the avant garde 24 construct the machine as “a demonic supplement, an addition to the body that threatened a subtraction from it.” 25 Such constructions rely Karl Marx’s formulation of the machine in Capital , but I do not think this is the same machinic imaginary we see at work in Loy’s Songs . When Loy introduces Joannes as “a clock-work mechanism/Running down against time,” she highlights the antiquated nature of his vision of the machine (which was, as Foster reminds us, extremely radical at the time) by connecting this stereotypical Italian Futurist to a historic curio, such as a clockwork automaton. 26 Instead, Loy’s vision of the machine (“fragmentary, fl uid, feminine”) is more in line with the portion of Marx’s Grundrisse labelled “The Fragment on Machines.” 27 , 28 Unlike the discussion in Capital , in which workers are alienated from their labor through the intervention of machines, the “Fragment” is revolutionary for showing the smooth integration into a totality, a totality that comes to absorb the social, that the machine facilitates in the workplace. Marx writes of the machine in this piece:

Labour appears, rather, merely as a conscious organ, scattered among the individual living workers at numerous points of the mechanical system; sub- sumed under the total process of the machinery itself, as itself only a link of “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 251

the system, whose unity exists not in the living workers, but rather in the living (active) machinery, which confronts his individual, insignifi cant doings as a mighty organism. 29

In this vision of the machine, the workers may be alienated from their own labor, but, more importantly, the machine comes to be the ultimate organizing, cognitive agent. As Marx writes, “the machine which pos- sesses skill and strength in place of the worker, is itself the virtuoso, with a soul of its own in the mechanical laws acting through it.” 30 In this way, Marx shifts the emphasis of the term “machine” to the organization of consciousness and virtuosity that develops within the network of humans and individual pieces of equipment comprising the hybridized agent of this model of a machine. While consideration of this particular portion of Marx’s thought would not become central to Marxism until long after World War II, it documents the fi rst stirrings of a machinic imaginary that bypasses the masculine avant gardes of modernism. This vision of the machine as a kind of social network is decorporeal and therefore capable of tying together multiple subjectivities into a functional whole that exceeds their individual capacities. For Marx, this dehu- manization of the laboring body is wholly irredeemable, but, I argue, Mina Loy utilizes this specifi cally new vision of the machine in the kind of fragmentary network consciousness she documents in her poems critical of Italian Futurism. When she contrasts the “clock-work mechanism” that is Joannes to the machines that the poem’s mysterious “we” become in order to “grow to the level of cool plains,” she is suggesting that the mechanistic model of the machine is not nearly as revolutionary as the Futurists she is rebuking would seem to think. Instead, Loy seems to suggest that this cosmic machine—this network of affect and information—is the truly revolutionary understanding of the word “machine.” Thus, the typesetting that makes possible this understanding, lost in extant online editions of Loy when I began my project, is crucial to understanding the machinic culture she anticipates; at the same time that that machinic culture nullifi es the force of this insight. This process of decorporealizing the idea of the machine would become increasingly relevant to our present moment when, twenty years after the publication of Songs to Joannes , Alan M. Turing outlined the thought experiment that would come to defi ne the digital computer. In “On Computable Numbers,” Turing suggests a thought experiment useful for calculating “computable numbers,” which he defi nes as “real numbers 252 A. PILSCH whose expressions as a decimal are calculable by fi nite means.” 31 Asking readers to “compare a man in the process of computing a real number to a machine which is only capable of a fi nite number of conditions,” he then launches into his groundbreaking description of this machine, including its tape and the means by which this infi nite tape might store manipulable symbols. 32 Turing argues that such a machine might be used to calculate a whole range of solutions, but the under-appreciated, yet crucial, takeaway from Turing’s description is that this machine is not, and never has been, real. The abstract machine Turing describes here is the idealized model of the real work a modern computer performs everyday; however, the actual description offered in “On Computable Numbers” could never actually exist (it depends on an infi nite paper tape, for instance). This machine is only meant to stand as a thought experiment, a model for modeling other real machines, including the code-breaking equipment Turing helped design, build and operate during WWII. I end this discussion of Loy’s critique of the modernist machine with Turing’s abstract machine because the real computers described by his abstract machine have come to in turn defi ne our current moment. What Loy, Turing and the Marx of Grundrisse all share is a belief that a machine is not defi ned by a single mechanism or constrained within a single technical object. As much as Marinetti celebrates the speed of the motorcar in the “The Founding and Manifesto of Futurism,” his account of being “hurled along roads as deep and plunging as the beds of torrents” leads to a feeling of being “vital and throbbing.” 33 However, Marinetti’s experience of absolute autonomy and phallic empowerment, crucial to the rhetoric of car culture throughout the twentieth century, ignores the deep enmesh- ing in a petroecological that this putative autonomy requires. The web of paved roads, petrochemicals, laboring bodies in automobile factories, and the various geological records shaped into the body of the car all shape Marinetti’s experience of himself as an autonomous agent. The abstract, decorporealized, network machine that we see in Loy (not to mention Turing and the Grundrisse ) contrasts Marinetti’s limiting understanding of the machine as mechanism with this more radically cosmic understanding of the machine as an ecology. From this ecological or networked perspective on Loy’s machinic poetry, I turn in the next section to the fate of this poetry online. As I have suggested, Loy’s spaces, so crucial to articulating this emergent understanding of the abstract machine, are extremely diffi cult to contain in HTML, the language of information exchange on the web. I show how “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 253 her critique of the machine as mechanism, along with its origin in the cult of male genius in Futurism and its embodiment through typesetting, are still just as crucial to resisting the oppressive rationality of digital life.

NON-BREAKING SPACES: HTML, TYPOGRAPHY AND THE &NBSP Turning from this fragmentation of space and its relation to the fragmented (or cohesive) reader in Loy’s poetics, I want to more specifi cally focus on the creation of my Loy archive and the challenges her method presents for creating HTML documents. As I argue in this section, HTML is designed for two regimes of textual arrangement. In one, the one I call the “linear-textual,” a document is marked up in the classic sense of the word, with features of the text described by a series of computer-readable tags delineated by less-than and greater-than symbols (< >). In the other, which I call the “patch-work,” text is divided into regions that are positioned and styled, as chunks, by CSS (Cascading Style Sheets). As I explain below, these two regimes of textuality result from HTML’s origins as a markup language for writing computerized documentation and the focus on creating linear documents. This linearity is in contrast to the rhetoric of hypertext that partly underscores the language’s design. In “A File Structure for the Complex, The Changing, and the Indeterminate,” the 1965 paper in which Theodore H. Nelson fi rst coined the term “hypertext,” Nelson articulates a vision of a new kind of fi le system “for personal fi les and as an adjunct to creativity … wholly different in character from those customary in business and scientifi c data processing.” 34 Such a fi le system would need to be asso- ciational, idiosyncratic; as Nelson writes it would need to “provide the capacity for intricate and idiosyncratic arrangements, total modifi ability, undecided alternatives, and thorough internal documentation.” 35 The system of linked, or in Nelson’s term “zippered” lists, draws on theoretical work done in the 1940s by Vannevar Bush on the Memex (fi rst described in an extremely infl uential essay in The Atlantic entitled “As We May Think”) and comes to provide the central organizational metaphor for the World Wide Web as we know and use it today. 36 Most curious, however, is the mutation in Nelson’s thinking as he introduces hypertext to his audience at the Association for Computing 254 A. PILSCH

Machinery conference. After describing the basic features of hypertext, he suggests a little of his project’s history:

The original idea was to make a fi le for writers and scientists, much like the personal side of Bush’s Memex, that would do the things such people need with the richness they would want. But there are so many possible specifi c functions that the mind reels. These uses and considerations become so complex that the only answer is a simple and generalized building-block structure, user-oriented and wholly general-purpose. 37

As with so many discoveries, a seemingly simple problem—how to make a fi le system for individual people instead of machinic corporations—begins to suggest larger systemic shifts. From an image of novelists and absent- minded professors, Nelson’s vision moves to libraries, movies and large- scale programming projects, fi nally suggesting that, because of the insights hypertext grants, “it is almost everywhere necessary to deal with deep structural changes in the arrangements of ideas and things.” 38 What is most directly relevant (and perhaps worrisome for some literary scholars and writers of literary fi ction) in Nelson’s original paper on hypertext is the way the fi gure of the writer slowly drops out of his account. Originally designed so that novelists and individual researchers might better organize their ideas in idiosyncratic ways, hypertext is modeled on Nelson’s observations of writers at work: “Many writers and research pro- fessionals have fi les or collections of notes which are tied to manuscripts in progress. Indeed, often personal fi les shade into manuscripts, and the assembly of textual notes becomes the writing of text without a sharp break.” 39 However, Nelson soon realized that he had invented a paradigm that, when computerized, changes the way information is organized and interacted with by humans. Nelson’s observations of the writing process are relevant for modernist critics due to their apparent ties to a codex-based textual economy. In the image of the writer’s fi le collection, the ideas, often written on note cards as Nelson points out, are reasonably complete thoughts. The textual fragments here are still more or less cohesive textual units that can be arranged into some semblance of a text. Moreover, this arrangement is the product of a very stable understanding of a singular author. Yet, given modernism’s many critiques of such an understanding, how would something like Mina Loy’s practice of textual fragmentation spilling over into cosmic consciousness factor into Nelson’s vision? “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 255

Working with Mina Loy’s poetry in HTML demonstrates the problematic foundations of Nelson’s concept of hypertext. It also underscores the fact that hypertext, for all of its legitimate paradigm-breaking power, still basically functions with an idea of unifi ed textuality. At the level of source code, for instance, when HTML renders—converting text marked up in HTML into the displayed text we interact with in our browsers—the rendering program condenses all white space (no matter how many line breaks, tabs, spaces, etc.) into a single space. So, for instance, something that looks like:

Document title Document Author

Will be converted to:

Document title Document author

before being rendered on screen by a web browser. In HTML, all consec- utive whitespace characters are converted into a single space before being displayed. While there is nothing inherently wrong with this approach, it underscores the fact that HTML, as a regime of text, does not care about space. Further enforcing this position is, as philosopher Brian Willems suggests in “An Accidental Imperative: The Meaning Presence of ,” the widely held belief in web design communities that the non-breaking space character in HTML ( ) is a fi gure of “lawlessness.” This belief alludes to the denigrated practice of using this character to add space to HTML before the emergence of CSS positioning in the late 1990s and early 2000s. 40 In the technical whitepapers that govern the language, the condensation of whitespace by HTML renderers is called “breaking,” and the character creates a space that resists this breaking. As Willems details, this character, for instance, can be used to prevent the guillemets (« »; used in French to indicate quotation) from wrapping onto the next line, as those characters always have a space between them and the beginning or ending of a quotation. 41 However, the has a “lawless” quality because this non-breaking behavior was frequently exploited in the early days of web design to position elements a certain number of spaces from the left or right margin (by using a cluster of commands to create horizontal white space). 42 256 A. PILSCH

Using the as a design element runs in contrast to the development of HTML as a data description language. However, at the time that became used for positioning elements—in other words, the moment that web design emerged as a distinct discipline—the language lacked the facility for sophisticated arrangement of text. In order to implement Nelson’s vision of hypertext in an early version of the Internet, HTML was developed by Tim Berners-Lee while at CERN (the European Organization for Nuclear Research) to provide technical undergirding for his vision of a worldwide hypertextual web. Lee based this set of commands on an SGML system (Standardized General Markup Language) that CERN used in-house for documentation. SGML, an ISO standard for defi ning markup languages and, outside of HTML, is primarily used for writing hypertextual documentation systems for technical products. This kind of hypertext is similar to reading the Help documentation that ships with a program such as Microsoft Word. At this time, HTML was primarily focused on marking up documents that might resemble these documentation systems, allowing CERN’s scientists to easily share data with colleagues at other research centers. These early versions of HTML offer a linear-textual perspective on document design, with most tags (other than the hyperlink tag that actually linked documents online) focusing on standard document elements (paragraphs (

