ChapterPDF Available

Learning Through Interactive Digital Narratives



One of the most effective ways of conveying information and learning is through storytelling and narratives. Thus, naturally, narratives have generated great interest and are being incorporated into educational computer games. Storytelling in educational games is used as a method for improving students' motivation and is considered an important component of learning. This chapter explores the research undertaken to integrate educational content into games that make use of narratives. It examines how a predefined set of Learning Objectives (LOs) are integrated into an interactive detective story using the Storytelling for educAtional inteRventions (STAR) framework. The STAR framework proposes to have a sequence of puzzle and red herrings during which LOs are taught. The chapter then describes our approach to integrating the educational content in an interactive digital narrative (IDN) game. Finally, it presents the results of evaluating explicit knowledge acquisition through the gameplay of the Global Hamdwashing Day (GHD) game.
Interactive Digital Narrative
The book is concerned with narrative in digital media that changes accord-
ing to user input—Interactive Digital Narrative (IDN). It provides a broad
overview of current issues and future directions in this multi-disciplinary
eld that includes humanities-based and computational perspectives. It
assembles the voices of leading researchers and practitioners like Janet
Murray, Marie-Laure Ryan, Scott Rettberg and Martin Rieser. In three
sections, it covers history, theoretical perspectives and varieties of prac-
tice including narrative game design, with a special focus on changes in
the power relationship between audience and author enabled by interactiv-
ity. After discussing the historical development of diverse forms, the book
presents theoretical standpoints including a semiotic perspective, a proposal
for a specic theoretical framework and an inquiry into the role of arti-
cial intelligence. Finally, it analyses varieties of current practice from digital
poetry to location-based applications, artistic experiments and expanded
remakes of older narrative game titles.
Hartmut Koenitz is Assistant Professor of Mass Media Arts in the Grady
College of Journalism and Mass Communication at the University of
Georgia, USA.
Gabriele Ferri is a Postdoctoral Researcher in the School of Informatics and
Computing at Indiana University Bloomington, USA.
Mads Haahr is Lecturer in the School of Computer Science and Statistics at
Trinity College Dublin, Ireland.
Didem Sezen is Assistant Professor in the Faculty of Communications at
Istanbul University, Turkey.
Tonguç brahim Sezen is Assistant Professor in the Faculty of Communica-
tion at Istanbul Bilgi University, Turkey.
Routledge Studies in European Communication Research
Edited by Nico Carpentier, Vrije Universiteit Brussel, Belgium and Charles
University, Czech Republic, François Heinderyckx, Université Libre de
Bruxelles, Belgium and Claudia Alvares, Lusofona University, Portugal.
Series Advisory Board: Denis McQuail, Robert Picard and Jan Servaes
Published in association with the
European Communication Research
and Education Association (ECREA),
books in the series make a major
contribution to the theory, research,
practice and/or policy literature. They
are European in scope and represent
a diversity of perspectives. Book pro-
posals are refereed.
1 Audience Transformations
Shifting Audience Positions in
Late Modernity
Edited by Nico Carpentier, Kim
Christian Schrøder and Lawrie
2 Audience Research
Between Innovation and
Edited by Geoffroy Patriarche,
Helena Bilandzic, Jakob Linaa
Jensen and Jelena Juriši
3 Multiplayer
The Social Aspects of
Digital Gaming
Edited by Thorsten Quandt and
Sonja Kröger
4 Mapping Foreign
Correspondence in Europe
Edited by Georgios Terzis
5 Revitalising Audience Research
Innovations in European
Audience Research
Edited by Frauke Zeller,
Cristina Ponte and Brian
6 Radio Audiences and
Participation in the Age of
Network Society
Edited by Tiziano Bonini and
Belén Monclús
7 Interactive Digital Narrative
History, Theory and Practice
Edited by Hartmut Koenitz,
Gabriele Ferri, Mads Haahr,
Didem Sezen and Tonguç
brahim Sezen
Interactive Digital Narrative
History, Theory and Practice
Edited by
Hartmut Koenitz, Gabriele Ferri,
Mads Haahr, Didem Sezen
and Tonguç brahim Sezen
First published 2015
by Routledge
711 Third Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2015 Taylor & Francis
The right of the editors to be identied as the authors of the editorial material, and of the
authors for their individual chapters, has been asserted in accordance with sections 77 and 78
of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any
form or by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or registered trademarks,
and are used only for identication and explanation without intent to infringe.
