Modernist informatics literature, information, and the state

Modernist Informatics


Modernist Literature & Culture
Kevin J. H. Dettmar & Mark Wollaeger, Series Editors
Consuming Traditions
Elizabeth Outka
Machine Age Comedy
Michael North
The Art of Scandal
Sean Latham
The Hypothetical Mandarin
Eric Hayot
Nations of Nothing But Poetry
Matthew Hart
Modernism & Copyright
Paul K. Saint-Amour
Accented America
Joshua L. Miller
Criminal Ingenuity
Ellen Levy
Modernism’s Mythic Pose
Carrie J. Preston
Pragmatic Modernism
Lisi Schoenbach
Unseasonable Youth
Jed Esty
World Views
Jon Hegglund
Americanizing Britain

Genevieve Abravanel
Modernism and the New Spain
Gayle Rogers
At the Violet Hour
Sarah Cole
Fictions of Autonomy
Andrew Goldstone


The Great American Songbooks
T. Austin Graham
Without Copyrights
Robert Spoo
The Degenerate Muse
Robin Schulze
Commonwealth of Letters
Peter J. Kalliney
Modernism and Melancholia
Sanja Bahun
Digital Modernism
Jessica Pressman
In a Strange Room
David Sherman
Epic Negation
C. D. Blanton
Modernist Informatics
James Purdon


Modernist Informatics

Literature, Information, and the State

James Purdon


Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research,
scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in
certain other countries
Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of
America
© Oxford University Press 2016
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any
means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms
agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be
sent to the Rights Department, Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same condition on any acquirer
Library of Congress Cataloging-in-Publication Data
Purdon, James, 1983–
Modernist informatics : literature, information, and the state / James Purdon.
p. cm. — (Modernist literature & culture ; 25)
Includes bibliographical references and index.
ISBN 978–0–19–049332–5 (ebook) — ISBN 978–0–19–021170–7 (pdf)
1. Modernism (Literature) 2. Government information—Access control. 3. Mass media and literature—History—20th century. I. Title.
PN56.M54P87 2016
809’.9112—dc23
2015012270

9 8 7 6 5 4 3 2 1
Printed in the United States of America on acid-free paper

for my parents


One can conceive of Heaven having a Telephone Directory, but it would have to be gigantic, for it would include the Proper Name and
address of every electron in the Universe. But Hell could not have one, for in Hell, as in prison and the army, its inhabitants are identified
not by name but by number. They do not have numbers, they are numbers.
—W. H. Auden, The Dyer’s Hand


Contents

Series Editors’ Foreword
Acknowledgments
Introduction: The Government of Information
1. Secret Agents, Official Secrets
Inspector Heat’s Postal Power
Korzeniowski’s Dead Letter Box
Insecurity
Inside Knowledge
2. Dossier Fiction
Abwehr’s Looking at You, Kid
Ford’s Forms
Black Books, Black Markets
Papers, Please
Thrillers
3. Information Collectives
The National Point of View
… a Mass of Unrelated Facts
This Little Old E. of Ours

Connectivity and Collectivity
… Paper of All Kinds Broadcast Everywhere …
4. Public Information
Information Film
Utility Aesthetics
A Pattern of Thought and Feeling
Informatic Voices
5. Information Blacked Out
In the Gloaming …
Keep It Dark
Keep Mum
Frankie’s Papers, Bowen’s Notes
Between the Lines
Coda: Information Machines


Notes
Selected Bibliography
Index


Series Editors’ Foreword

To indulge in ruthless telegraphic shorthand, Modernist Informatics provides a prehistory of
cybernetics. James Purdon’s richly detailed book–the range of reference is astonishing–develops its
argument through what sometimes resembles a chapter by chapter scatter plot, the technique of data
visualization that demonstrates (where possible) correlation and coherence within a mass of data that
otherwise might seem random.
Thus in chapter one, “Secret Agents, Official Secrets,” Joseph Conrad’s The Secret Agent shares
space with nineteenth-century postal scandals, underpaid copyists, Italian revolutionary Giuseppe

Mazzini, and the Official Secrets Acts of 1889, 1911, and 1920. In chapter two, “Dossier Fiction,”
Ford Madox Ford’s Parade’s End constellates standardized forms, the National Registration Act of
1915, Arnold Bennett’s overlooked modernist gem Riceyman Steps, the concept of the data double,
fingerprinting, passports, and the thriller. In the third chapter, “Information Collectives,” Stephen
Spender’s journal Fact introduces Mass-Observation (M-O) and responses to M-O by satirical
essayist G. W. Stonier, novelist and screenwriter John Sommerfield (if you don’t know him, you’ll
want to), and anthropologist and occultist T. C. Lethbridge. Chapter four, “Public Information,”
emplots documentary film maker John Grierson, W. H. Auden, concepts of patterning, and the picture
language known as Isotype (International System of Typographic Picture Education). Chapter five,
“Information Blacked Out,” offering the most sustained literary analysis of a single text, locates
Elizabeth Bowen’s The Heat of the Day within a matrix that includes World War II propaganda
posters and films, optical and informational black-outs, and E. M. Forster’s essays. Finally, “Coda:
Information Machines,” connects the dots between George Orwell, Claude Shannon, and Norbert
Wiener, who brings us back to cybernetics.
The coherence of these data points is made possible by the clarity of Purdon’s central claim:
informatics–that is, the science of managing information–does not begin, as is commonly thought, with
digitization but with the emergence of new controls on information around the turn of the twentieth
century that Purdon calls (playing with the ambiguous genitive) “the government of information.”
Here’s where Purdon’s difference from a number of related books becomes clear. Literary treatments
of information control from which Purdon’s book emerges include Alexander Welsh’s George Eliot
and Blackmail, Mark Wollaeger’s Modernism, Media, and Propaganda, and Mark Goble’s
Beautiful Circuits; literary treatments attuned, as Purdon is, to the material infrastructure of
information include Carolyn Marvin’s When Old Technologies Were New , Richard Menke’s
Telegraphic Realism, and Jay Clayton’s Charles Dickens in Cyberspace; and enabling non-literary
histories of information include James Beniger’s The Control Revolution and Daniel Headrick’s
When Information Came of Age.
But none of these focuses so intently on the government of information, which is to say on the way
that information has emerged as both the basis of modern political power and one of its primary
objects of attention. Thus the sustained story of Modernist Informatics is how the politics of
information became one of the principal ways through which both literary culture and the state came