), tabular data (

), quotations (

), and computer code () to name but a few). An example of this model would be the many texts digitized Project Gutenberg, which just have long, single columns of text as their content. Despite the hyperlinks that make web browsing possible, these documents defi ne a single stable text that may (and often should) be read in a linear fashion. For the most part these early versions of HTML focused on producing technical documents: documentation, reports, white papers, etc. As mentioned above, however, once the web moved beyond the research centers that it originally linked, people increasingly began designing documents in more complex fashion, especially once the idea of web commerce emerged. The tag, in early HTML, was the primary option available to design-oriented web content producers to position elements on the screen in an aesthetic rather than purely functional manner. However, with the increasing sophistication of Cascading Stylesheets (CSS) for handling layout, the need to use was both discouraged by the World Wide Web Consortium (W3C) that governs the HTML and CSS standards and largely abandoned in favor of easier, CSS-based “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 257 solutions. This combination of CSS for aesthetic design and HTML for functional description of content caused a second textual model to emerge in the HTML standard, the patch-work model. In this latter approach to web textuality (Pinterest is a good example of an extreme application of the patch-work model), HTML block elements (any element in HTML that causes a line break at its end is a block element) cause things such as paragraphs, lists and the generic HTML block element to be abso- lutely positioned anywhere on the two-dimensional coordinate plane of the user’s screen. While this patch-work model of allows for a whole range of stunning and sophisticated websites (and underscores most of the modern web), it simply allows a web designer to position linear-textual blocks of content on the page as needed. As I discovered while converting Mina Loy’s poetry into HTML, the markup language allows no lawful way to fragment linear-textual blocks of content in the manner that is so essential to Mina Loy’s art. Moreover, the kind of rational, linear, limited perspective Loy associates with boorish masculinity in Songs to Joannes is literally encoded into the very structure of HTML itself. In converting her poems, I ended up using large numbers of entities to create the proper line spacing that structures her poetry. The way Loy employs space in her poetry forces a digitizer to engage in the unlawful. The spaces of her poetry, like many modernist works, resist linearity, but her typographic appeals to a cosmic subjectivity also, as I show below, resist the stable understanding of a textual unit implied by the seemingly avant-garde possibilities of the patch-work model of HTML page design. This recourse to the lawless became necessary because, early in the project, I found myself attempting to fi nd logical units within Loy’s fragmentation. I was, at this point, thinking more like a web designer than a textual scholar: looking for ways to carve up Loy’s poem into block elements I could position using CSS. These units do not often exist in her poetry, though certain passages seem to suggest logical blocks of this kind. Inevitably, sections of the document that appear to be chunkable in HTML’s patch-work model would have a line or two that does not fi t such a scheme or, equally problematically, such chunking would break up the textual units of the stanzas themselves. To illustrate this conundrum, Fig. 11.1 , presents on the left two stanzas from Songs to Joannes . While we, as trained readers of poetry, can clearly recognize these two stanzas, in order to position them using CSS’s patch- work approach to textuality, I would have to break the stanzas into six 258 A. PILSCH Fig. 11.1 Comparison of printed and HTML block stanzas for Songs to Joannes chunks, which have little coherence with or attachment to the logical stanzas of the poem. As such, to divide the poem into units and use the lawful design vernacular of HTML, I would have to violate the coherence of the text’s organization. While this is enough of an affront to the coherence of the poem, it further violates HTMl’s origins in the coherence suggested by Nelson’s original hypertextual metaphor of the card catalog. In this way, I argue, Loy’s poetry constitutes a crisis within the rationality of hypertextual systems, a crisis her poetic method anticipates in its original context of futurist critique. AVANT-GARDE SPACES: MODERNIST MARKUP AT THE LIMITS OF HUMAN PERCEPTION As Alan Liu argues in The Laws of Cool , much of what passes for “good” web design today is, essentially, the same as the laws of design invented by modernist avant-garde movements and then codifi ed into “the generalized Swiss or International Style that dominated business advertising and ‘corporate identity’ campaigns in the 1950s through at least the 1970s or 1980s.” 43 In Liu’s account, the avant garde invented a style of design based on rhythm, contrast and the internal demands of a single piece of content—a style that contrasted with the older Victorian model in which everything in a design was centered against the middle of the page. This new style of design, now internalized as merely “good” or, to use Liu’s key term, “cool” graphic design, was revolutionary at its time. Following computer scientist Lev Manovich, Liu goes on to suggest that “cool” websites reproduce much of this avant-garde logic, absorbing this model of spatial relations in building the avant-garde web. In other words, Liu suggests, web design is an aesthetics of information “mate- rialized” and “naturalized” (Manovich’s words) from modernist avant- “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 259 garde aesthetics. 44 Thus, the two textual models we see in HTML (the linear-textual and the patch-work) both emerge from this modernist quest for the internal logic of a specifi c content. As Liu explains, for all of its fragmentation and radical content, modernist design, ultimately sought to create, through very creation of designer as a profession, a “single perspective … charged with overseeing the total form of the composition.” 45 This functional approach (in which design served content fi rst and fore- most) stands in opposition to Victorian modes of design, which privilege a central textual axis. As Liu goes on to explain, “only the functional relation of form to content could rationalize the bolding, upper-casing, or placement of a word here as opposed to there while keeping sight of the need to coordinate effects around a single, desired impact.” 46 However, Mina Loy’s poetic usage of spaces, as we have seen, does not conform to this “cool” model of design. The layout of texts such as “Parturition” specifi cally agitate against the very idea of a “single perspective” or a “total form,” as Liu suggests was the goal of modernist design. Consequently, I argue that one of the results of the shared lineage of web and modernist design ideologies is the unrecognized retention in HTML of the kind of male genius Loy’s poetry critiqued in male Futurists’ vision of the machinic body. The diffi culty I had rendering her poetry in HTML therefore demonstrates not only the radicality of Loy’s style, but also the obstacles in making “lawless,” non-hegemonic texts available online. Unsurprisingly, then, one growing thread in feminist analyses of online technology addresses this extension of the deeply troubling gender politics of Italian Futurism into the Internet. Writing for The B2 Review , boundary2 ’s blog, educational technology expert Audrey Watters suggests that the continuation of mansplaining 47 online and the growing harassment of women and minorities in these online spaces results from a structural bias toward white men at the core of the technology itself. She writes that: there is a problem with computers. Culturally. Ideologically. There’s a problem with the internet. Largely designed by men from the developed world, it is built for men of the developed world … Despite all the hype and hope about revolution and access and opportunity that these new technologies will provide us, they do not negate hierarchy, history, privilege, power. They refl ect those. They channel it. 48 260 A. PILSCH Watters concludes that “neither the internet nor computer technology writ large are places where we can escape the materiality of our physical worlds.” 49 For Watters and similarly minded feminist critics, the problem with the invisibility of women and minorities online is not something that can be easily dismissed. Instead, she argues, the very infrastructure of these technologies encodes a specifi c identity: “a presumption of male- ness, whiteness, and perhaps even a certain California-ness.” 50 We can see Watters’s argument resonating with Liu’s: the perspective of cool web design is the modernist cult of the designer—which is also, I argue, the cult of male genius that Loy agitated against in her poetry. In each of these modes, the singular perspective of a stable genius-subject is the causeway through which aesthetic experiences of information travel. In infl icting modernist design protocols on HTML, we forget that the language is primarily a semantic markup language for describing the shape of information for both human viewers and machinic readers. And when we neglect this fact, this network of subjectivities is replaced by the idea of the singular genius. However, in forcing HTML beyond the patch-work and the linear-textual models of information design, digitizing Loy’s poetry forces us to confront the limits of the stable-subject-as-reader presumed by HTML’s seeming avant-garde textual mode. As Liu suggests, space—in the web and in modernist design—is a means of creating a textual rhythm and a single perspective, but as I have argued, Loy uses space to break up totality rather than to establish it. A single perspective is not her point. This challenge is carried over into the realm of digital design. Using HTML to encode Loy’s poetry is a messy operation not focused around questions of unity. Rather than treat a text as a fi nely tuned clockwork machine (the position on the machine Loy critiques in Futurism), her texts use space to break up the linear nature of masculine informational machines. Loy’s poetry acknowledges the messy, fragmented nature of the machine that would come to dominate the latter half of the twentieth century, during the shift from Marx’s understanding of the machine as a mechanism for alienating labor to Turing’s model of an abstract computational machine. Despite Loy’s recognition of the limited machinic imaginary pro- pounded by the masculine cult of futurist genius, this model underscores, as Watters shows, the deeply suspect cultural logic of the technological systems in which we are enmeshed. We can see that there is an oppressively masculine machinic imaginary operating online that directly inherits from the futurist position on the machine critiqued by Loy. Further, Loy’s more “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 261 cosmically minded vision of the machine—“fragmentary, fl uid, feminine” as Hal Foster fi gured it—is closer to the reality of life online, despite the persistence of a masculine logic that acts as a defense against this other position. 51 Mina Loy’s poetic critique of Futurism and her manufactur- ing of a cosmic machine in opposition to Futurism still stands today as a viable, important, potent and necessary critique of the cult of masculine genius and its imagined perfect machines. NOTES 1. Online at http://oncomouse.github.io/loy . 2. In Manifestoes , Lyon identifi es Loy’s Futurist canon as “Aphorisms on Futurism” (1914), “Feminist Manifesto” (1914), “Parturition” (1914), The Sacred Prostitute (1914), The Pamperers (1915), and Songs to Joannes (1917). 3. Sourced from Cosmic Consciousness: A Study in the Evolution of the Human Mind by Richard Maurice Bucke, the word “cosmic” entered certain vocabularies of the European avant garde (especially Theosophy). Bucke, who turned to Buddhism to better understand the suffering he saw as the directory of a mental hospital in Canada, began to discuss the tension between self and other of Cartesian mind–body dualism as a cosmic unity in which inside is always outside and vice versa. Loy uses this term in this way twice in “Parturition,” both times in stanza 10. See Richard Maurice Bucke, Cosmic Consciousness: A Study in the Evolution of the Human Mind , 1st edn (New York: E. P Dutton, 1901), http://www.sacred-texts.com/ eso/cc/index.htm . 4. Mina Loy, The Lost Lunar Baedeker: Poems of Mina Loy , ed. Roger L. Conover (New York: Farrar, Straus, Giroux, 1997), 4. 5. Ibid., 5. 6. Alan Liu, Laws of Cool: Knowledge, Work and the Culture of Information (Chicago, IL: University of Chicago Press, 2010), http://site.ebrary. com/lib/alltitles/docDetail.action?docID=10431287 , 195–230. 7. Janet Lyon, Manifestoes: Provocations of the Modern (Ithaca, NY: Cornell University Press, 1999), 157. 8. Ibid., 157. 9. Loy, The Lost Lunar Baedeker , 56. 10. Ibid., 56–7. 11. Ibid., 57–8. 12. Ibid., 58. 13. Ibid., 58. 262 A. PILSCH 14. Filippo Tommaso Marinetti, Critical Writings: New Edition , ed. Günter Berghaus, trans. Doug Thompson (New York: Farrar, Straus, Giroux, 2008), 14. 