Library of Congress Cataloging in Publication Data
Interactive digital narrative : history, theory, and practice / edited by Hartmut Koenitz,
Gabriele Ferri, Mads Haahr, Digdem Sezen and Tonguc Ibrahim Sezen.
pages cm. — (Routledge studies in European communication research and education ; 7)
Includes bibliographical references and index.
1. Digital storytelling. 2. Interactive multimedia. 3. Narration (Rhetoric)
4. Digital media—History. I. Koenitz, Hartmut, editor. II. Ferri, Gabriele, editor.
III. Haahr, Mads, editor. IV. Sezen, Digdem, editor. V. Sezen, Tonguc Ibrahim, editor.
QA76.76.I59I549 2015
006.7—dc23 2014046169
ISBN: 978-1-138-78239-6 (hbk)
ISBN: 978-1-315-76918-9 (ebk)
Typeset in Sabon
by codeMantra
Foreword ix
Acknowledgments xv
1 Introduction: Perspectives on Interactive Digital Narrative 1
Introduction: A Concise History of Interactive
Digital Narrative 9
2 The American Hypertext Novel, and Whatever
Became of It? 22
3 Interactive Cinema in the Digital Age 36
4 The Holodeck is all Around UsInterface Dispositifs in
Interactive Digital Storytelling 51
Introduction: The Evolution of Interactive
Digital Narrative Theory 67
vi Contents
5 Narrative Structures in IDN Authoring and Analysis 77
6 Towards a Specic Theory of Interactive Digital Narrative 91
7 Emotional and Strategic Conceptions of Space in
Digital Narratives 106
8 A Tale of Two Boyfriends: A Literary Abstraction Strategy
for Creating Meaningful Character Variation 121
9 Reconsidering the Role of AI in Interactive
Digital Narrative 136
Introduction: Beyond the Holodeck:
A Speculative Perspective on Future Practices 151
10 Interaction Design Principles as Narrative Techniques for
Interactive Digital Storytelling 159
11 Posthyperction: Practices in Digital Textuality 174
12 Emergent Narrative: Past, Present and Future of
an Interactive Storytelling Approach 185
13 Learning through Interactive Digital Narratives 200
14 Everting the Holodeck: Games and Storytelling in
Physical Space 211
Contents vii
15 Narrative Explorations in Videogame Poetry 227
16 Artistic Explorations: Mobile, Locative
and Hybrid Narratives 241
17 Remaking as Revision of Narrative Design
in Digital Games 258
Contributors 272
Index 000
Nick Montfort
The essays in this collection further our understanding of how computers
can narrate responsively and in profoundly new ways. While the articles
are organised into those emphasising the historical, the theoretical and the
practical, the editors have helpfully identied ways to also read across the
three sections by following common threads. In addition, many essays deal
quite substantially with two or all three aspects.
Consider a few of the ways in which these three categories overlap: Scott
Rettberg’s ‘historical’ essay on the American hypertext novel documents his
‘practical’ experience as an author of The Unknown; Marie-Laure Ryan
develops her ‘theoretical’ discussion with reference to the ‘history’ of play-
elds in digital games; Janet Murray situates her ‘theoretical’ discussion by
considering a rich ‘history’ of women and their boyfriends in interactive
and noninteractive media; Nicolas Szilas draws on his system-building prac-
tice to ‘theorise’ highly interactive digital narrative; and those whose work
appears in the practice section have informed their system-building and art-
making with an awareness of both history and theory. While this book is
organised according to whether history, theory and practice are stressed—
and it needs to be somehow organised—these are not exclusive silos, and
they are not shown to be. Rather, they are the bases from which connections
are made. The work the authors and editors have done moves us beyond the
popular chant of “theory and practice, which really does nothing but jux-
taposes these two without explaining their relationship, to actually connect
them, and history as well, allowing all three to inform one other.