to define themselves in the twentieth century. Roughly coincident with common periodizing accounts
of modernism, the British government began to undertake coordinated administration of a state
understood as dependent on information beginning around 1889 with the first of a series of Official


Secrets Acts, and it had to change approaches after World War II, when digital computing required
the development of a new set of control protocols, or cybernetics proper. The literary data points
here do not coincide with the usual high modernist suspects–the men of 1914 or, in Bonnie Scott’s
revision, the women of 1928. Like Wollaeger, Purdon routes his story through Conrad and Ford, both
of whom were closely involved with official networks of information control, but also through many
less well known writers and film makers. These chapters will undoubtedly spur others to extend his
approach to additional figures, from Joyce, Eliot, and Rebecca West, who surface occasionally, to
those who do not, such as H. G. Wells, Virginia Woolf, or Dorothy Richardson.
How did we arrive in today’s media ecology, in which biometric data collection, information
insecurity, and global hacking scandals dominate not only the headlines but our lives? And what kinds
of writing and film in the modernist period anticipate today’s “glitch” or “archive” art? In Purdon’s
rich (pre)history, Conrad, exploring leaks and the emergence of “preprocessing” (Beniger), emerges
as a kind of proto-hacker; Ford, attentive to war-time systems for the storage and cross-checking of
personal data, adumbrates the concept of the data double and the “dividual” long before the terms
were coined, respectively, by surveillance studies and Gilles Deleuze; and Bowen, in Purdon’s tour
de force analysis of her torqued syntax, unidiomatic double negatives, and insistent redundancies,
comes to look like a signal jammer. Modernist Informatics offers more surprises than sysadmins are
willing to tolerate, and for that reason will make its readers very happy.
—MARK WOLLAEGER AND KEVIN J. H. DETTMAR


Acknowledgments

Long before it came to describe a set of virtual patterns, “information” meant the process of
fashioning a character, a consciousness. First and foremost I am grateful to the teachers who have

informed me in that older sense of the word, and most especially to David Trotter, whose influence
and example helped to shape this book from start to finish.
Leo Mellor played a key role in the early stages, when the ideas expressed here were still in the
making; later, Alex Houen and Adam Piette offered invaluable criticism and guidance. Many of the
arguments made in these pages were tested and strengthened in the course of lively discussions with
Beci Carver, Charlotte Charteris, Michael Englard, James Fox, John Gallagher, Olivia Laing, Robert
Macfarlane, Rod Mengham, Robbie Moore, Ian Patterson, Beryl Pong, Mark Rawlinson, Jordan
Savage, Philip Sidney, Alfie Spencer, and James Wade. For their support and encouragement at other
times and in other ways, I am grateful to Jon Day, Susanna Hislop, Arthur House, Thomas Marks,
Peter Scott, and Dan Stevens.
Mark Wollaeger and Kevin Dettmar were encouraging readers of the manuscript, while the two
anonymous readers who reported on it offered a wealth of constructive suggestions.
The Arts and Humanities Research Council funded the doctoral thesis that began the project.
Thanks for practical help are due to the archivists of Mass-Observation at the University of Sussex
and the John Grierson Archive at the University of Stirling, and to the curators of the British Postal
Museum. I am also grateful to the fellows and staff of Emmanuel College, Cambridge, and Jesus
College, Cambridge, where most of the writing took place.
Finally, I owe a special debt of gratitude to Kristen Treen.
Material from the Mass-Observation Archive is reproduced with permission of Curtis Brown
Group Ltd., London, on behalf of The Trustees of the Mass-Observation Archive.
Quotations from John Grierson’s unpublished writings are reproduced with the kind permission of
the John Grierson Archive, University of Stirling.
Part of Chapter 1 first appeared as “Secret Agents, Official Secrets: Joseph Conrad and the
Security of the Mail,” The Review of English Studies 65, pp. 302–320; parts of Chapter 4 appeared
in “Electric Cinema, Pylon Poetry,” Amodern 2 (October 2013).


Modernist Informatics

Introduction: The Government of Information
Where is the wisdom we have lost in knowledge?
Where is the knowledge we have lost in information?1
—T. S. Eliot

In the July edition of The Idler for 1892 there appeared a short story by Israel Zangwill. Part science
fiction, part sting-in-the-tail parable, “The Memory Clearing House” is narrated by a “poor, unhappy,
struggling, realistic novelist” who, having allowed his realist ambitions to lapse, has at last found
popularity with a successfully sentimental romance.2 Using his new-found wealth, he has moved to a
more expensive address, and circulated the usual cards to let his friends know where to reach him.
Yet all is not well. Before long, one of those friends—an Irish parliamentarian named O’Donovan—
accosts the narrator in the street to complain about the difficulty of replacing the memory of the old
address with that of the new:
Just imagine what a weary grind it has been to master—“109, Little Turncot Street, Chapelby Road, St. Pancras.” For the last
eighteen months I have been grappling with it, and now, just as I am letter perfect and postcard secure, behold all my labour
destroyed, all my pains made ridiculous. It’s the waste that vexes me. Here is a piece of information, slowly and laboriously
acquired, yet absolutely useless. […] It cannot be scotched—it must lie there blocking up my brains, a heavy, uncouth mass,
always ready to spring at the wrong moment; a possession of no value to anyone but the owner, and not the least use to him. (p.
187)