15. Ibid., 14. 16. Ibid., 16. 17. Loy, The Lost Lunar Baedeker , 63. 18. Ibid., 63. 19. Ibid., 63. 20. Ibid., 53–4. 21. Hal Foster, “Prosthetic Gods,” Modernism/Modernity 4, no. 2 (1997): 5–8. 22. Ibid., 9. 23. Ibid., 8. 24. Foster’s discussion of Marshall McLuhan’s debt to and inspiration in Wyndham Lewis is particularly important for constructing a pre- and post- WWII continuum in the machinic imaginary, a connection I will exploit extensively in this essay. 25. Ibid., 5. 26. Loy, The Lost Lunar Baedeker , 54. 27. Generally indebted to the work of the Autonomist Marxists in Italy in the last half of the twentieth century, though specifi cally sparked by the revital- ization of interest in Grundrisse by Antonio Negri in Marx Beyond Marx (published in English in 1991), “The Fragment on Machines”—along with all of Grundrisse in general—has become increasingly important for thinking through the role Marxist politics can play in the twenty-fi rst century. The “Fragment” documents how Marx could have argued for the machine as a force of social organization and a means of more fully inte- grating workers into a global circuit. With the rise of informatics and what the Italian Autonomists call “semiocapital,” this vision, rather than the haunted logic of the prosthesis Foster documents, becomes so much more central to anti-capitalist thinking in the present. 28. Foster, “Prosthetic Gods,” 9. 29. Karl Marx, Grundrisse: Foundations of the Critique of Political Economy , trans. Martin Nicolaus, Reprint edition (Penguin Classics, 1993), 693. 30. Ibid., 693. 31. Alan M. Turing, “On Computable Numbers, with an Application to the Entscheidungsproblem,” Proceedings of the London Mathematical Society s2–42, no. 1 (1937): 230. 32. Ibid., 231. 33. Marinetti, Critical Writings , 12. 34. Theodor H. Nelson, “A File Structure for the Complex, the Changing, and the Indeterminate,” in The New Media Reader , ed. Noah Wardrip- Fruin and Nick Montfort (Cambridge, MA u.a.: MIT Press, 2003), 135. “WE TWIDDLE … AND TURN INTO MACHINES”: MINA LOY, HTML... 263 35. Ibid., 135. 36. Vannevar Bush, “As We May Think,” The Atlantic (July 1945), http:// www.theatlantic.com/magazine/archive/1945/07/ as-we-may-think/303881/ . 37. Nelson, “A File Structure for the Complex, the Changing, and the Indeterminate,” 135. 38. Ibid., 135. 39. Ibid., 135. 40. Brian Willems, “An Accidental Imperative: The Meaning Presence of &Nbsp;” in From a to : Keywords of Markup , ed. Bradley Dilger and Jeff Rice (Minneapolis: Minnesota University Press, 2010), 83. 41. Ibid., 83. 42. Prior to the introduction of CSS and its wider adoption in browsers, there was no way to control for things like element margin or spacing (components of what are called the CSS Box Model). To add additional, horizontal space between two elements (say two links), prior to CSS, an HTML developer would have to use the nonbreaking space to manually insert a certain number of spaces between the two elements. With CSS, the developer could, instead, set numeric values for each elements margins to exactly space the elements without cluttering the HTML document with non- semantic markup. 43. Liu, Laws of Cool , 207 & 196. 44. Ibid., 205. 45. Ibid., 198. 46. Ibid., 198. 47. Mansplaining is the practice, frustratingly common amongst male academ- ics, of explaining basic facts to persons in a perceived lesser position of power (often determined based on the listener’s gender and race) in a manner that, more than answering the question, is meant to enforce the speaker’s perceived position of greater power through aggressive, hostile, dismissive or patronizing language. 48. Audrey Watters, “Men (Still) Explain Technology to Me: Gender and Education Technology,” The B2 Review (April 22, 2015), http://bound- ary2.org/2015/04/22/men-still-explain-technology-to-me-gender-and- education-technology/ , n.p. 49. Ibid., n.p. 50. Ibid., n.p. 51. Foster, “Prosthetic Gods,” 9. CHAPTER 12 CGI Monstrosities: Modernist Surfaces, the Composite and the Making of the Human Form Eunsong Kim In this chapter I will examine contemporary scholarship that focuses on the representational monster in fi lmmaking and CGI (Computer Generated Imagery). Taking up Derrida’s invocation of the function and the political space of the beast —specifi cally, his central question in The Beast and the Sovereign, “[O]f knowing who can die. To whom is this power given or denied? Who is capable of death, and through death, of imposing failure on the super- or hyper-sovereignty of Walten ?” 1 —I will examine how the representational digital “monster” carries forth modernist mythologies and methodologies regarding sovereignty, possibilities and conclusions for capitalism, and the fabricated, idealized human. Though this essay will closely examine CGI scholarship and CGI fi lmmaking, and their connections to algorithmic fi nance, the necropolitics of modernism manifested I would like to thank Fatima El-Tayeb, Page duBois, Grace Hong, Dorothy Wang, Elizabeth Losh, Max Haiven and Joel Nishimura for feedback throughout every stage of this article. Jason Hirata and Shen Yuan Su provided invaluable comments throughout my writing process. Shawna Ross has been the most gracious and supportive editor. E. Kim () University of California, San Diego, La Jolla, CA , USA e-mail: [email protected] © The Editor(s) (if applicable) and The Author(s) 2016 265 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0_12 266 E. KIM as legibility in digital representation and digital humanities scholarship will be my materialist horizon. I will argue that current arguments made concerning computer generated graphics imaging, fi lmmaking and online cultures, ascribe to a positivist understanding of legibility and visuality often inhering the digital humanities. Because CGI remakes shadows and color-corrects bodies to fi t into the frames, and these frames are layered and composited, every frame with the CGI “monster” is a frame with an equally algorithmic driven “human.” The CGI monster is situated to leg- islate the human and to hide the manufactured monstrosity of the representational “human” construction. Yet digital scholarship on CGI refl ects how racial capitalism 2 continues to not be narrativized within modernist frameworks. The legacy of modernist methodologies assures us that we can recognize the human—be it in narrative or in the digital frame. Current digital humanities methodologies cannot attend to what it has deemed illegible—and what it has deem illegible are the material conditions of racial capitalism. Visibility driven analysis is a continuation of an orientalist tradition of the monster/human divide that has been the topic of so much postcolonial criticism since Edward Said. In Joseph Jeon’s “Neoliberal Forms: CGI, Algorithm, and Hegemony in Korea’s IMF Cinema,” 3 Jeon argues that the CGI monster (which he believes is concentrated to the legible monster) and neoliberal fi nance capitalism are linked via an abstract notion of “algorithm.” According to Jeon, the CGI monster is connected to transnational fi nance and the IMF (International Monetary Fund). This is partially true. The digital and transnational capital might constitute a couplet, but the digital—much like capital—is not an abstraction. There is tremendous labor required every step of the way for the construction of digital technology in all aspects of fi lmmaking. What Jeon seems to consistently assume—and what refl ects the author/name position that I fi nd common in contemporary Western digital humanities—is that he, the critic and the writer, can accurately recognize what has and has not been digitally altered, what may or may not be “algorithmically” processed. As a result, his thesis that the algorithm is the recognizable CGI monster might be more useful inverted: all digitalized visual culture is algorithmic processing, including what we believe is “human.” Throughout the chapter I will argue that concentrating the algorithm to the “least human” or the “monstrous” falls in line with articulations of the human implicit and explicit in contemporary scholarship—what Edward Said described as Orientalism 4 and what Derrida described as the construction of sovereignty. A critical interest in its dichotomy should CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 267 attune to questions of racialized and gendered manifestations, as well the labor and material make up in its divide. In addition, to show these assumptions at work in the more intimate context of contemporary modernist scholarship, Jeon’s argument practices the traditions of “Surface Reading” 5 —a practice that modernist scholars such as Heather Love, Stephen Best and Sharon Marcus have advocated for in tending to the surface of the object/text, with no assumption of its depth. Jeon’s thesis and the “surface” methodology he implicitly uses to read CGI replicates the ideologies of Orientalism 6 and such positivist modernist scholarship. Additionally, reducing the question of the algorithm as an allegory of neoliberal fi nancial capitalism to the non-human form effectively erases how the digital, fi lmic human form is created. I will push against Jeon’s thought-provoking argument problematizing these premises—premises so often perpetuated in modernist studies, new media studies and the digital humanities—through a series of counterarguments. First, algorithmic computations are not isolated to the “visibility” of the monster. Second, the computer-generated monster exists because it is a composite, much like every other part of the digital frame: the CGI monster cannot exist alone. 7 Third, all software processing is the production of algorithmic computation; the process of digitalization leaves no trace of the “organic.” Fourth, all contemporary fi lms and digitalized imaging are processed in software and or coded in and therefore all digitalized images are the manifestations of unseeable algorithmic production. Fifth, in software driven post-production, the “algorithmic” is not a novelty but the norm. 8 Sixth, we—the legible human viewers—cannot decipher between the altered and the unaltered digital image. Finally, the recognizable monster solely exists to verify the modifi ed human. With these points in mind, I will argue that legibility centered allegories of our capitalist contemporary moment ultimately evade the structural conditions of marginalized and impoverished communities. In order to critique the function of the digital, algorithm and legibility, I will use a “behind the scenes” special effects video produced by the post production crew working on John Adams , a show produced by HBO (Home Box Offi ce). 9 The clip openly displays how each scene was pre- shot, rendered and composited. I will examine this process to show why it may be more useful to approach both CGI and fi nancial capitalism as a composite frame. Rather than analyzing what we can see and recognize, we should read the digital landscape as algorithmic, and therefore often fully contaminated. To assume that the landscape is algorithmic would not be a 268 E. KIM “novelty,” as “algorithms” are not novelties: they are formulas that makes the screen possible: from the extracted raw material 10 to the formation of hardware, 11 and from all the variegated levels of software 12 processing 13 and development 14 to its eventual waste. 15 Take, for example, the algorithmic processing that goes into image capture technology such as their: sensors, color registers, data transfer processing and render functions. 16 Rather than seeing them as abstract, mysterious, monstrous fi gures, situating them as computations for pre-conceived longings of modernity 17 will be, I argue, far more useful. DE-CENTERING THE REPRESENTATIONAL MONSTER Jeon’s insightful essay uses contemporary South Korean fi lmmaking to defi ne his concept of “neoliberal forms.” Jeon specifi es what he calls South Korean “IMF Cinema” and provides an example of these “neoliberal forms,” as representations where the invisible (the IMF, economic crisises, fi nancial algorithms) are made visible (as monsters). Jeon states that these fi lms/neoliberal forms “offer allegories of American-Korean relations at its juncture—relations of capitalism, of late empire, and of late (and now strained) partnership in massive cycle of accumulation— through the optic of digital production” (88). Though Jeon emphasizes the American-Korean relationship (hereafter, SK-US), his argument concerning algorithmic fi nancial capitalism is a consciously transnational argument, therefore linking contemporary South Korean cinema incorporating CGI with the ascendency of transnational approaches in current digital humanities scholarship. His argument reads the SK-US economic and military relationship as allegoric to the condition of neoliberal fi nancial capitalism, witnessed in expressions of digital fi lmic production. I would agree that the SK-US military and its economic relationship will always be useful in understanding the networks of neoliberal transnational capitalism. 