Having noted the above, this collection is also important because of how it
reveals a lack—not any lack inherent in this particular project overall, or in
any essay, but rather something that is incomplete in this intellectual area,
which is still in the process of coalescing into a eld. This nascent eld has
framed but not solved its initial conceptual problems. That is part of what
makes it exciting. This book, after all, is not a festschrift for a body of work
that has essentially been concluded, but it is an important part of the foun-
dation for those of us still establishing major ideas, directions and practices.
x Nick Montfort
A sign that this is not a well-established and xed eld is that those of
us in this area have not yet agreed on its name. Interactive digital narrative,
whether in short form (IDN) or long, is in certain ways similar to intelligent
narrative technologies (the name of a workshop series), interactive drama
(pioneered by Oz Project participants but also taken up by David Cage),
interactive storytelling (the name of a conference series and considered
so conventional that it is abbreviated to IS in Chapter 12), and narrative
games (a term often used at the Game Developers Conference, for instance,
although it appears only in passing in this book). These different terms sug-
gest their own different emphases and connections—to articial intelligence
or narrative theory, or traditions and theories of drama, or ludic interaction
and videogaming—while, at the same time, the people working under these
banners, and others, do truly share many common assumptions, use many
similar techniques, and are often informed by each other’s work.
At certain points, developing some sort of neologism to serve as a stan-
dard term can be extremely valuable. A new name can serve to produc-
tively expand some concept, dened by previous terms, that is now seen to
be too narrow. This happened in different ways with both cybertext and
electronic literature, two categories that have rich intersections with each
other and with interactive digital narrative. These two terms were also suc-
cessful because they hearkened back to a relevant history and foundation
(-text, literature), contextualising the eld they described and grounding it
in known cultural practices, while also embracing recent changes and the
future cyber-, electronic).
While this type of naming activity can be valuable, and may eventually be
important, there seems to be no need to immediately unify everyone’s name
for this activity or eld, or to standardise on a single term, old or new, for
all that we do. Interactive digital narratives is a ne term, existing alongside
several that are slightly different, because it too makes connections to well-
known aspects of culture, to history, while also looking forward. It gathers
together academic work from a wide variety of disciplines and from the
practice of artists, writers and game-makers within and beyond industry.
Certain other terms—as long as they aren’t attempts to retrench and to wall
off and ignore relevant, related work—can be ne, too, and can show how
many different perspectives are involved in the same enterprise, or at least,
very similar enterprises.
Yes, there are some related activities that would seem to be left out of
IDN strictly interpreted: Interactive digital poetry and art that is nonnarra-
tive, story generation systems that are noninteractive, live interactive drama
experiences that may be scripted and rule-based but are enacted by peo-
ple without the use of computers. Of course, some focus is always needed
for conversation and progress. The existence of other, related elds simply
suggests to me that this diversity in names is a good thing and can allow
researchers and practitioners to make different connections at different
Foreword xi
That is, our diversity of names, and the use of IDN here, is productive.
These are not serving as exclusive decrees; rather, they facilitate intersections
and connection, just as the conjunction of history, theory and practice do.
Since there is no need to create history/theory/practice intersections or to
worry ourselves about establishing a single standard name for the whole
eld and everything related, what needs to change in IDN?
The urgent question for interactive digital narrative is what it truly aspires
to. What are its goals? How does it seek to change or maintain society? Is it
here to reassure us or provoke us? In this sense, it is time to rene our asso-
ciations, to discern, and to distinguish, even as we celebrate intersections in
other ways.