Frustrated, O’Donovan resolves to turn his own difficulty into an opportunity. With the help of
“psychical science” he succeeds in contriving an apparatus which he calls “the noemagraph, or
thought-writer” for transferring memories from one mind to another (p. 191). This device, which
imprints memories “on a sensitised plate,” preserves discrete, decontextualized, transferable records
of memory just as a camera preserves discrete, decontextualized, transferable records of light: what
the photograph does for the eye, the noemagraph does for the mind.
Before long O’Donovan is in the information business. He establishes the titular clearing house,
setting the methods of the stock-exchange (or, equally well, the pawn shop) to work on memory itself,
and is soon doing a roaring trade among those whose minds unhelpfully store useless information,
particularly “politicians, clergymen, and ex-examinees” (p. 193). When the business expands beyond

the trading of superfluous memories to deal in trade-ins, memory rentals, and amnesia, however, it
begins to expose problems in the relationship between information and the integrity of the self.
Customers who are “tired of themselves” begin to call at the clearing house “to get a complete new
outfit of memories, and thus change their identities” (p. 195). In the end, the formerly struggling
realistic novelist caves in and invests in the memories of a murderer, hoping to write “the most
veracious novel the world has seen” (p. 202). Having done so, he finds himself facing trial for the
very murder he had hoped to use in his realistic fiction.
Zangwill’s story satirizes fin-de-siècle London’s posing litterateurs as well as its upwardly mobile
middle classes, but what makes “The Memory Clearing House” doubly interesting is that it plays one
anxiety of excess against another, making information overload into a way of talking about the
expansion of consumer culture, and vice-versa. “Information,” which had once meant the shaping of
character, especially through moral instruction or religious inspiration, had by the 1890s lost most of


those old associations. But it had not yet come to designate a mathematical quantity or a virtual
domain. In the meantime what it increasingly resembled, at least to the primed gaze of late-Victorian
capitalism, was a commodity.
Certainly, information can be understood to satisfy the essential requirements of the commodity
form as Marx had expressed them.3 It has a use-value: O’Donovan’s problem is not that his
laboriously acquired information is inherently useless, but that—like many other commodities—it is
perishable. It can satisfy a human need or desire. And, as O’Donovan realizes, it has an exchangevalue: it can be traded for other commodities, or for a money equivalent. Yet information is a strange
kind of commodity, one that can exist simultaneously in multiple locations. “If I have a disk and make
a copy for you,” writes N. Katherine Hayles, “we both have the information. Like the fabled magic
pot, information promises to proliferate virtually without cost.”4 This strange quality of informationas-commodity—that information exchanged may also be retained—has led theorists such as Hayles
and Mark Poster to speculate about the relationship between the increasing ease with which
information can be reproduced and the radical reorganization of capitalism in post-industrial
economies. The “dream of information,” in Hayles’s account, opens the way to “a realm of plenitude
and infinite replenishment, in sharp contrast to what might be called the regime of scarcity.” In this
new realm, which both Hayles and Poster associate with the increasing ease and speed of
reproduction enabled by digital technology, terms such as “exchange value” and “surplus value”

become nonsensical: the value that inheres in information seems to transcend the laws of supply and
demand. That, at least, is the fantasy. Yet the regime of scarcity, both theorists also point out, has a
habit of reasserting itself, bringing the dream of informational plenitude and freedom back under the
control of market relations and political hierarchies. The genius of the noemagraph, which furnishes
Zangwill’s story with its conceit, is that it does just that. In response to information’s dizzying
increase, it reinstitutes the regime of scarcity by technological means, allowing information to be
treated like any other commodity. By removing memories from one mind and installing them in
another, the apparatus transforms information into a thing that can be exchanged in the same way that
bales of linen or hats can be exchanged. Customers at the memory clearing house forfeit their
superfluous information for cash, or for new information. Either way, they no longer possess the
original, and for O’Donovan this is all to the good. Fixed as a finite substance—“a heavy, uncouth
mass”—information finally makes sense to his entrepreneurial mind.
To those without the benefit of a noemagraph, however, information remained profoundly
troubling. Returning to Hayles’s analogy, it helps to recall the cautionary aspect of the Grimms’ magic
pot, which goes on uncontrollably producing porridge beyond all bounds: “and there was the greatest
distress, but no one knew how to stop it.”5 The anxiety that attends most discussions of information in
late nineteenth- and early twentieth-century writing arises precisely from the fact that it cannot be
secured and controlled in the same way as other commodities.6 Information is transmitted rather than
traded: its governing logic is not exchange but contagion. (The virus, Jussi Parikka has suggested, is
the informational form par excellence.)7 Like other real and imaginary epidemics of the fin-desiècle—criminality, homosexuality, syphilis, absinthe-drinking—information was understood both as
a moral and as an economic problem, a disturbance of the relationship between the self and the world
and a threat to the structure of society. The epidemic had somehow to be brought under control. The
commodity form had to be secured.
We no longer think of information as a substance. Or, if we do, we seldom think in terms of the
metaphors used by O’Donovan. Heaviness, uncouthness, mass: these describe a set of associations


very different from those which circulate at the beginning of the twenty-first century, when
information tends to figure as weightlessly abstract, culturally dominant, the most sought-after of all
commodities. How did that transvaluation happen? How was information transformed from the