18 However, in the analysis what makes these fi lms neoliberal forms is not the CGI “monster,” but the modifi ed “humans” the composited landscapes. In such scholarship, there is a tradition of focusing on the legible monster, 19 as the monster is a fi xture of linear, modernist narratives. Jeon writes: From Grendel to Frankenstein’s monster to Godzilla, one strategy that literature and cinema have often returned to is the monster that fi gures an everyday or ordinary social problem in terms that are distinctly out of the CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 269 ordinary. A more refi ned fi gure for the present context is the CGI (computer- generated imagery) monster of contemporary action cinema, which not only represents the anxieties surrounding today’s massive capital fl ows. (88, emphasis mine) While the Western, positivist emphasis on the monster is expected, I would like to push the logic of this routine circulation by pairing it with Derrida’s articulation of the politics of the animal realm. Of the representation of the beast, Derrida argues: Just where the animal realm is so often opposed to the human realm as the realm of the nonpolitical to the realm of the political [...,] the state and sovereignty has often been represented in the formless form of animal monstrosity, in the fi gure without fi gure of a mythological, fabulous, and non-natural monstrosity, and artifi cial monstrosity of the animal. 20 The animal—the representational non-human—serves as the vehicle of social anxiety according to Jeon, and this anxiety, according to Derrida, is the delegated realm of the nonpolitical. This realm, pushed further, might actually be the formal expression of sovereignty as non-human monstrosity, allowing us to grapple with the artifi ce of the representational monstrosity. We can thus situate Gozilla, Frankenstein and the beast in the The Host 21 as the monstrosity of fi nancial sovereignty which produces our ongoing cultural anxiety. However, if neoliberal capitalism is the manifestation of one such anxiety, the CGI human’s ability to destroy it in the form of the beast, sovereignizes the CGI human: the allegory of the CGI monster is then about the transference of power, not its invisibility. If we were to examine Jeon’s allegory of the CGI as the “visible” moment of fi nancial capitalism closely, we could read into its narrative that its political ramifi cations might be, in the near horizon, that neoliberal, fi nancial, algorithmic capitalism is the creation of human intelligence, human materiality, and then, in a further horizon, that neoliberal, fi nancial, algorithmic capitalism can and will be destroyed by the representational human. A project that “unmasks” the invisible without accounting for the altered human is one that refuses to account for the politics of modernist representation and image technologies. The CGI monster-focused allegory is compacted to make neoliberal fi nancial capitalism manageable; it assumes that we can manage its invisibility, into visibility and in its visible form: in short, that we can kill it. Through manmade software, within the 270 E. KIM human imagination, only the human form survives. All manmade problems, man will manage. To chart the modernist tradition of visual representation, to build a narrative of image history and imagining technology, is to chart the development of the altered, idealized human. The complete dissolution of the algorithms involved in idealizing and compositing the human form in Jeon’s argument highlights how trusted the fi gure of the human in digital fi lmmaking has become. Aside from the vast implication of the “beast” and the “monster” in Western narrative and mythology, the development of CGI has a multifaceted function beyond the monster on screen. In inspecting the expansive history of photographic representation, we might be more inclined to believe that CGI—or rather, all imaging technologies—was created to better alter and idealize the human form. Indeed, altering the human to appear more ideal has been the primary function of modernism and visual representation, as witnessed by some of the earliest photo manipulation techniques. In “The Legs of the Countess,” Abigail Solomon-Godeau 22 looks at some of the earliest nineteenth-century daguerreotypes used in self-portraiture. The subjects and photographers utilized various lighting, draping and post-coloring techniques in order to “liven” and “aestheticize” the appearance of the model. Similar to the youth-enhancing, body-contouring lighting techniques, and manipulat- ing the color in the fi nished photograph/fi lm, the ideal representation of the legislated human has been of utmost importance in all image history (modernism) and its technological development. Is it even possible to imagine visual developments that did not account for how the human might appear, interact and remain in the frame? 23 CGI is no different: what good is CGI if the human form cannot be deposited idealized into the frame in order to conquer the monsters of its creation? It is as if say: I am more like you, dear viewer—we dreamed of killing this monster together . If the viewer is expected and demanded to give solidarity to the human form in the screen, it might bemoan us to wonder how we are distinguishing between the human form and the monster—as they exist in precisely the same frame (Fig. 12.1 ). It is important to remember the monster is a narrative construction. The representational digital monster is algorithmically and narratively constructed to die; the representational digital human is algorithmically and narratively constructed to live. The political weight of the monster and its human are derived from their role in the narrative. In this light, Derrida reminds us that Plutarch wrote that fundamentally, “You don’t CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 271 Fig. 12.1 Pierre-Louis Pierson, “La Frayeur” (1861–67) (Reprinted courtesy of the Metropolitan Museum of Art ( http://www.metmuseum.org/ )) fi nd animals begging or pleading for mercy or admitting defeat” (45). Modernist narratives such as Frankenstein , the environment described in Heart of Darkness , and the methodology of Orientialism have constructed the Other, the beast, as fi gures without defeat. Rarely does the representational beast, and the representational other come to an understanding about one’s need for mercy. This is possibly because the beast/other is without consciousness , and without the ability to come into consciousness. The narratively ideal monster is without defeat; the ideal human’s constructions are hidden. Derrida situates that the beast in the fable exists as the vehicle of affective devourment, as a transitional site for the power of The Man. He writes, “[T]he beast is on this account devouring, and the man devours the beast. Devourment and voracity” (46). Devourment and voracity: the representational human exists to rule the representational beast, but the representational human is without the affective and political power it seeks, without the representational beast. The beast exists as the vehicle in which the human form is granted the authority to destroy, kill, devour—in the speed and through the weapons of its choosing. The beast is sliced, 272 E. KIM the monster exploded. The narrative is without trial or remorse: its direc- tive was the transference of devourment and voracity—a task that required the beast, as both the fi gure to be narratively and allegorically destroyed, as well as the visual composite to measure the human besides. In the history of Western visual iconography, we can argue that the monster exists because the protagonist prevails. We are to side with him; we are enveloped in his gaze. Thus we can link the history of image manipulation as technological advancements in modernist narratives to the visual logic 24 of empire, and to the accelerated militarization of modern nation states. Jeon acutely describes the symbiotic links between militarization, fi nancialization and digitalization. The military has fi nanced the majority of action/war fi lms. Jeon writes, “Having many current military applications, CGI was originally derived from military weapons technology—fi rst adapted from analog, anti-aircraft computers—and developed through military-funded research and defense contracts” (90). Jeon describes the intimate relationship between imaging technologies and militarization, but its historical contextualization feels redacted. Could it be argued that almost all modern and contemporary technological developments have been derived from military research and development? 25 Photography, of course, has a long and dense military history; 26 CGI is not singular in that the military industrial complex funded its earliest developments. Rather, the unique history of CGI lies in the shared modern history of military developments that have since become popularized, personalized and turned into user-driven software. CGI is militarized knowledge in that its omnipresence is unknown. 27 It may be more useful to contextualize CGI as part of the industrial military complex, 28 and within modernist image technologies, and to read its development as a relationship to the politics and aesthetics of Western empires. The US military industrial complex has been developing the aesthetics of “whole scale annihilation.” 29 Military technological developments and fi nance can and should always be seen as a symbiotic formation. Bombs 30 do not exist before economic rescue packages, and these do not exist outside of the visual regimes of domination. The technological needs of the US military industrial complex are to prepare the soldier for this command. At the heart of CGI as a military tool and as civilian entertainment is the confi guration of the dynamic visualization of the protagonist human. Without the human in the frame, or controlling the frame (as with video games), the monster, the enemy is of no use to the goals of the military, or, I would argue, to those of digital fi lmmaking. CGI and the history CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 273 of photography are linked in its military trajectories and in how much the technology works to hide its touch, its reach. As militarization and visualization technologies often work to remain hidden structurally, positivist approaches to data visualizations and algorithmic renderings are at best a limited approach. Media theorist Elizabeth Losh has written about feminist approaches to positivist methodologies in visualization projects. In “Feminism Reads Big Data,” Losh 31 examines Lev Manovich’s Selfi ecity to formulate feminist reading methodologies for data visualizations and visualization projects. Selfi ecity collected worldwide selfi es on Instagram to plot out a visual database exhibition to be held in São Paulo, Brazil. In describing the project Manovich situates that Selfi ecity makes possible “social physics” (a term he pulls from Auguste Comte)—where science can be utilized to analyze the atoms (human beings) and output visual data. Losh argues that Manovich’s acceleration for a “social physics” that comprises visible data points, as legible human subjects, to a directly translatable “quantifi able” form is a positivist data visualization scheme. Losh states, “With his analogy to atoms, Manovich also depicts human individuals as discrete elemental particles, which also happens to be a common strategy in visualizing networked relationships to make social graphs more legible” (1649). In order to “make social graphs more legible” to other humans, the human subject (object?) is concentrated to a data point. The visuality of one thing renders all other components (labor performed by Mechanical Turks in the case of Selfi ecity ) fl at. Visuality is performed through the fl attening of complex material, political components. Manovich’s approach to visualization is similar to Jeon’s methodology concerning the CGI monster. Both theorists centralize the act of rendering the invisible, visible—which is a process media theorist Wendy Chun has argued to be the delicate function of software. The spectrum of visibility and invisibility is the predetermined terrain of the interface; to provide us with readings of the additional visibility of this interface (as if to do so lessens the burden/labor of the invisible) is a modernist project, dependent on the notion that legislated human beings might be able to read the interface better, and theorize the visibility of the surface more succinctly. Further, Donna Haraway’s pivotal work, “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective/ Feminist Studies,” 32 addresses the connections between visuality (“human” sight), technology and militarism. She writes, 274 E. KIM The eyes have been used to signify a perverse capacity—honed to perfection in the history of science tied to militarism, capitalism, colonialism, and male supremacy—to distance the knowing subject from everybody and everything in the interests of unfettered power …. Vision in this technological feast becomes unregulated gluttony; all seems not just mythically about the god trick of seeing everything from nowhere. (582) Haraway argues that the eyes (the gaze that personifi es and perpetuates technological advancements) constitute the “perverse capacity” for the “god trick” of “unregulated gluttony.” The a priori for the “technological feast” as well as for the critical examination of ongoing technological feasts lies in vision. The eye is broadened and lengthened—its deregula- tion accepted as foundational insight. Relatedly, Chun suggests that software’s “invisibly visible” (10) condition is expressed as the interface, and linked to contemporary racial formations. Chun argues, “Race and software therefore mark the contours of our current understanding of visual knowledge as ‘programmed visions’” (180). How the “god trick” becomes programmed, circulated and situated both as sight (digital technologies) and sight (scholarship) is in need of further examination. In discussing the role of the beast in modernist discourse, and the Other in the construction of the modern human, I am hoping to display how a linear allegory of the beast as fi nancial capitalism and the humans/ environments as unknown sets up an uneasy relationship between digital representation to fi nancial capitalism. As Derrida suggests, will narrative humans devour further, the visual eruption of the artifi ce? If the beast is an allegory of fi nancial capitalism, what is the allegory of the CGI human form? Let us imagine: Gozilla touches the woman; the constructed white woman is so beautiful. The monster in the Host appears, disappears and ultimately is destroyed by a band of legislated composite humans. Imagine the shooting frames as an actress—running away. Screaming. Imagining the monster in front of you. Waving your sword your guns your fi sts. Sounds a bit like Don Quixote —except we are all Don Quixote. We cannot distinguish between the windmill and the (human) monster. We attack the windmill and are winners. We can visualize the danger, the enemy, and we trust the victory. Sancho Panza does not exist in this landscape. There is no Panza to ask again: Really, that’s the monster? Come again? CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 275 FULL PROCESSING, FULL CONTAMINATION Thus far I have argued that to center and focus on the monster is a modernist, orientalist approach to analyzing representation. I have alluded to how tending to the dynamics of the digital composite might be a more useful methodology in digital humanities and software scholarship. In contrast to the CGI monster, I wish to concentrate on the digital composites. 