Will the work that we are doing end up as a (possibly protable) exten-
sion of the worst properties of current media—as with “The Family” in Ray
Bradbury’s Fahrenheit 451, in which the numbed Mildred Montag reads
a script to characters on three wall-sized displays, becoming ever more
enthralled? In such a scenario, which is participatory although not truly
interactive, the absence of fourth wall is hardly what is important—no mat-
ter that Mildred bemoans having only three walls installed. The real issue
is that the system provides a sort of engagement, immersion and ability to
play along in supercial ways, without offering any true agency, any new
aesthetic vision or any new philosophical or political concept.
Our eld doesn’t lack passion and enthusiasm on the part of researchers,
theorists and artists who seek to dig into every possibility. But, why are we
doing this?
What this eld needs now is to become a movement, or perhaps more
than one if there is no single common cause. Is IDN out to bust up the game
industry and fundamentally change it, or does it seek to simply buttress
what exists? Is it supposed to be pure entertainment or something beyond
that? Does it care more about empowering the disenfranchised or about
training the military, and does the answer to this question depend upon
how much funding is available for each option? Is it about human nature
and our cultural world, or a distraction from them? Is IDN fashion, always
seeking to set and follow trends, or can it open a chasm to the timeless? Is it
a pleasing but meaningless salve to soothe workday aches or a way to model
a better society?
Something I have heard Scott Rettberg speak about is that electronic lit-
erature is a migration rather than a movement. This description is inclu-
sive, and it too connects the past with the future, as there is somewhere to
migrate from (print literature) and new possibilities of where to migrate to
(the digital medium). It means that people do not have to have a common
artistic vision or shared goals at the highest level to discuss their experiences
xii Nick Montfort
and learn from one another. Whether this is precisely true of the related eld
called IDN or not, IDN is certainly more a migration than a movement. The
migrants may come from a wider or at least different variety of disciplines
and practices, but it is the arrival at user-controlled computational narra-
tive, not the underlying drive, that denes this eld.
People making videogames, interactive TV, interactive story generators and
all manner of other interactive narrative experiences on computers can con-
sider themselves in the same eld—the same migration—and they would
be correct. But people who have opposite answers to these questions I have
posed—some trying to retrench and focus entirely on earnings, some try-
ing to innovate and make progress toward new cultural developments and
understandings—may not be in the same eld. They certainly aren’t in the
same movement.
The technologies to model a utopia may not be the same ones needed to
entertain, or, at least, the intersection of these may be slight and not very
interesting. Let me suggest that people need to communicate, collaborate,
and be empowered with an understanding of computation to use interac-
tive digital narrative as a way of designing a new society or ameliorating
social ills. They have to be builders, makers and coders, rather than just
being given an experience. Working toward such a goal involves collec-
tive thought, deliberation and consciousness on the part of the interactors.
Mildred Montag did not need any of this in order to feel a part of her video
narrative and to be entertained. The techniques that entertained her, we
have to imagine, were very different. It was a fundamentally different design
The community of writers using Twine, an effective and straightforward
hypertext creation tool, one that is free software and outputs HTML for the
Web, provide one example of a more movement-like group. While authors
seem glad to be part of the interactive ction (IF) community (the former
participate in and do well in its annual competition, and the creator of Twine
is an IF author), the work they do is radically different as a writing practice
and as interactive design. Recent Twine work strongly focuses on issues such
as cultural desolation, personal identity, alienation, and lack of choice much
more thoroughly than has been done in parser-based interactive ction, even
though individual parser-based works have treated these themes.
One might try to locate the difference between Twine production and
parser-based production (most of which is done using the Inform 7 system)
in particular aspects of writing style, in Inform 7’s ability to accept input and
understand typed user commands, or in the choice of themes. Perhaps the
most signicant distinction is that Twine authors are more allied in terms
of what they want to accomplish. They create unusual fantasy worlds of
Foreword xiii
the sort that appear in other digital work, but theirs cut into personal and
cultural questions in obvious and jarring ways. They are often seeking to do
work that is personally or politically meaningful, to provoke the reader into
understanding other people’s cognition, emotion, and experiences.