burdensome residue of fact into the “universal solvent” of cybernetics and digital computing?8 What
cultural pressures went into the making of information as a concept before it was conscripted as a
theory? Those questions are at the heart of this book. Its general claims are that information and
literary narrative have a history of entanglement as well as antagonism, and that this relationship—the
productive challenge posed to literature by the genres of information—was a significant factor in the
cultural shaping of modernist narrative.
*
One good reason for beginning with “The Memory Clearing House” is that it successfully anticipates
many of the information problems that would become pressing for twentieth-century writers. In an
influential essay first published in the year 2000, the critic James Wood wrote that “Information has
become the new character.” 9 His point was that the vast novels of contemporary fiction have tended
to focus less on the elaboration of unique subjectivities than on the background hum of information—
information about manufactured objects, industrial processes, flows of commodities, and patterns of
consumption—against or through which the postmodern subject comes to be formed. But Zangwill’s
story should once again remind us that such anxieties about the proliferation and control of
information are not new. They weren’t new even to Zangwill. An 1853 editorial in The Times could
declare in passing that Victorian England was already “in an age of information,” and the late
nineteenth century’s array of new i-compounds easily matches that of the twentieth. 10 As early as the
1780s, British travelers could take advantage of an “information office” in Calcutta, but it was almost
a hundred years before vocabulary really caught up with the rising professions, producing new terms
including “information bureau” (first citation in the Oxford English Dictionary 1869), “information
agent” (1871), “information room” (1874), “information service” (1885), “information officer”
(1889), and “information work” (1890). In the company kept by the word—public service and
business, agency and control—we can begin to discern the outline of a modern understanding of
information as a form of mediation which structures relations between individuals, corporations, and
state bureaucracies.
In the middle of the nineteenth century, those relations were significantly reconfigured by a series
of interrelated administrative changes. Between them, the Penny Post of 1840, the end of the
newspaper stamp duty in 1855, the repeal of paper duties in 1861, and the Elementary Education Act
of 1870 formed the conditions for increasing literacy among the lower social classes, and fueled

demand for reading matter. 11 Meanwhile, extensive reforms were carried out in the Post Office, the
Civil Service, the British Museum, the British East India Company, and other state institutions. 12 The
sophisticated new classifying and sorting operations developed in the course of those reforms
gradually effected the transformation of Victorian statistics into modernist informatics. Where the
Victorians had instituted the much-satirized Society for the Dissemination of Useful Knowledge, their
children and grandchildren would build Ministries of Information; where statistics had come to be
understood as the science of rational government according to abstract data, informatics would
foreground the new material, social, and technological practices of official communications.
Informatics supplemented the bureaucratic fantasy of rational government-by-numbers with a more
complicated set of protocols, technologies, and social assemblages designed to mediate between


states and populations.
The Oxford English Dictionary defines “informatics” as the “branch of study that deals with the
structure, properties, and communication of information and with means of storing and processing
information.” The word entered English in the computer science boom of the mid-1960s on the model
of the Russian informatika (the theory of scientific information) and perhaps under the influence of
related but distinct terms in French (informatique) and German (Informatik). Informatics can be
regarded as the infrastructure of information, but it need not be limited to digital computing machines,
or indeed to machines alone. The most recent edition of the International Encyclopaedia of
Information and Library Science, for instance, allows that since “computers, individuals, and
organizations all process information, informatics has computational, cognitive, and social aspects.”
Informatics, that is to say, deals with the representation, processing, and communication of
information within and between systems of several kinds: not only “computer communications and
networking” but “paper, analogue and digital records,” “organizational processes,” and even “human
reasoning.”13
This more expansive concept of informatics has helped to shape contemporary theories of
technology and culture. In her “Cyborg Manifesto,” for instance, Donna Haraway argues that the
“polymorphous information system” of post-industrial capitalism has brought about “a massive
intensification of insecurity and cultural impoverishment.”14 The consolidation of what Haraway calls

“the informatics of domination” marks, decisively in her view, the transition from the settled form of
industrial capitalism in which power and ownership relations are relatively clear to the decentered
global flows of information that characterize the expansion of capital in its post-industrial mode.
Similarly, N. Katherine Hayles proposes a broad definition of informatics encompassing all of “the
material, technological, and social structures that make the information age possible.”15 Like Hayles I
find it useful to distinguish these structures from the “information” stored, processed, and transmitted
in accordance with their individual capacities and rules. Yet here I part company with both Haraway
and Hayles, who understand informatics as a relatively recent development, with primary application
to systems that have already become digital. The “Cyborg Manifesto” does not attempt to distinguish
between what Haraway has elsewhere called “post-industrial, post-modern, or other posted
locations,” nor does Hayles pursue the implications of her sweeping definition of informatics for our
understanding of an era before cybernetics.16 My own view is that the informatic revolution described
by both theorists was as much a legacy of early twentieth-century bureaucracies as a post-industrial
achievement. In what follows, I trace the development of informatic systems in the period between
steam and cybernetics, when the statistical analyses of nineteenth-century public offices gradually
became the informatic protocols of twentieth-century ministries, and the attention of government
shifted from the analysis of populations and quantities to the control of access and channels.17
Modernist Informatics explores the premise that informatics begins not with digitization but rather
with the development of new information controls in nineteenth- and early twentieth-century
government systems. Informatics might then be understood as the government of information,
allowing that phrase its full measure of ambiguity in order to call to mind the way information has
emerged as both the basis of modern political power and one of its primary objects of attention and
control.
Literary criticism, on the other hand, has been relatively slow to take an interest in information and
informatics. In March 2010 the journal Library and Information History could look back on “An
Information History Decade” during which important work had been done in evaluating the role and