33 Almost all professional level editing software comes with the capabilities of erasing, compositing, layering, keying in and out colors for the purposes of green/blue screening, adjusting, color-correcting, splicing and so forth. There are softwares specifi cally designed to adjust individual frame components and compositing new images into the frames. Basic commercial and proprietary software programs are utilized in editing large chunks of video: to rearrange sequences, to create/rupture linearity. Commercial programs like Adobe After Affects and Maya exist specifi cally to aid in creating motion graphics for specifi c frames in longer sequences by creating either new 3-D imaging or specifi c frame layers. For example, take a fi lm sequence that is fi lmed and needs to be edited. The editing for its narrative might happen on an editing program. However, if specifi c frames in the sequence need to be adjusted (such as, additional graphics need to be added to a series of frames), then the frames will be imported into a motion graphics software program, where the frames are altered, rendered and exported out into the editing program. Multiple software programs are utilized in almost all professional and amateur postproduction—frames are imported, adjusted, remade, rendered and exported and imported— again and again and again. While Jeon and, arguably, Derrida fi xate on the monster/beast for an analysis of empirical fi nancial capitalism (Jeon) or sovereignty (Derrida), I want to suggest that in an analysis of the digital, it may not be necessary to search for the representational beast. Rather than searching for the beast, search the monstrous digital imaging practice of compositing which fully contaminate of the screen. Interrogating “realist” digital representations, its human forms, their environments and constructions may be of more use in grappling with the narratives and allegories of empire, fi nancial capitalism, and sovereignty that we have inherited from the modernist era. In order to demonstrate the full emersion of 3-D and graphics imagining in “realist” representation, I turn to the HBO series John Adams. I am selecting this show for many reasons. The special effects rendering clips are widely accessible online and as part of the series packaging, but more 276 E. KIM importantly, there are no representational monsters in the series. However, every part of the show is composited: from the political/protest scenes to the backdrops and the extras. 34 A surface reading might conclude that this is a political, realist, period television series. Finally, I hope to make very clear that such graphics imagining is not the anomaly, but the absolute rule for commercial digital imagining. In the John Adams HBO FX special, 35 it is clear that neither the parent company nor its digital artists felt the need to hide the special effect processes. In this montage, CGI and digital alternation is not made invisible. It is perhaps hidden in a moment to the eyes of the viewers of the original series, but the postproduction artists have worked hard to bear the digitalization of every layer of the series, and to make this process searchable and accessible to interested audiences. The “behind the scenes” special makes clear that most of the characters were fi lmed with a greenscreen. Presumably, the greenscreen was keyed out and a “realist” period fi tting background was composited in. In mapping out this journey, the postproduction crew shows us to the extent in which the main actors were shot with a greenscreen (Fig. 12.2 ). This means that their acting shots took place without period specifi c back- grounds or period specifi c extras. The behind-the-scenes frames however, are not meant to alleviate the anxiety of better sight. In discussing the function of sight and visualizations, Chun utilizes the term interface. Chun argues, “interfaces—as mediators between the visible and the invisible, as a means of navigation—have been key to creating ‘informed’ individuals who can overcome the chaos of global capitalism by mapping their relation to Fig. 12.2 CGI screenshot of John Adams showing initial shot of primary actors CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 277 the totality of the global capitalist system” (8). Chun argues that the interface (which I would extend to be the screen ) is the negotiation between visible and invisible. The empowered user—be it the fi lm editor, viewer or even the software developer—navigate their “relation to the totality of the global capitalist system” (8) via their negotiation through the perimeters of visibility and access to a particularized, empowerment based knowledge systems. Arguably, behind-the-scenes clips and the digital humanities critic offer this exposure; it includes the fi lmic viewer into a particularized interface—both the production and the viewing is to empower the viewer/maker/critic into a sense of knowing more, seeing more. They have access to a visibility still invisible to others—and through this knowledge, capital systems feel more manageable/contained. Chun argues, “Freedom here stems from individual knowledge and actions, a central tenet of neoliberal governmentality” (176). How much one can see within the interface—and how the problems of sight are negotiated—is how neoliberalism works in digital technologies as well as digital scholarship dependent on bettering sight. Rather than being the vehicle of enlightenment or better insight, I hope that Fig. 12.2 might materialize the labors of compositing in digital imaging technologies. Throughout the footage, viewers are shown where the green screens may have been placed: outside, near buildings and in the pathways the actors were to walk in. We are also provided glimpses of the material layers of the compositing process. These are the layers involved in compositing: green screen, layers, frames and in this case, 3-D rendering (Fig. 12.3 ). Fig. 12.3 The compositing process of CGI animation 278 E. KIM Fig. 12.4 Constructing extras in John Adams For reasons that should be explored by our fi eld, “extras” it seems, were graphically made (Fig. 12.4 ). Such composited extras comprise, presumably, members of this political gathering (Fig. 12.5 ). Meanwhile, the actors are looking at an entirely different image composite (Fig. 12.6 ). The actors are eventually composited to appear—politically speaking—to 3-D renderings of idealized digital human forms. In the HBO FX Special, the FX team makes this point—this construction of human fi gures and personal and public interaction through compositing—clear (Fig. 12.7 ). Ultimately, this multiple layer of composition makes it possible for this scene to come the surface (Fig. 12.8 ). John Adams displays how we do not need to seek a CGI monster to fi nd a commercial visual production that entirely relies on computer graphics imaging. These frames can illustrate the how the representational human/frame is an ideal construction: all fraying hair wisps removed, all unscripted blemishes erased, all skin brightened or whitened. Frame by frame, the representational human is worked on, adjusted and radically altered. 36 John Adams may be a starting point in which we might ask many questions: why have extra actors been replaced with 3-D renderings? Why CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 279 Fig. 12.5 Crowd scene of composited extras in John Adams Fig. 12.6 Idealized renderings of political fi gures in John Adams are the protestors duplicated? How can cultural theorists and digital scholars think about the politics of color correction? How did fi lm and imaging move to be constructed this way? Such legislating against the human form is, I am arguing, the legacy of modernity. While I agree with Chun that particularized software knowledge functions as neoliberal articulations of freedom, I also believe that 3-D renderings should not be dismissed as 2-D surfaces. I have suggested that despite the resurgence of interests in modernist-driven methodologies 280 E. KIM Fig. 12.7 Interpolating human actors in a CGI crowd scene in John Adams Fig. 12.8 A fi nal composited scene in John Adams such as “surface reading”—a method that advocates against the depth of the text and advocates for descriptive readings of the surface/object—surface readings of digital objects are not possible. Surface readings’ desire for objectivity, with its putative avoidance of cultural, structural, historical and contextual readings (Marxist, Freudian, Feminist, etc.) and its denial of “political” and materialist renderings of the text or its circulation, is in itself a political project. Digital objects, arguably much like most objects, CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 281 are multi-layered and structured. Like other objects, the blueprinting for a 2-D surface begins inside of 3-D frameworks. Surface readings of digital objects, which facilitate the transference of modernist, abstracted methodologies, should be fundamentally discouraged. Surface reading espouses the belief in the truthfulness of the surface. Yet the positivist belief that certain sensory abilities, when utilized rigorously, may guide us to the truth of the structure is a methodology that refuses the materiality and the labor/circulation histories of the object. Our connection to the representational human and our willingness to trust this category is dependent on the legacies of Western humanism, particularly Orientalism. When deployed in digital humanities scholarship, these positivist and orientalist methodologies become re-articulated to visualize “contemporary” novelties. If the CGI and the “algorithmic” must be paid particular attention, I am suggesting an easy transference of modernist methodologies not be utilized to analyze its scope. Otherwise, what digital humanities scholarship is grappling with is the how to “modernize” digital texts and objects, rather than working to materialize and contextualize digital cultural production and the economies of digital forms. Taiwanese, Chicago-based digital artist Shen Yuan Su articulates that he operates within technological forms with the understanding that “I don’t believe in technology but I’m using it. I try to insult technology, but I’m using it.” Rather than utilizing proprietary, commercial based software programs, he uses open source software and open source code. 37 While CGI may be hidden from the viewer’s eye, as software it controls the editor’s digital visual form. The control is hierarchical, beginning from the workfl ow to how images are rendered. He explains of such programs, “Commercial software programs do not want you to build your own system, they don’t want you to build your workfl ow, but follow theirs.” For this reason, coding into the commercial program is a useful way of altering the workfl ow or its commands. 38 In providing an example of how coding into open-source programs alters his projects, Su states, “I don’t wanna do color correction, why should I color correct—I can code into it and shade it. And change the computer graphics. It’s a different way to think about it. The logic is totally different.” Su’s approach—to fundamentally code into the computer graphics, the hardware of the system, in order to alter the colors on the screen—is a radical provocation of how we might begin to think about fi nancial allegories attendant on digital imaging as a series of hacks, rather than a series of visualizations. 