If the Twine community agrees with my (outsider) assessment or some-
thing like it, they might realise that although it’s good to trade tips and
technical tricks with other Twine users, it isn’t the development tool, but
their more abstract goals, that unite them as a group. They might join with
others working in different sorts of IDN and other sorts of poetic, narra-
tive, and digital development. They might decide that someone writing a
Twine game that is, for instance, a humorous Choose-Your-Own-Adventure
parody aimed only at laughs isn’t really part of their movement, while a
digital poet working with a different system could be.
The more retrograde and more aspirational types of IDN can, of course,
both exist, and there can be groups, even large and organised ones, working
in both of these ways. But as people in the eld determine who to confer-
ence with and collaborate with, they should both consider the particular
valence of IDN being undertaken, as well as think about who the others
are, or would be, in their movement. I am not suggesting that we shun those
with different goals, but rather take the opportunity to form alliances based
on deep values and principles, noticing where we have already done so, and
reaching out across different practices and media.
The essays in this book show that IDN has tremendous capability and is
continuing to excel. As someone compelled to learn more about this eld,
while you read and learn from what is included here, consider, please, how
this capability will be used. Consider in what direction you would like to
move the eld, and what you and like-minded researchers and artists—in
your movement—can do to help it progress.
The editors would like to give their sincere thanks to the ECREA book series
editors, especially Nico Carpentier, for their positive critique and helpful
advice during the formative period of this volume, as well as their support
in seeing this work to completion.
The editors are also grateful for Benjamin Larkin, whose outstanding
abilities in copy-editing and proofreading were crucial for the readability of
this volume.
1 Introduction
Perspectives on Interactive Digital
Hartmut Koenitz, Gabriele Ferri, Mads Haahr,
Didem Sezen and Tonguç brahim Sezen
Interactive Digital Narrative (IDN) connects artistic vision with technology.
At its core is the age-old dream to make the fourth wall permeable; to enter
the narrative, to participate and experience what will unfold. IDN promises
to dissolve the division between active creator and passive audience and
herald the advent of a new triadic relationship between creator, dynamic
narrative artefact and audience-turned-participant. Within this broad vision
of fully interactive narrative environments through the use of digital tech-
nologies, IDN aggregates different artistic and research directions from mal-
leable, screen-based textual representations to the quest for virtual spaces in
which human interactors experience coherent narratives side by side with
authored narrative elements and synthetic characters.
The IDN vision is as much about narrative and control as it is about bal-
ance. Indeed, the quest for the right artistic measure, for equilibrium between
agency and a coherent, satisfying experience, might be the ultimate chal-
lenge of the eld. Yet, the artistic challenge does not exist in isolation and
is joined by technological and analytical challenges. IDN is a truly interdis-
ciplinary eld, which includes scholars and practitioners with backgrounds
in multiple disciplines: from literary studies to computer science and ne
art. While guiding visions have been described, sometimes even heralded, in
various forms for quite some time—for example, the image of Alice entering
the rabbit hole or Borges’ innite labyrinth in the form of a novel—it is only
with the advent of computer technology that its realisation seems possible,
and constant developments in computer technologies seem to put them ever
closer to our reach. Indeed, digital media has radically changed the way nar-
rative content is being created, shared, experienced and interpreted.
In her seminal work Hamlet on the Holodeck (1997), Janet Murray notes
that digital media is inherently procedural and participatory, referring to the
capacity of computers to execute a series of commands and react to user
input. While procedurality affords digital creators the expressive power to
dene initial conditions and rules under which an interactive work executes
and reacts to input, IDN bestows cocreative power on its users through
2 Hartmut Koenitz et al.,
interaction and therefore reshapes the relation between creator, work and
audience in a way that far surpasses aspects of interpretation and reader-
response theory, but whose exact extent is a subject of scholarly debate. The
complex relation between authorial control and the power of interactive
agency is therefore an underlying topic in all three parts of this collection of
essays, which focus on history, theory and practice.