importance of the concept of information in the development of modernity. 18 Not until the end of that
decade did modernist studies begin seriously to broach the subject of information, by way of an

increasing interest in the role of the state in cultural production. In an influential 2008 article for
PMLA, Douglas Mao and Rebecca L. Walkowitz reviewed the work of “The New Modernist
Studies,” paying particular attention to scholars’ growing interest in international modernisms and
political modernisms, and calling for scholarly investigations to “probe much further the effects of the
state on Modernist production.” “It would be surprising,” they concluded, “if modernist studies,
centered as it is on times and places marked by especially dramatic changes in the politics of
information, ignored this pressing challenge.”19
One way of responding to that challenge is by directing attention toward phenomena other than the
consumable public media of entertainment and the audio-visual technologies of storage and
dissemination. With a tacit understanding that the most heated debates regarding twenty-first-century
data culture involve the retention and use of information by state institutions and transnational
corporations, this book posits a new media presence in early twentieth-century narrative: one which
has received little critical attention for the very reason that it operates semi-covertly, mediating not
between an artist and an audience but between citizens and the institutions of the state. “In our epoch,”
notes John Guillory, “large numbers of people write, but they do not for the most part write poems or
scientific papers; they fill out forms, compose memos or reports, send interoffice emails.”20 These
“information genres” as Guillory calls them—along with post-marks, records and reports, files and
licenses, passports and mugshots—have tended to remain covert in the sense that their reading has
been imagined, when imagined at all, as an official activity. Their ideal reader, typically conceived
as a professional bureaucrat or state functionary, is one whose role is to process conventions and data
according to particular institutional rules and with a view to defined outcomes: marriage, arrest,
admittance, payment, conscription, and so on. The representation of such forms in literary works
tends, therefore, to depend upon an interruption in the usual circulation of documents in institutional
sign systems that otherwise remain closed and invisible as a condition of their proper functioning.
Alexander Welsh has shown how, for instance, the increasing prominence of blackmail plots in the
novels of George Eliot reflects the emergence of a characteristic “pathology of information” in the
latter half of the nineteenth century. By making blackmail intrinsic to its narrative structure, Eliot’s
fiction is able to uncover the contradictions within a new set of social conventions regarding privacy,
publicity, and reputation. 21 Drawing on Welsh’s work, Mark Wollaeger goes on to argue that as
blackmail was to the Victorians, so propaganda—the “chief information pathology of the twentieth

century”—was to their modernist successors.22 Yet here, it seems to me, pathology encounters a slight
difficulty. For it remains unclear why propaganda should necessarily constitute a pathology of
information. Indeed, from an official perspective, successful propaganda might rather appear to be the
sign of a national information system in rude health. If a symptomatic analysis of this kind is to work
effectively, it needs to account for the fact that information systems are fundamentally political in
ways that the human body is not, and that what appears pathological from one perspective can look
perfectly proper and even desirable from another. Where information systems themselves become
objects of attention, we are dealing not simply with moments of error, but with what Lisa Gitelman
calls “moments of innovation, dispute, breakdown, transfer, and the like, moments in which the
grounds of meaning itself seem to have been most clearly at stake.”23 Is the leaking of official
documents a political pathology or a public service? Are identification technologies necessary for the
safety of law-abiding citizens or illiberal impositions of control? These questions are still relevant,


and making sense of them requires political as well as symptomatic analysis.
Modernist Informatics charts the structures of information in modernist culture by attending to
moments at which those structures emerge into view, but it also insists that “the politics of
information,” far from constituting a new field for the exercise of forms of official power later to be
represented in cultural artifacts, was one of the principal ways in which both the state and literary
culture came to define themselves in the new century. Information did not spring suddenly upon the
world at the end of the nineteenth century, yet from the work of writers such as Joseph Conrad, Ford
Madox Ford, Graham Greene, and Elizabeth Bowen there emerges a new understanding of the
relationship between the technologies of information management and the conditions of everyday life
and thought in modern societies. Conrad’s interest in interception and leaking, Greene’s horror of the
lost identity, and Bowen’s structural and syntactic jammings are all attuned to the contested
frequencies of information in ways that remind us forcefully of the difficulties of reading, thinking,
and knowing in the situation of modernity.
Regarding the scope of that modernity, some studies have pushed the history of information back
further still. Welsh himself has argued for the crucial role played by the Foudrinier paper machine
(1799) and the rotary press (1843) in encouraging increased literacy and meeting the consequent

demand for printed material. “For better or worse,” he concludes, “the age of information commenced
about two hundred years ago.”24 For Toni Weller, too, “The 1800s formed the nascent years of our
modern information age.”25 Extending the timeline, Peter Burke notes that “The commodification of
information is as old as capitalism.”26 Despite their advantages, such wide optics have tended to
elide real historical differences in how information was conceived, managed, and represented. As
Daniel R. Headrick has pointed out, “information” has come to designate a category so vague as to be
critically useless: he therefore turns his attention to “a more manageable concept, the study of
information systems.”27 Yet I am less confident than Headrick that we can point to a threshold
moment when information (as the title of his study puts it) “came of age.” There was no one moment
when information became a governing force in people’s lives, nor even a significant date after which
it suddenly became widely understood as such.
But if we cannot set a date for the beginning of the information age, we can at least identify the
crucial moment when the British government began to construe itself as administering a state
predicated on information. This moment was the passing into law in 1889 of An Act to Prevent the
Disclosure of Official Documents and Information, generally known as the Official Secrets Act, the
first piece of legislation of its kind anywhere in the world. The Conservative government of the day,
led by Lord Salisbury, had been preparing the legislation for almost two years, spurred by the
increasing commerce between newspaper editors who felt they were being deprived of information,
and clerks and draftsmen in government service who felt they were being deprived of an income
consistent with their responsibilities. Information security was thus both an inward-looking and an
outward-looking phenomenon, in which the danger posed by external enemies was compounded by
internal grievance.
By claiming this moment as crucial, I do not mean to suggest that literary texts simply produce or
respond to moments of sudden convulsion in media or in the politics of information. Following the
example of James Beniger, whose The Control Revolution has long been the standard work on
Anglo-American communications and control technologies, I take it that revolutions can be ongoing
processes as well as sudden events. From Beniger, too, I draw corroboration for the intuition that we
have tended retrospectively to post-date the development of “information society” in seeing