282 E. KIM THE VISUALIZATION OF NEOLIBERAL FINANCIAL CAPITALISM: FOR WHOM? I wish to conclude this essay by arguing that when we practice a surface reading of digital objects, we create limited allegories that fl atten our trajectories. This thesis stems from Derrida’s invocation that began this essay: “To whom is this power given or denied? Who is capable of death, and through death, of imposing failure” (290). I wish to suggest that in the analysis of digital visual cultures, power and priority are given to modernist narratives and methodologies. Modernist approaches will not offer the methodologies and narratives we need in order to think about the denial and distribution of power. Additionally, we do need narratives, allegories and stories that imagine an analysis that fundamentally centers the composite make up of racial capitalism. In Jeon’s description of allegory, the CGI monster makes the invisible visible, which is neoliberal, transnational, fi nancial capitalism. In the context of The Host, the monster is located as the IMF crisis, making it the focalizer of IMF cinema. Jeon locates the CGI in the monster, and with it, the allegory of algorithmic, fi nancial capitalism. This allegory is useful for a number of reasons in examining modernist methodologies in digital humanities scholarship. Through a positivist understanding of “other” as “monster,” it delegates visibility as the discourse of crisis. The International Monetary Fund entering Asia, for example, was a crisis for the neoliberal Asian nation states. While South Korea is both a neocolony and a necolonizer, the 1997 IMF crisis fi rmly visualized this standing. It marked its neocolony status globally, or to put it a different way, the IMF went into these neoliberal nation states and visually marked the order of global power. The 1997 IMF crisis was also a constructed crisis. South Korea’s capitalist economy was not in crisis; there was no anti-capitalist, socialist or communist plan underway. 39 The loan provided by the IMF to repay South Korean’s national debt to its trade partners was, much like the composite political frame, entirely manufactured. 40 This loan was provided with the understanding of ongoing trade. Consider, too, China’s loans to the US, as the US has borrowed signifi cantly from China, as well as its other allies. These loans are not necessarily a sign of a failing capitalist state, but rather signifi ers of mutually assured trade partnerships; they are given with the expectation of ongoing repayment in trade. They are not provided under the assumption that the debt will be paid in full, if ever. CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 283 Aside from the fact that the IMF crisis was a global, visual display of the prominence of Western nation states above its Asian neocolonies, the 1997 IMF crisis, as with all fi nancial crises, should be thought of as a continuum and not an event. The income disparities between the South Korean rich and poor have remained consistent since the 1990s. 41 Perhaps the IMF entering Asian nation states is a visualization of something , but the visualization of neoliberal capitalism cannot rest with the representational fi gure of the monster, but rather in the institutionalized maintenance of poverty, which transnationally is the maintenance of racial capitalism. I would like to pair the centering the IMF crisis with the terminol- ogy of the 2008 economic crisis. Scholars, journalists and politicians have become familiar with describing the corruption of fi nancial capitalism as the 2008 economic crisis. To state the “2008 economic crisis” is akin to deriving meaning and material only from the monster. 2008 was not the economic crisis; it was not a moment where the invisible was made visible. Derivative and software heavy fi nancial capitalism is not invisible: it works as corruption, and it searches for loopholes and profi ts through planned devastation. Manipulative, predatory fi nance is not invisible, but instead protected by neoliberal capitalism and purposefully hidden. This corruption is not invisible; it is part of the system, institutionalized. The corruption is the spectrum of coherence in corruption, the logic and rationale, the imagination 42 of capital as fi nance. Purposely hidden corruption is not invisibility—or rather, if it is “invisible,” it is invisible in the spectrum of legibility. What is invisible and illegibility remains invisible and illegible. The predatory condition of institutionalized poverty and the theft of black and brown properties is not described as a crisis : it is rarely accounted or theorized by digital humanities scholars as part of the narrative or the allegory. To situate 2008 as the crisis, centralizes the effect of white investment banking white economies. This language situates white fi nance as invisible, rather than accounting for what it deems invisible. In large respects, this language privileges whiteness and centers its visibility. The crisis has been the economic condition for black and brown families in the United States; the crisis is the condition for racial capitalism. The crisis is neither singular nor exceptional, but rather constant and ongoing: it is the composite. Likewise, it is important to remember that speculative trading does not begin in the transition into neoliberal fi nancial capitalism. Though arguably digital technological advancements have accelerated the voracity for devourement (to return to Derrida’s terms), the blueprint for speculative 284 E. KIM trading begins not with digital technology linked to CGI but with chattel slavery. In The Half has Never Been Told , Edward E. Baptist writes of speculative trading of slave bonds in Europe as a primary foundation of Wall Street. 43 The importation of chattel slavery into the Americas was a speculative European fi nancial market, in which US and European fi nancial fi rms could facilitate and profi t from. Granted, these early Wall Street fi rms were without seven-nanosecond fi ber optic cables, 44 and therefore, the speculative trading would have been dependent on analog technology, but the impetus to trade theft 45 and damage, and to group risk into bundles for the purposes of risk-transfer does not begin in the decades preceding the 2008 “crisis,” but with chattel slavery and the various markets created through racial slavery. Speculative trading has been accelerated, though was not invented by digital technologies. Rather, speculative trading is another derivative form of racial capitalism. The corrupted limitation of investment banking culture becomes our crisis, as the populations most damaged by the crisis, most needed in the crisis, are termless, without narrative, treated as backdrops. This is why the idea of the composite is fundamental to the creation of new allegories. What makes the frame possible, what will always make fi nance possible, is not the expensive, non-visible made visible monster banking system, but the extras (both hired and rendered), the layers, the renderings who interact with, get eaten by and fi ght off the monsters to be situated inside the composite. What the system does not count as theirs, but without it, capital and representation would not be possible. To make the invisible visible, or rather the illegible legible, is an impossible project of racial capitalism because it would mean that the entire system is being overturned. It would be wonderful if the digital humanities took on this approach 46 —but such a task requires us to re-imagine our tools, our ideologies and how the legacies of modernist forms continue to shape our readings. NOTES 1. Jacques Derrida, The Beast and the Sovereign , vol. 2, trans. Geoffrey Bennington (Chicago: University of Chicago Press, 2010): 290. I would like to thank Elizabeth Losh in particular for reminding me of this text. 2. For historical analyses on racial capitalism, see Cedric J Robinson, Black Marxism (University of North Carolina Press, 1983); Jodi Melamed, “The Spirit of Neoliberalism: From Racial Liberalism to Neoliberal CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 285 Multiculturalism,” Social Text 24: 1–24 (2006); and Nancy Leong, “Racial Captialism.” Harvard Law Review. 126 (2011). 3. “Neoliberal Forms: CGI, Algorithms, and Hegemony in Korea’s IMF Cinema,” Representations 126.1 (2014): 85–111. 4. Said explained that “Orientalism is a style of thought based upon an ontological and epistemological distinction made between ‘the Orient’ and (most of the time) ‘the Occident’” (2), and that “[T]he Orient has helped to defi ne Europe (or the West) as its contrasting image, idea, personality, experience. Yet none of this Orient is merely imaginative. The Orient is an integral part of European material civilization and culture” (2). Said argued that when looking at Western representation, “Orientalism is a cultural and a political fact” (13). This materialist, colonial dichotomy becomes useful in analysis of representational “other.” 5. See Stephen Best and Sharon Marcus, “The Way We Read Now,” Representations 108.1 (2009): 1–21. 6. Said explained, “For Orientalism was ultimately a political vision of reality whose structure promoted the difference between the familiar (Europe, the West, “us”) and the strange (the Orient, the East, “them”). This vision in a sense created and then served the two worlds thus conceived. Orientalist lived in their world, “we” lived in ours” (44). 7. In my November 13, 2015 interview with the artist and editor Jason he usefully pointed out that monsters do exist in visual and digital isolation in children’s animation and 3-D fi lms such as Monster’s Inc . Perhaps a secondary paper on the isolated, idealized, child-friendly “monster” is eventually necessary. 8. Leigh Claire La Berge argues against reading fi nance and fi nancial algorithms as “novelty.” See Scandals and Abstraction: Financial Fiction of the Long 1980s . London: Oxford University Press, 2014. 9. John Adams , directed by Tom Hooper (New York, NY: HBO Studios, 2008), DVD. 10. See Edward B. Barbier Scarcity and Frontiers: How Economies Have Developed Through Natural Resource Exploitation (Cambridge University Press, 2011), for a history on the linkage between material extraction and development. 11. On the labor that goes into hardware production see Lisa Nakamura “Economies of Digital Production in East Asia: iPhone Girls and the Transnational Circuits of Cool” Media Fields Journal (Feb. 2011). 12. In Programmed Visions: Software and Memory , Wendy Chun argues that software is “invisibly visible” (10), and that the term “soft ” is gendered. See Programmed Visions: Software and Memory (MIT Press, 2011). 13. In the same personal interview mentioned above, artist and fi lm editor Jason Hirata remarked, “Just to get an image through a lenses onto a 286 E. KIM memory card utilizes countless patens, [AU: should ‘patens’ be ‘patterns’?] algorithms, lines of code, through licensing … every capture device (sensor) utilizes a global production of technological, corporate licensing orchestra.” He noted that the network of transnational corporations should be of interest to digital humanities scholars, from Texas Instruments to GE and Sony, which hold a majority of the licensing and patents. 14. For an in-depth analysis of software processing and outsourcing in neocolonial corporations see Kalindi Vora, Life Support: Biocapital and the New History of Outsourced Labor (Ann Arbor, MI: University of Minnesota Press, 2015). 15. For a brief article on waste sites see Jakob Schiller, “Inside the Hellscape Where Our Computers Go to Die,” by Jakob Schiller published in Wired , 23 Apr. 2015. http://www.wired.com/2015/04/kevin-mcelvaney- agbogbloshie/ 16. For an overview of these functions, see Milan Sonka, Vaclav Hlavac and Roger Boyle, Image Processing, Analysis, and Machine Vision (Stanford, [AU: should ‘Stamford’ be ‘Stanford’?]Cengage Learning, 1993). 17. For a discussion on prescriptive/ideological modernity see, Fredric Jameson, Singular Modernity: Essays on the Ontology of the Present (New York: Verson,[AU: should ‘Verson’be ‘Verso’?] 2002). 18. Jodi Kim reminds us that, “Secretary of State Dean Acheson called the opening weeks of the Korean War the greatest four weeks in American history” (26). For an in depth analysis of US empire in East Asia see, Jodi Kim. Empires of Empire: Asian American Critique and the Cold War (Minneapolis: University of Minnesota Press, 2010). 19. The monster has been of concern in analysis of capitalism. Utilizing materialist methodologies, David McNally theorizes the “monster” and cultural representations of capitalism. See David McNally. Monsters of the Market: Zombies, Vampires and Global Capitalism (London: Haymarket Books, 2011). 20. Jacques Derrida, The Beast and the Sovereign , vol. 1, trans. Geoffrey Bennington (Chicago: University of Chicago Press, 2009): 49. 21. The Host , directed by Bong Joon-ho (2006; Austin, TX: Magnolia Home Entertainment, 2007), DVD. 22. See Abigail Solomon-Godeau, “The Legs of the Countess,” October 39 (Winter 1986): 65–108. 23. Much has been written about the racialized history of photographic development. For a brief overview of this see Rosie Cima, “How Photography was Optimized for White Skin” ( Priceonomics Apr. 24, 2015). In addition to this history, Hirata argues that this photo history is carried in digital fi lmmaking sensor technology. He states, “Overloaded brightness is han- dled in a nuanced way—forehead and noses on light skin—it’s to make CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 287 these things natural and good looking. Even in the best sensors—the shadows are where the noise occurs—and this is interpolated as grain.” Regarding algorithm in fi lm, Hirata explains, “The way camera sensors are tuned to light and color, and are calibrated to the spectrum—is an algorithm. The sensors algorithmically tuned to accept a specifi c slice of the spectrum and then to interpret it in as a particular set timbre of colors, grain, or noise. This technology is, of course, racist.” 