While IDN has been an artistic practice and a topic of scholarly inquiry
for more than two decades, it is still in its infancy compared to other nar-
rative forms like the stage drama, the printed book or the moving image.
As a technical and artistic challenge and opportunity, advances in the IDN
eld depend on the combined effects of developments in different parts
within the greater eld. However, progress in these different areas has his-
torically been uneven. For example, graphical representation has seen rapid
improvements from the humble beginnings as text on the screen to current
cinematic- quality 3D depictions driven by simulations of highly realistic
physics. Whilst the progress in graphics and physics is no less than aston-
ishing, the same cannot be said for the larger challenge of creating specic
narrative forms to produce compelling and captivating experiences: in this
regard, the pace of development has been unsteady and slow. The virtual
environments used for many contemporary interactive narratives are real-
istic, dynamic and feature high delity in terms of their visual presentation
and physical mechanics. However, the narratives and characters they host
remain shallow, static and lacking in believability, dramatic engagement and
narrative development in comparison.
Indeed, while clearly eclipsed in visual presentation, the strong narrative
of early titles like Zork (1982) holds up well even today. Maybe this fact
should not surprise us, as resources for work on improving the graphical
representation have been more readily available than for the more artis-
tic problem of narrative development, spurred originally by the US Air
Force’s interest in convincing visuals for ight simulators (Myers, 1998).
Research in IDN ideally combines technical development and advances in
artistic expression, as well as the expansion of analytical perspectives; and
historically, it has been difcult to nd resources for such interdisciplinary
projects. Funding, however, is only one aspect of the problem. Cinematic
visualisation and real-world perceptions provide an ideal to aspire to for
graphical representations. A comparable, shared goal on the side of narra-
tive development and resulting form is elusive. Janet Murray’s proposal of
the ‘Holodeck’ (Murray 1997), an imaginary future form of entertainment
rst depicted in the TV series Star Trek: The Next Generation that immerses
its audience in a dynamic, reactive narrative, has perhaps been rejected more
often (Ryan, 2001; Aarseth, 2004; Spector, 2013) than it has been tacitly
embraced (Mateas, 2001; Nitsche, 2008). Other visions, like constructive
hypertext or interactive drama, share this fate. However, while the absence
of a canonical set of narrative structures specic to IDN can be problem-
atic, the lack of a unanimously shared vision also represents an opportunity
Introduction 3
because it provides space for experimentation and creative license to create
new forms.
If the dening artistic moment of the book was the advent of the novel
in the 17th century
and of lm was the invention of montage (Eisenstein,
1949), a similar breakthrough is still elusive in IDN, and maybe there never
will be a comparable moment in this eld. Instead, we might see existing
design modes (e.g., third-person versus rst-person perspective, modes of
audience participation and novel narrative structures) grow into mature
artistic conventions applied in a conscious way by a new generation of
authors. A possible dening milestone for IDN might even be the emergence
of a consistent group of practitioners, IDN auteurs or cyberbards, to use
Murray’s term (1997), who feel more condent with the notion of relin-
quishing some of their authorial control to users, players and interactors,
and see themselves not as the creators of singular visions, but as designers
of expressive potential.