twentieth-century instruments (and computer microprocessors in particular) as marking a decisive
break with pre-existing technologies rather than accelerating an ongoing process. Against “prevailing
views, which locate the origins of the Information Society in World War II,” or in “the commercial
development of television,” or in computers, or in “computer-based telecommunications in the 1960s
and early 1970s,” or in “microprocessing technology in the late 1970s,” Beniger argues—rightly, in
my view—that “the basic societal transformation from Industrial to Information Society had been
essentially completed by the late 1930s.”28 While the rate and method of this transformation differed
significantly between the United States and the United Kingdom, Beniger’s account nonetheless
provides a persuasive, if counterintuitive, periodization. Where my argument differs from his is in
emphasizing the role of cultural production, as well as the official discourses of parliamentary
debates and legislative instruments, in the relationship between governments and publics. Official
policy in Britain was neither solely driven by technological change nor simply imposed upon citizens
according to political expediency. Rather, political discourse about information security, and about
national security more broadly, was enmeshed with social, technological, and cultural changes.
Novelists affected government policy; governments censored books and paintings; ministries
produced films and sponsored radio broadcasts. Technologies of photographic and cinematic
representation developed at the same time as, and in dynamic interaction with, “official” technologies
of identification and surveillance.
In the case of the visual arts, Sven Spieker has proposed that the aesthetics of early twentiethcentury visual modernism can be understood “as a reaction formation to the storage crisis that came in
the wake of Beniger’s revolution, a giant paper jam based on exponential increase in stored data, both
in the realm of public administration and in large companies whose archives were soon bursting at
the seams.”29 The archival imagination became a focus for artists because of its rich possibilities for
modeling the relationship between time, memory, and chance. Where the eighteenth-century
information systems of Linnaeus, Diderot, and the philosophes sought to classify, organize, and
discriminate, the archives of the nineteenth century became places “where historians hoped to find the
sediments of time itself […] in flux and ongoing.” For Spieker, the expansion of archives in the
nineteenth century reflects an increasing interest in registering the visible traces of what normally
eludes representation, so that the modernist project may be said to begin by turning that high ambition
back on the archive itself, revealing how contingency not only determines its contents but affects the
intellectual work it accomplishes and enables. Spieker’s thesis is appealing, not least because it

allows him to account in new and persuasive ways for the upsurge of interest in dossiers, files, and
bureaucratic records in the work of twentieth-century artists from Marcel Duchamp to Gerhard
Richter. My own enquiry begins from a similar hypothesis about the proliferation of information, but
it approaches the problem through analysis of texts from several apparently discrete domains of
discourse. My aim in these readings has been to recognize not only the contingency of archival
processes but also the real continuity between those realms—public/private, official/unofficial,
literary/popular—whose notional boundaries were shaped by the circulation of writing.
So much for Informatics: why modernist? It should be clear from the outset that comparatively little
attention is devoted here to those writers—Eliot, Joyce, Pound, and Woolf, to name only four—
whose work even after several decades of wide-ranging re-examination and recuperation remains
central to the institutional reception of modernism. My intention has been to make visible a rather
different set of concerns within the writing of modernity. Modernist Informatics traces a circuit in
which the term “modernism” is not to be regarded as a stable name for a single mode or genre of
writing, but rather—like “information”—as the residue of a series of definitions, none of them


adequate to encompass all possible forms or instances of the protean phenomenon they aim to
circumscribe. That said, the main figures discussed here do have one important feature in common, in
that all of them at one time or another were intimately involved with official networks of information
control. Joseph Conrad—the child of exiled political dissidents—spent his early working life
navigating the communication routes of the British Empire. Ford Madox Ford saw service both in the
War Propaganda Bureau and as an officer in the Welsh Regiment. Graham Greene worked for the
Ministry of Information and for the British Secret Intelligence Service, while the energetic founders
of Mass-Observation found themselves absorbed into more mundane government work. The director
and impresario John Grierson worked as a telegraph operator on a minesweeper during the First
World War before becoming the British Commonwealth’s leading theorist of public information. And
Elizabeth Bowen reported covertly on morale in neutral Ireland. Each of the figures considered here
was in some way uniquely positioned on the margins of class or nationality. All found ways to reinvent themselves through creative work that drew upon intimate knowledge of the informatic
structures of the British state. By reading their works together, I aim to demonstrate how a specific
kind of modernist culture emerged at points of intersection between private, public, and official

media channels in the first half of the twentieth century.
*
Anglophone novelists have been struggling to understand the nature of information at least since
Joseph Conrad’s Chance (1914), in which Charles Marlow pauses to reflect on the epistemological
conditions of his own storytelling. The passage is part of Marlow’s commentary on the story of the
disgraced banker de Barral, which is itself the context for a romance plot involving de Barral’s
daughter Flora:
“You must not think,” went on Marlow after a pause, “that on that morning […] I went consciously in my mind over all this, let us
call it information; no, better say, this fund of knowledge which I had, or rather which existed, in me in regard to de Barral.
Information is something one goes out to seek and puts away when found as you might do a piece of lead: ponderous, useful,
unvibrating, dull. Whereas knowledge comes to one, this sort of knowledge, a chance acquisition preserving in its repose a fine
resonant quality…. But as such distinctions touch upon the transcendental I shall spare you the pain of listening to them. […]”30