24. For an examination on empire’s cultural formations, see ed. Amy Kaplan and Donald E. Pease, Cultures of United States Imperialism (Durham: Duke Universiry Press, 1993). 25. This is part of the argument that Paul Virilio and Sylvère Lotringer make in, Pure War (Los Angeles: Semiotext(e), 1997). 26. From the advent of the wet-collodion, the US military has comissioned photography in wars. For the expansive Civil War collection, see “Photography and the Civil War, 1861–1865,” accessible online through the Heilbrunn Timeline of Art History (New York: The Metropolitan Museum of Art, 2000–). http://www.metmuseum.org/toah/hd/phcw/ hd_phcw.htm . 27. It is important to state that activists and scholars have done tremendous work around the omnipresence of militarization (for example, Network of Concerned Anthropologists). See also The Counter-Counterinsurgency Manual, Or, Notes on Demiltarizing American Society. (Chicago: Prickly Paradigm Rress, 2009); Catherine Lutz, ed, The Bases of Empire: The Global Struggle against U.S. Military Posts (London: Pluto Press, 2009). 28. For a broader history of the political economy of the war and military efforts see, William S. Borden, The Pacifi c Alliance: United States Foreign Economic Policy and Japanese Trade Recovery, 1947–1955 (Madison: University of Wisconsin Press, 1984); Chalmers Johnson, The Sorrows of Empire: Militarism, Secrecy, and the End of the Republic (New York: Metropolitan Books, 2004); Paul A. C. Koistinen, Mobilizing for Modern War: The Political Economy of American Warfare, 1865–1919 (Lawrence, KS: University Press of Kansas, 1997); and State of War: The Political Economy of American Warfare, 1945–2011 (Lawrence, KS: University Press of Kansas, 2012). 29. Cathy Schlund Vials, “Vertiginous Sights and the Military Sublime: Cambodia as Spectacle in Marvel’s The ‘Nam,” lecture, January 21, 2015 at 2015. 30. For a detailed examination of the political logic and rhetoric around bombing and “rescue,” see ed. Yuki Tanaka and Marilyn B. Young. Bombing Civilians: A Twentieth-Century History (New York: The New Press, 2009). 31. “Feminism Reads Big Data: ‘Social Physics,’ Atomism, and Selfi ecity.” International Journal of Communication 9(2015): 1647–59. 288 E. KIM 32. Donna Haraway in “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective/Feminist Studies” Feminist Studies 14(3) 1988: 575–99. 33. In Chardin Material (Berlin: Sternberg, 2011), Ewa Lajer-Burcharth argues that Chardin’s painting technique consisted of processes for laying together materials, which is arguably a composite. Extrapolating from Lajer-Burchartch, one could argue, then, that the method of the composite in digital fi lmmaking stems from a longer history in Western representation. I want to thank Jason Hirata for pointing me to this text. 34. I fi nd the compositing of the protestors and/or politically conscious representational subjects to be fascinating. The FX shows how a grid of the protestors/subjects were duplicated, and the politician’s speech shot sepa- rately from this action. The separate acting shots, the graphics imagining and then compositing these frames to create what is supposed to appear as a unifi ed political scene; this seems to be a particularly rich description of the political campaigns and their compositions. 35. HBO, John Adams: Visual FX , YouTube video, 6:39, posted by HBO Studios, August 11, 2008. https://www.youtube.com/ watch?v=oTUs7hDq2PA . 36. A crude and gendered example of this is Herbie Fully Loaded , directed by Angela Robinson (Burbank, CA: Buena Vista Pictures, 2005), DVD. Lindsay Lohan’s breasts were digitally reduced in this fi lm. Typically, animation sequences are shot/made on twelve frames a second. Model. 35 mm technology worked on twenty- four frames a second, with many digital models adopting this template. Recently, digital cameras include modes that record at sixty frames per second, offering, essentially, a frame every nanosecond. I list this all out to state that every second of breast- reduction rendering meant working with at least twenty-four frames a second—if not more. That’s 1,440 individual frames every minute, However, it may have been possible that new software was made to target and overlay Lohan’s breasts. Either way, the labors involved in this gendered edit were astronomical. 37. Jeon seems to believe that editors, software users and artists are not interested in coding into the existing software. He states, “[E]ven in the aspects of fi lmmaking like CGI that most explicitly involve the manipulation of digital forms, the artist does not engage so much with the mechanical apparatus of the machine, but rather with an abstracted version in the form of interfaces, which distance the user from the mathematics. Digital fi lmmakers work increasingly at the level of image with the help of software and less at that of code” (97). However, in my experience, there is a level of expectation that professional and working fi lm editors will know how to code into commercial software programs. CGI MONSTROSITIES: MODERNIST SURFACES, THE COMPOSITE... 289 38. Su brings up that coding into open source software, or coding into the computer graphics of the hardware is a way to rupture the linear progression to color correcting in commercial software programs. Personal Interview with the artist, November 15, 2015. 39. For in depth explanations, see Kang-Kook Lee, “Neoliberalism, the Financial Crisis, and Economic Restructuring in Korea,” in New Millennium South Korea: Neoliberal Capitalism and Transnational Movements , ed. Jesook Song (New York, 2011) and Krishna Gidwani, “Korea and the Asian Financial Crisis,” accessed Nov. 23, 2015. http://web.stanford.edu/class/e297c/trade_environment/global/ hkorea.html 40. For an analysis of the situation, see The Cato Institute’s formal recommen- dation, Ian Vásquez, “Why the IMF Should Not Intervene,” Feb. 25, 1998. http://www.cato.org/publications/speeches/why-imf-should-not-intervene 41. For an exhaustive report see, Jongil Kim (Dongguk University), “Piketty Fever and Income Distribution in Korea: Reality and Prescription,” East Asia Foundation (EAF) Policy Debates (October 28, 2014). 42. Max Haiven argues that fi nance is capital’s imagination. See Haivan, “Finance as Capital’s Imagination,” Social Text 29.3 (2011): 93–124. 43. Edward E. Baptist, The Half Has Never Been Told: Slavery and the Making of American Capitalism (Philadelphia: Basic Books, 2014). I would like to thank Dorothy Wang for bringing this book up during the Modernist Studies Association Panel. “Financialization or Revolution?” November 2015. 44. See the Nanex report on high frequency trading, “The Rise of the HFT Machines,” accessed Nov. 23, 2015. http://www.nanex.net/aqck/2804. html 45. I am deriving the word theft from Hortense Spillers who writes, “[T]heir New-World, diasporic plight marked a theft of the body—a willful and violent (and unimaginable from this distance) severing of the captive body from its motive will, its active desire” (60). See “Mama’s Baby, Papa’s Maybe: An American Grammar Book,” The Black Feminist Reader. Ed. James, Joy and T. Denean Sharpley-Whiting (Malden: Blackwell Publishers, 2000), 57–87. 46. Anna Munster makes an argument regarding this need for materialist methodologies in new media studies. See Munster, Materializing New Media: Embodiment in Information Aesthetics (Dartmouth: Dartmouth University Press, 2006). INDEX 1 A Arnold, W. , 9 Adobe After Affects, 275 A Room of One’s Own (Woolf), 8, 110, Advances in Radio, 156 126, 129 advertisements, 135, 139, 143, 144, Artaud, A., 23 146–8, 150–4, 156–60 Auerbach, E., 8, 51–5, 65, 71, 72 Affectcrawler, 10, 226–30, 237–9, Autobiography of Alice B. Toklas 241n20 (Stein), 25 affective imagination, 235–7 Autodesk Mudbox, 89 Afternoon Pawnbroker , 167, 172, 178 avant-garde spaces, 258–60 Aiden, E. L., 16, 33, 34 Aimée Furniss, Scholar (Glasier), 125 algorithmic fi nancial capitalism, 268 B ALTO . See Analyzed Layout and Text “bag of words” model, 205 Object XML schema (ALTO) Bakhtin, M., 8, 52–5, 71 Analyzed Layout and Text Object Baptist, E. E., 284 XML schema (ALTO) , 142–4 Bardrick, C., 183n43 Angel Arms , 165, 166, 170–3, 176, Barnard, R., 172, 181n22 178–9, 183n43 Barnes, D., 8, 80–3, 90–7 The Antinomies of Realism (Jameson), Barney, N., 94, 97 224, 238 Barr, Jr., A. H., 22–3 Antonia, A., 227 Baudelaire, C., 198 archival-standard metadata markup, Bauer, J. , 3 141 Bauhaus, S., 16, 18, 21–3, 29, 30 1 Note: Page numbers followed by “n” refers to notes. © The Editor(s) (if applicable) and The Author(s) 2016 291 S. Ross, J. O’Sullivan (eds.), Reading Modernism with Machines, DOI 10.1057/978-1-137-59569-0 292 INDEX Bauhaus: Weimar, Dessau, Berlin, Catholic-Protestant tensions, 229 Chicago (Wingler), 30 Cavender, K., 10 Bayer, H., 23 CAVS . See Center for Advanced Visual The Beast and the Sovereign , 265 Studies (CAVS) Bech, H., 105n26 Cecire, N., 3 Bedient, C., 62 Celtic , 10, 206, 208, 215–17 Being Digital (Negroponte), 30–1 Center for Advanced Visual Studies Berners-Lee, T., 256 (CAVS) , 29 Berry, R., 226 “A Chat With Our Readers”, 147–9 Best, S., 267 Christie, A., 8, 226 The Big Clock (1946), 9, 165 Chun, W., 90, 273, 274, 276, 277, Bildungsroman , 224 279 Blanch, S., 139 Churchill, S. W., 39 Bloomsbury Group, 21 Clement, T. , 50, 57 Bolter, J. D., 17, 147, 148 Cole, S. , 226 Bolt, S. , 209, 210 Collected Poems of Kenneth Fearing , Bouchard, M., 143 166, 167, 173, 178 Boyle, T., 216 Collier, P., 137, 138 Bradley, A. J., 9 Comentale, E. P., 160 Breton, A., 84 commercial digital imagining, 276 Brian, M., 204 commercial software programs, 275, Bridges, R., 190 281, 288n37 Brinkema, E., 223, 224 Comparative Textual Media , 6 British Mass-Observation movement, 24 Complete Poems of Kenneth Fearing , Brown News Corpus, 194, 195, 207, 168 208 ComposedBlock identifi ers , 143 Brown, S., 29 computer, 1, 4, 28, 30, 31, 50, 52, The Brown Stocking , 53, 65 56, 59, 88, 191, 230, 251, Bruno, G., 19 252, 256, 259, 266, 272, Bucke, R. M., 261n3 278, 281 Burgess, A., 217 Computer Generated Imagery (CGI), Byrne, D., 233 265–8 processing and contamination, 275–81 C representational monster, Cahn, H. R. , 179 de-centering, 268–74 Canadian Writing Research visualization of neoliberal fi nancial Collaboratory/Le capitalism, 282–4 Collaboratoire scientifi que des computer-generated monster, 267 écrits du Canada (CWRC/ Coney Island , 166, 170, 172, 173, CSEC), 29 177, 179 Cartesian mind–body dualism, 261n3 Connolly, T. , 210 Cascading Style Sheets (CSS), 253, conventional affect theory, 225, 226 255–7, 263n42 Cooper, M., 29–31 INDEX 293 corporate identity campaigns, 258 digital remediation, 18, 138, 140, Critical Inquiry , 223 141, 161 Croxall, B., 4 of The Western Home Monthly CSS . See Cascading Style Sheets (CSS) (WHM) , 147–55 Cuda, A., 188 digital scholarship cultural modularity, 35 on CGI, 266 cutting-edge techniques, 52 on Ulysses , 226 “The Digital Text”, 56, 75n28 digital WHM , 140–7, 153, 160, 161 D Dilthey, W. , 72 Dale, B., 142 “Discourse in the Novel”, 53, 54 “Damned Good Poet: Kenneth displacement mapping, 88 Fearing” (Halliday), 168 distant reading, 7–9, 32–3, 49–53, Davidson, C. N., 28–9 73, 104n13, 137, 147, The Dead (Joyce), 53, 65–72 153, 160, 161, Dead Reckoning , 172–4, 176, 178, 215, 226 179 distant reading tools, 147 Dedalus, S., 203, 204, 206, 210, 230, Document Type Defi nitions (DTDs), 231 113, 116 de Melo, G., 206, 221n42 Donne, J., 198–9 Derrida, J., 265, 266, 269–71, 274, Drucker, J., 2, 16, 31–2, 84 275, 282, 283 Dryden, J., 189, 199 de Saint-Point, Valentine, 245 DTDs . See Document Type dialogism, 8, 52–6, 72 Defi nitions (DTDs) Dick, S. , 134n37 digital humanists, 1–2, 5, 16, 26, 32, 33, 44n48, 50, 85, 226 E digital humanities (DH), 1, 3, 5, Earhart, A. E., 26–7, 44n48 7, 12n6, 32, 80, 85, 89, EAT . See Experiments in Art and 169, 201n26, Technology (EAT) 277, 284 Editing Modernism in Canada at the fi eld-specifi c, 3–12 University of Alberta (EMiC formation of, 16 UA), 141, 143, 146 laboratories , 27 The Egoist (Marsden), 120, 122 pedagogy of, 37 “The Electronic Word” (Richard), 3 scholars, 283 Eliot, G. , 127, 133n35, 238 scholarship, 266, 282 Eliot, T. S. , 7, 9, 10, 25, 53, 55, 64, digital imaging technologies, 277 72, 76n42, 126, 154, 185–202, Digital Initiatives, 141 225 digital literary studies, 4 empowerment based knowledge digital media, 16–18, 30, 38 systems, 277 Digital Modernism (Pressman), 3 Empson, W., 189–90 digital reading methods, 144, 153, Erlebte Rede , 205 161 experimental humanism, 3 294 INDEX Experiments in Art and Technology free indirect discourse (FID), 7, 53, (EAT), 29 54, 65–72 extensible markup language (XML), The Freewoman , 8, 120–2 38, 85, 86, 88, 95 Freewoman Discussion Circle, 120 extensible stylesheet language Fried, D., 181n18 transformation (XSLT), 85, 86 Fry, R., 21 full-color advertisement, 158, 159 “The Future of Women in F Modernism”, 112 Fada Radios advertisement, 158, 159 Futurism, 4, 245, 246, 248, 250–2, Fawcett, M. G., 110, 118, 119 260 Fearing, K., 165–6, 182n36, 183n39, 184n47 literary criticism on, 166–9 G study and procedure, 169–74 Gaeley, A., 85 type/token ratio (TTR) and Gaelic Lexicon (O’Hehir), 215 n-gram repetition , 174–8 Garnett, E., 212 feminism, 71, 109–34 Gascoyne, D., 24 feminist digital modernisms, 7 Gawthorpe, M., 118–20 “feminist institutions of modernism”, A Genealogy of Modernism (Levenson), 112 59 Fernald, A., 112, 131n9 Georgian Poetry (Marsh), 9, 188, 190, A Few Don’ts by an Imagiste (Pound) , 191, 194 190 Germanic, 10, 206, 208, 211–13, FID . See free indirect discourse (FID) 216, 218, 219 fi eld-specifi c digital humanities, 3–12 Gitelman, L. , 17 fi nancial capitalism, 267–9, 274, 275, Glasier, K. B., 124–5, 128 282, 283 Goldberg, D. T., 28–9 Finnegan, R., 167 Gore-Booth, E., 110, 118 Finnegans Wake , 204, 218 Graphs, Maps, Trees (Piazza), 33, 34 Flanders, J., 50 Graves, R., 188, 190, 192 fl exible collective identity, 5 Green, B., 112 Ford, H., 35 Gregory, H., 167 forensic materiality, 17 Gropius, W., 22, 23, 30 formal materiality, 17 Grusin, R., 17, 147, 148 Forster, C., 3 Guillory, J., 136 Fortuna, D., 210 Foster, H., 4, 5, 249, 250, 260 “The Foundation and Manifesto of H Futurism”, 248, 252 Haiven, M., 289n42 “The Fragment on Machines”, 250, The Half has Never Been Told (Baptist), 262n27 284 Frankenstein , 271 Halliday, M., 168 INDEX 295 Hammill, F., 156 intradisciplinarity/interdisciplinarity, Haraway, D., 273, 274 27 Harley, J. B., 81, 104n6 investment banking culture, 284 Harrison, J., 126–8 Irish-British tensions, 229 Harvard lab, 18, 19, 25–6, 34 Irish nationalism, 10 HASTAC . See Humanities, Arts, Irvine, D., 6–8 Science, and Technology Italian Futurism, 245, 247, 251, 259 Advanced Collaboratory “Ivy Day in the Committee Room”, (HASTAC) 204 Hayles, N. K., 151 Hayman, D., 210 Heart of Darkness , 271 J Hedberg, J., 216 Jacob’s Room (Woolf), 128 Hellenic, 10, 206, 208, 213–15, James, H., 169 221n36 Jameson, F., 36, 79, 223–6, 228, 230, Herbert, G., 198–9 237–9 hermeneutic circle, 72–3 Jameson, S., 118–20 Heuser, R., 51, 52 Jennings, H., 24 Hirata, J., 285n13, 286n23, 287n23, Jeon, J. , 266–70, 272, 273, 275, 282, 288n33 288n37 Hoover, D. L., 169 Jockers, M., 16, 32, 34, 49–52, 227, Horace, G., 167 230 The Hospital , 179 John Adams , 11, 267, 275, 276, 278–80 Howes, M., 205 Johnson, J. , 212, 217 “Humanities Approaches to Graphical Jones, S. E., 3 Display” (Drucker), 84 Joyce, J., 8, 10, 53, 65–7, 72, 80, 152, Humanities, Arts, Science, and 154, 203–23, 226, 229, 239 Technology Advanced JoyceWays , 226 Collaboratory (HASTAC), 28–9 Humanities-sponsored panel, 4 hypermediacy, 147–8, 155, 158, K 162n16 Kaufman, J. C., 182–3n36 HyperText Markup Language Kees, W. , 168 (HTML) , 10, 11, 243–63 Kenner, H. , 205 Kenneth Fearing: Complete Poems , 168, 169 I Kepes, G., 29 immediacy, 147–8, 162n16, 155, 160, Kim, E., 11 213 Kim, J., 286n18 International Monetary Fund (IMF), King, L. A. , 170–1 266, 268, 282, 283 Kirschenbaum, M., 17 International Surrealist Exhibition in Klopfer, D., 183n45 London 1936 , 24 Klüver, B. , 29 296 INDEX L Making Canada New: Editing, laboratory modernisms, 18–26 Modernism, and New Media Laforgue, J., 198 (Irvine, Lent and Vautour), 6 Lajer-Burcharth, E., 288n33 The Making of Americans (Stein), 26, Lancichinetti, A., 145 50 Lanham, R., 3 MALLET . See Machine Learning for latent semantic analysis (LSA), 57, 60 Language Toolkit (MALLET) Latham, S., 16, 38, 40, 41, 138–9, The Man Against the Sky (Robinson), 151, 153 175 Latimer, M., 180, 182n36 Manifestoes: Provocations of the Modern Latinate, 10, 51, 203, 206–12, 219 (Lyon) , 243 Latour, B. , 7, 26, 28, 40–1 Manitoba Legislative Library, 141 La Ventre de Paris (Zola) , 237 Manovich, L., 18, 35, 85, 258, 273 The Laws of Cool , 245, 258 Mansfi eld, K., 247 Le Corbusier, 99 Mao, D., 138 Lectures in America (Stein), 26 mapping methodology, 87–91 Le-Khac, L., 51, 52 Marcus, S., 267 Le roman expérimental (Zola) , 15 Margret: A Twentieth-Century Novel Levenson, M., 59, 75n33 (Glasier), 125 Lewis, S., 136, 156 Marinetti, F. T., 4, 17, 248–50, 252 Lewis, W., 4, 205, 249, 250, 262n24 Marino, M., 227 Lexical Analysis Software Ltd., 169 markup methodology, 85–7 Leys, R. , 223, 225, 226, 228 Marsden, D., 8, 119–22 Library of Congress, 38, 142 Marsh, E., 188, 190–2 life document type defi nition, 114 Marx, K., 11, 250–2, 262n27 Liu, A. , 2, 89, 245, 258–60 Massachusetts Institute of Technology London-based Omega Workshops, 21 (MIT), 29–31 Long, H., 116 mass cultural, 136 Losh, E., 273 Massey, D., 103 Love, H., 267 Mass-Observation, 24, 25 Loy, M., 10, 25, 243–63 McGann, J., 16, 31 LSA . See latent semantic analysis (LSA) McGregor, H., 9, 141 Lyon, J. , 243, 245, 246 McKible, A., 39 McKinna, W., 227 McLuhan, M., 262n24 M McNally, D., 286n19 Machine Learning for Language Media Lab, 30–1 Toolkit (MALLET), 145–6 “media paradox” of modularity, 36 Mackay, J., 140, 162n17 “media translation”, 38, 40 MacLeod, K., 137 Metadata Encoding and Transmission Macroanalysis (Jockers), 49–51 Standard (METS) , 142–4 Macro-Etymological Analyzer, 206–7, Metadata Object Description Schema 215, 217–18, 221n42 (MODS), 37, 38 INDEX 297 “The Metaphysical Poets”, 10, 198 “Mud March”, 109, 110, 129 METS . See Metadata Encoding and Munster, A., 289n46 Transmission Standard (METS) Münsterberg, H., 15, 18, 19, 25, 26 Michel, J.-B., 16, 33, 34 Murphy, J. S., 131n13, 161 Middlemarch (Eliot), 10, 238, 239 Museum of Modern Art (MoMA), Milesi, L., 203–4 22–3 Mills, N., 172 Mussell, J., 139, 140, 143 Mimesis (Auerbach), 51–4, 65 Mirrlees, H., 126–8 MIT . See Massachusetts Institute of N Technology (MIT) National British Women’s Temperance MJP Lab . See Modernist Journals Association, 110 Project (MJP) Lab National Guardian (Rosenthal), 169 modernist dialogism, 49–77 National Research Council (NRC), 28 Modernist Journals Project (MJP) National Union of Women’s Suffrage Lab, 36–40, 113, 120 Societies (NUWSS), 109, 110, modernist self-referentiality, 17 116–22, 129, 130 Modernist Versions Project, 6, 79–80, Negroponte, N., 30–1 104n4, 105n18, 113, 226 Nelson, R. K., 145, 147 ModLabs, 15–16 neoliberal capitalism, 269, 283 laboratories, collaboratories, neoliberal fi nancial capitalism, 266–9, observatories, 26–35 282–4 laboratory modernisms, 18–26 “neoliberal forms”, 268 modularity, 35–41 neologism, 34, 185, 187, 211, 218, new-media modernism, 17–18 219, 228 MODS . See Metadata Object neomodernism, 3 Description Schema (MODS) New and Selected Poems , 166, 167 modularity, 28, 35–41 Newbolt, Sir H., 190, 197 Moholy-Nagy, L., 1, 2, 12, 22, 29–30 new-media laboratories, 16 MoMA . See Museum of Modern Art new-media modernism, 17–18, 32, 36 (MoMA) New York-based Design Laboratory, 22 Monroe, H., 188, 191, 200n3 n-gram repetition, 174–8, 180 Monumental Map, 83, 98 Ngram Viewer, 33, 34 Moonan, S., 216 Nightwood (Barnes), 8, 80, 83, 90–7, Moretti, F., 16, 32–4, 104n13, 114, 100, 103 145 nineteenth-century realism, 223 Morris, W., 127, 133n35 Norman Conquest of 1066, 208 motion graphics software program, Nouveau Paris Monumental map, 83, 275 89, 90, 93 Moving Through Modernity: Space and Nowviskie, B., 16, 31 Geography in Modernism NUWSS . See National Union of (Thacker), 79 Women’s Suffrage Societies Mrs. Dalloway (Woolf), 5, 81, 129 (NUWSS) 298 INDEX O Parsons, D., 99 Observatoire, 98 “Parturition”, 11, 244, 246, 259, OCR . See Optical Character 261n3 Recognition (OCR) Pechey, G., 66 oeuvre , 9, 168, 169, 203–4 Peel’s Prairie Provinces, 135, 141, O’Hehir, B., 215 149, 152, 159 Ohmann, R. , 158, 160 Pei, I. M., 31 OlandoVision (OViz) , 112–15 Penguin Book of Victorian Verse , 194 National Union of Women’s Pennebaker, J., 170–2, 178 Suffrage Societies, 117–22 photo manipulation techniques, Newnham College, 123–9 270 “paratextual machine”, 115–17 The Photoplay (Münsterberg), 19 Olson, E., 189, 190, 200 Piazza, A., 34 Omega Workshops, 21 Picken, E., 216 “On Computable Numbers”, 251, 252 Pierson, P.-L., 271 “onomatopoeic theory”, 212, 213 Plutarch, 270 open-source programs, 281 polyphiloprogenitive, 185–7 optical character recognition (OCR), A Portrait of the Artist as a Young 85, 141–3, 169, 201n23 Man (Joyce), 10, 203–6 Oral Poetry: Its Nature, Signifi cance calibration, 207–8 and Social Context (1977) experiment, 206 (Finnegan), 167 languages, 203, 208–17 Orientalism, 266, 267, 271, 281, unknown words, 217–19 285n4 postcolonial digital modernisms, 7 Orlando: A Biography (Woolf), 113 post-realist modernism, 224 Orlando Project, 113, 114 Pound, E. , 25, 120, 137, 188, 190, Orlando: Women’s Writing in the 193 British Isles from the Beginnings “precinematic design”, 19 to the Present , 112–15 Prescott, J., 218 National Union of Women’s preservation-quality digital images, Suffrage Societies, 117–22 141 Newnham College, 123–9 Pressman, J., 3, 160 “paratextual machine”, 115–17 Principles of Literary Criticism Ostenso, M., 155, 156 (Richards), 32 “productive convergences”, 3 Project Gutenberg, 61, 256 P proprietary software programs, 275 Panza, S., 274 “paratextual machine”, 115–17 Parisian space, 83, 91, 93, 95 Q Paris Monumental et Métropolitain Quamen, H. , 146 map , 82, 103 Quartet (Rhys), 8, 80, 87, 97–103 INDEX 299 R Schuchard, R., 188 racial capitalism, 266, 282–4 “science laboratory model”, 27 Rainey, R., 112 The Secret Life of Pronouns , 172 Rakosi, C., 173 Selected Poems of Carl Sandburg Ramsay, S., 50, 57, 67, 70 (Sandburg), 175 Rando, D. , 241–2n25 Selfi ecity (Manovich), 273 RapidMiner, 145–7 Sellars, J., 51 Rauschenberg, R., 29 sentiment analysis (SA), 10, 227 Ray, M., 23 Sexton, J. D., 182n36 Reeve, J., 10 Simon, M., 190 Reinfurt, D., 29 “single-personed hegemony”, 54 Remediation (Bolter and Grusin), 17 “Sketch of the Past” (Woolf), 130 Reminiscences of a Student’s Life Society of Young Artists, 20 (Harrison) , 127 Songs to Joannes , 246, 248–51, 257, Rhys, J., 8, 80–2, 87, 90, 97–103 258 Richards, I. A., 32, 189 So, R., 116, 132n18 Riquelme, J. P., 209, 211 Sour Grapes , 175 Robinson, E. A., 175, 288n36 South Korea, 282 A Room of One’s Own (Woolf), 8, 110, SpecLab . See Speculative Computing 126–7, 129 Laboratory (SpecLab) Roper, E., 110 Speculative Computing Laboratory Rosenthal, M. L., 169 (SpecLab), 16, 31, 32 Ross, S., 3 Spillers, H., 289n45 Rourke, J. , 232 Standardized General Markup Rousseau, V., 136 Language (SGML), 256 Ruecker, S., 85 Stanford Literary Lab, 7, 16, 32–4, 51 Ryley, R. M., 168, 173, 175, 179, Stead, C. K., 190 184n47 Stein, G. , 17, 25–6, 50 Rynell, A., 216 Stephen Hero , 204, 205 Stieglitz, A., 15, 19–20 Stirman, S. W., 170 S Stommel, J., 2 Sadler, S., 84 Stovel Printing Company, 136, 145, Said, E., 266, 285n4, 285n6 150 Salverson, L. G., 136 Strachey, J., 110, 118, 119 Sandburg, C., 175 strip-mining, 39–40 Sayers, J., 3 “structural rhythm”, 209 Schehr, L. , 92, 106n32 “The Study of Languages”, 204 Schmidgen, H., 19 Sullivan, M., 139 Scholes, R., 37, 40, 41, 120, 137 surface readings of digital objects, 280, Schreiner, O., 124, 128 281 Schrivjer, L., 83–4 Su, S. Y. , 281, 289n38 300 INDEX T W Tanigawa, K., 8, 226 Wald, A. , 168 Tansley, C., 70 Wales, K., 218 Text Encoding Initiative (TEI), 37–9, 65 Walkowitz, R. L., 138 textual markup, 85 Wang, D. , 289n43 Thacker, A., 79 warped 3D map mesh, of Paris, 89 three-dimensional mapping The Waste Land (Eliot), 7, 53, 55–65, methodology , 79–80 67, 72, 126, 137, 154 three-dimensional modelling Watters, A., 259, 260 methodology , 85, 87–91 Waugh, A. , 189 The Three Provincialities, 186 The Waves (Woolf), 50 The Three Taverns (Robinson), 175 Weinroth, M., 27 Tobler, A. , 66 Wells, H. G., 120, 212 To the Lighthouse (Woolf), 53, 54, Wells-Lynn, A., 94, 97, 105n24 65–72, 129, 154 Wernimont, J., 116 traditional disciplines, 4, 28 Western digital humanities, 266 “Tradition and the Individual Talent”, The Western Home Monthly (WHM) , 10, 198 135–8 transhumanism, 245 digital remediation of, 147–55 Tschichold, J., 38–9 digitization of, 140–7 Turing, A. M., 11, 251, 252 modernism, 138–40 type/token ratio (TTR), 174–8 remediating new and old media, 155–61 Western visual iconography, 272 U West, N. , 181n22 Ulysses (Joyce), 5, 10, 80, 104n3, 154, West, R., 120 216, 218, 223–42 White, S. E., 241n21, 241n22 Ulysses Meets Twitter 2011 project Whittaker, S., 204 (Cole), 226 “whole scale annihilation”, 272 Underwood, T., 51 Wild Geese (Ostenso), 155, 156, 158, “United Procession of Women”, 109, 160 111 Willems, B., 255 Williams, A., 128 Williams, W. C., 175 V Wingler, H., 30 van Orden, N., 8, 9 “A Woman’s College from Outside” Victorian model, 258 (Woolf), 128–9 The Virtual Laboratory , 19 Women’s Freedom League, 109 Visconti, A., 226 Women’s Liberal Federation, 110 Visible Language Workshop (VLW), Women’s Social and Political Union 29, 30 (WSPU), 109, 116, 117, 119, Vorticism, 4 130n1 INDEX 301 Women’s Trades’ Council, 110 XML markup of text and map-based Woogar, S. , 26, 40–1 locations, 85, 86 Woolf, V., 7, 8, 50, 53, 54, 71, 72, 81, XSLT . See extensible stylesheet 110, 113, 126–30 language transformation Woolgar, S., 7, 40 (XSLT) WordSmith, 169, 170, 173–5 Wordsworth, W., 188, 189 World’s Columbian Exhibition (1893), Y 19 Yeats, W. B., 110, 188 World Wide Web Consortium (W3C), 256 WSPU . See Women’s Social and Z Political Union (WSPU) Zaturenska, M., 167 Wulfman, C., 37, 40, 137 z-axis mapping, 81–5, 105n24 Wulf, W. A. , 28 modernist Paris on, 103 of Nightwood (Barnes), 80, 92 z-axis research, 80–2, 85, 90, 103, X 104n3 XML . See extensible markup language Zeikowitz, R., 99, 105n17 (XML) Zola, É. , 15, 225, 237