With his fastidious revisions, Marlow seeks to articulate a distinction between information and
knowledge as differing modes of possession. Knowledge, he decides, isn’t the kind of acquisition that
can be “had” in a way that keeps it separate from the self, but rather something that “exists” in him in
some unspecified way. Slipping into the financial idiom that permeates Chance, he describes
knowledge as a “fund,” a word he elsewhere uses only for affective states. (Mrs. de Barral, he tells
us, was once a woman “with a fund of simple gaiety”; later he complains that women do not share
“that fund of at least conditional loyalty” with which men are said to regard each other.) 31 If
information, for Marlow, resembles base metal, the fund that is knowledge suggests active
(“resonant”) capital: capital invested in the expectation of a return rather than hoarded unproductively
in the form of raw material. Both information and knowledge, in this schematic view, are to be
understood as economic phenomena but with different implications: one is laboriously and tediously
sought and hoarded; the other simply “comes to one” as a “chance acquisition.” Marlow, evidently, is
thinking of financial matters; pausing for a moment to explain a point to his listeners, he does so under
the influence of a financial idiom.32
One problem with this distinction, of course, is that information is not at all like lead. It is neither

fungible (one piece of information is not interchangeable with any other) nor is it alienable in the
same way. That difficulty may help to explain Marlow’s rather de haut en bas tone—and the
authorial ellipsis—when he chooses to let the subject drop. In Conrad’s writing, the figure who most
frequently embodies the lamentable ascendancy of information is the newspaper reporter, and indeed
Marlow’s meditation on information and knowledge comes immediately after his description of an
encounter with a “pressman” acquaintance outside the trial of de Barral. In the course of their
discussion, he obtains a piece of information: the fact that de Barral, when sentenced, “permitted
himself his very first and last gesture in all these days, raising a hard-clenched fist above his head.”
The pressman, according to Marlow, has failed to grasp the significance of the gesture: “Is it ever the
business of any pressman to understand anything? I guess not. It would lead him too far away from the
actualities which are the daily bread of the public mind.” For Marlow, however, de Barral’s gesture
matters far more than as a simple “actuality.” He takes it as evidence of an “imagination […] at last
roused into activity,” and what interests him—as a matter not of contingent news but of universal
human nature—is the idea that the stolid and unimaginative de Barral has finally understood the tragic
aspect of his precipitous fall. The moment of anagnorisis is lost on the pressman who is obliged
simply to write “a readable account” and be done with it. But it is not lost on Marlow, in whom it
resonates between mind and body, provoking “a thrill very much approaching a shudder.”33
Such thrills and shudders mark, for Conrad, a psychic boundary to pass beyond which information
must become something else entirely. Sometimes he calls that something else knowledge, sometimes
sympathy, and sometimes understanding. In any case, art seems to be the only possible vehicle:
newspapers, for instance, remain firmly in the realm of information. Consider the essay “Autocracy
and War” (1905), which turns a scathing eye on the “cold, silent, colourless” press coverage of the
Russo-Japanese War:
In this age of knowledge, our sympathetic imagination, to which we can alone look for the ultimate triumph of Concord and
Justice, remains strangely impervious to information, however correctly and even picturesquely conveyed. As to the austere
eloquence of a serried array of figures, it has all the futility of eloquence without force. It is the exploded superstition of
enthusiastic statisticians.34

In the course of explaining how the “sympathetic imagination” is “impervious to information,” Conrad
makes the same stylistic move that he will later repeat in Chance, here turning not to the idiom of

finance but to a series of military metaphors. By superimposing the language of warfare on an
indictment of information-saturated media, the essay makes metaphor into critique: “serried array”
projects the image of ranked troops over the “austere eloquence” of enumeration, while “exploded”
holds the real violence of the war in the same frame as the “enthusiastic statisticians” whose abstract
calculations make that violence possible. In both cases, the prose demonstrates what it declares: that
literary language is capable of working upon the reader in ways that are quite different from the
straightforward model of affectless, objective transmission implicit in informational forms such as the
newspaper story and the statistical report.
News, according to Walter Benjamin, was the form of modern writing that had most powerfully
affected imaginative literature. The rising power of the middle class and its increasing control of the
press were, for Benjamin, the key factors in the development of a “new form of communication”
inimical not only to the oral tradition of the epic and folktale but also to the bourgeois narrative form
which had already supplanted that tradition: the novel. “This new form of communication,” he wrote,
“is information.”35 [Diese neue Form der Mitteilung ist die Information.] Benjamin used the term
die Information here to denote a specifically new form of communication, a specificity somewhat


obscured by the standard English translation of the collected essays, which collapses the distinction
by rendering both Information here and Mitteilung elsewhere as “information.”36 What Benjamin
chose to call die Information exhibited a distinct mode of social operation that he considered unique
to the industrial phase of modernity:
The intelligence [die Kunde] that came from afar—whether the spatial kind from foreign countries or the temporal kind of
tradition—possessed an authority which gave it validity, even when it was not subject to verification. Information [die
Information], however, lays claim to prompt verifiability. The prime requirement is that it appear “understandable in itself.” Often
it is no more exact than the intelligence of earlier centuries was. But while the latter was inclined to borrow from the miraculous,
it is indispensable for information to sound plausible. Because of this it proves incompatible with the spirit of storytelling [dem
Geist der Erzählung].

“Information” here gains its value not from the authority that communicates it but from its freshness,
verifiability, self-sufficiency, and plausibility. For Benjamin, information collapses space and

tradition into an undifferentiated simultaneity, unlike storytelling which “thrives for a long time in the
milieu of work—the rural, the maritime, and the urban” and “does not aim to convey the pure essence
of the thing, like information or a report. It sinks the thing into the life of the storyteller, in order to
bring it out of him again.” The value of information thus inheres “only in the moment in which it was
new.”37 (“Poetry,” Ezra Pound had insisted a couple of years earlier, “is news that stays news.”)38
Benjamin’s essay instantiates a more general trend in modern literary engagements with the
discourse of information, since it aims to define this apparently new phenomenon by contrast with
some favored quality that it threatens to displace. E. M. Forster—who was not altogether kind about
Conrad’s studiously uninformative prose style—considered that different forms of literary production
might partake more or less of the qualities of information. “Books are composed of words,” he wrote,
“and words have two functions to perform: they give information or they create an atmosphere.” As
an example of something close to “pure information,” Forster suggested the word “stop” on a
tramway sign, which he went on to compare with another hypothetical sign reading “Beware of
pickpockets, male and female.” The latter sign, he thought, could be said to contain something more
than information: “Who can see those words without a slight sinking feeling at the heart? […] Besides
conveying information it has created an atmosphere, and to that extent is literature.” It was Forster’s
view that dramatic writing contained on average a higher proportion of information than lyric poetry,
the novel a higher proportion than dramatic writing, the newspaper a higher proportion still, followed
by advertisements and price lists and signs. He had not yet had to contend with the traffic lights
installed across London in the early 1930s, which completed the reduction of sign to signal and set
Virginia Woolf’s dinner companions wondering whether the gallery-going public might lose its
appreciation of color as a result.39
A little later in the same essay, Forster turns to the more difficult question of what gets left over
when the information value in language is accounted for:
What is this element in words that is not information? I have called it “atmosphere,” but it requires stricter definition than that. It
resides not in any particular word, but in the order in which words are arranged—that is to say, in style. It is the power that words
have to raise our emotions or quicken our blood. It is also something else, and to define that other thing would be to explain the
secret of the universe. This “something else” in words is undefinable. It is their power to create not only atmosphere, but a world,
which, while it lasts, seems more real and solid than this daily existence of pickpockets and trams. Information is true if it is
accurate. A poem is true if it hangs together. Information points to something else. A poem points to nothing but itself.

Information is relative. A poem is absolute.40

What these moments in Conrad, Benjamin, and Forster have in common is the rhetorical invocation in


each instance of “information” to clear space in an increasingly congested media ecology for some
special quality, proper to literary art, which remains “undefinable” except by negation. Conrad’s
knowledge, Benjamin’s Erzählung, and Forster’s atmosphere each stake a claim to some additional
quality of resonance (Conrad), miraculousness (Benjamin), or creativity (Forster), over and above
those possessed by “information.” In each case, a concern with information surfaces where the
survival of literariness is itself in question. It is information against which the idea of the literary
must be defined, as the fantasy of what Forster calls “pure information” is put to work in order to
evoke by contrast the literary value it lacks.
In fact, as I shall argue, the relationship between information and fiction was rather more complex
and interesting than these basic oppositions would suggest. From the end of the nineteenth century
onward, powerful new technologies and vast systems were developed in order to govern and monitor
the new world of information. Far from maintaining a pure aesthetic distance from these new
informatic webs, many writers and artists found themselves increasingly entangled. Benjamin,
Conrad, and Forster had good reason to familiarize themselves with such information systems, as we
shall see. So too did Ford Madox Ford, Graham Greene, John Grierson, and Elizabeth Bowen. Each
of these figures was profoundly affected both by the political regulation of information systems and by
the use of information systems as instruments of political authority. Their writings began to respond to
those phenomena in new ways. Where “information” had usually been considered antagonistic to
aesthetic value, or at best as the inert workaday material upon which the work of art might stamp such
value, by the 1930s it had become possible to conceive of an information aesthetic, partly because of
the ways in which modernist writers and filmmakers had begun to reflect upon and make visible the
informatic structures governing their own work. Rather than occupying the opposite end of the
spectrum described by Forster, literary narrative in the early twentieth century interacted in
unexpected and fascinating ways with the government of information.
One writer who suddenly found himself at the center of a global information system was the young

T. S. Eliot, who in March 1917 began working in the Colonial and Foreign Department of Lloyd’s
Bank. In September 1919, however, Eliot moved to the new “Information” Department on the first
floor of the bank’s Lombard Street headquarters. 41 (The address for cables, he told Dorothy Pound,
was “Eliot, Information, Branchage, Stock, London.”)42 Early the following year, in a letter to his
mother, he noted that the department was short-staffed:
Next week I shall have an assistant and a typist to write my letters and do card indexing, but last week I have had to struggle
through chaos myself, receiving hundreds of reports from Branches of the bank, classifying them, picking out the points that
needed immediate attention, interviewing other banks and Government Departments, and trying to elucidate knotty points in that
appalling document the Peace Treaty.43

Lawrence Rainey has pointed out that this comment about “card indexing”—a “relatively new office
procedure”—is interesting in that it places Eliot at the heart of the bureaucratic control revolution
described by Max Weber (and, more recently, by Beniger). 44 But Rainey also misdates Eliot’s letter
to September 1919, perhaps because that was the month when the first part of “Tradition and the
Individual Talent” was published in The Egoist. It might be thought that Eliot’s letter clarifies one of
the more abstruse points in that essay: the idea that the “existing monuments” of a culture “form an
ideal order among themselves, which is modified by the introduction of the new (the really new)
work of art among them.”45
The reason this passage remains puzzling is that “monuments” is an odd word in an essay