Analytical perspectives have developed considerably since the 1980s,
when the rst scholars with backgrounds in design and the humanities
became interested in the topic. Where early treatments of the topic focused
on the comparison to older narrative practices, later works have become
increasingly more focused on specic aspects like space (Jenkins, 2004;
Nitsche, 2008; Ryan, in this Volume), on the particular manifestations
(Montfort, 2003), specic theoretical concepts (Koenitz, 2010) and the con-
nections to larger frameworks (Ryan, 2006; Koenitz et al, 2013a) and most
recently on particular theoretical aspects (Bruni and Baceviciute, 2013;
Mason, 2013; Ferri, 2013). Amongst this much needed focus, scholars in
this eld are also engaged in a meta-reection on the dening characteristics
of IDN (Murray, 1997; Aarseth, 1997, 2012; Juul, 2011; Eskelinen, 2012;
Mateas, 2001; Ryan 2001, 2006; Frasca, 2003b; Crawford, 2004; Koenitz
et al., 2013b). A particular example of this discussion emerged in the early
2000s with the advent of computer game studies as a discipline. In that
debate, narrative-oriented and game-oriented approaches were framed as
a dichotomy, painting games through the simulative aspects as a “radically
different alternative to narratives as a cognitive and communicative struc-
ture” (Aarseth 2001). A group of game studies scholars (Aarseth, 2001,
2004; Juul, 1999; Eskelinen, 2001; Frasca, 2003a), opposing narrative-
centric views, adopted the name of ludologists; and thus the discussion
is often referred to as the ‘narratology vs. ludology debate.The very rst
ludological perspectives not only opposed the use of narratological con-
cepts to describe video games but, in their early forms, also described inter-
active narrative as practically impossible: “computer games [are] simply not
a narrative medium” (Juul, 1999, p. 1). Jesper Juul’s argument conated
two claims; notions derived from narratology—or related disciplines—are
not effective to read games, and games cannot convey narratives. The rst
claim followed from the need to legitimise game studies as an independent
academic discipline, thus dening it by contrast with others and establishing
4 Hartmut Koenitz et al.,
its own vocabulary. This was a move understood by Stuart Moulthrop
(2003) as a necessary “defensive maneuver (sic), however at the cost of
an “alarmingly narrow” point of view, one that carries the danger of creat-
ing “conceptual blind spots” (Jenkins, 2004). As game studies became a
recognised academic discipline in the following years, a gradual softening
of perspectives nally allowed Janet Murray to pronounce the end of the
debate (Murray, 2005). The second claim about the constitutive dichotomy
between play and narration—although retracted by Juul himself (2001)—
today remains inuential, especially in the professional practice of game
design where gameplay and narrative are often seen as opposing param-
eters. In this vein, game designer Ralph Coster, for example, denes narra-
tive in contrast to gameplay: “The commonest use of a completely parallel
medium that does not actually interact with the game system is narrative”
(Koster, 2012). He categorises the narrative parts of a game experience as
linear, noninteractive and in the sole function of rewarding players.
Even after years of research and discussion, the coupling of narration and
interaction can still spark provocative debates that require our attention.
Therefore, the practical and ontological analogies and differences between
interactivity and narration warrant further academic inquiry. Likewise, the
relationship between static and procedurally generated narratives calls for
more attention. In this respect, a more holistic view of IDN, foregrounding
how digital means enable interactive forms of narrative, could also contrib-
ute to the ludological discussion.
This volume covers a diverse and vibrant eld that has continually grown
since the late 1970s, from the rst text-based Interactive Fiction to such
forms as Hypertext Fiction, Interactive Cinema, Interactive Installations,
Interactive Drama and Video Game Narrative.
The book is structured in three parts. The rst part is historical and
addresses how forms of IDN emerged over the years as distinct phenomena
and how the transformations of digital media shaped the current forms.
Scott Rettberg examines hypertext novels and poems, offering an histori-
cal perspective on their technical development and literary fruition, while
Chris Hales describes the historical development of interactive cinema with
a focus on the impact of digital technology on this form of IDN. Finally,
Udi Ben-Arie and Noam Knoller offer a diachronic perspective on the user-
facing aspects in IDN, foregrounding the aesthetic, experiential and herme-
neutic dimensions.
The book’s second part is theoretical. Theoretical enquiry into IDN started
with adaptations of established narratological perspectives, for example
neo-Aristotelian poetics (Laurel 1986, 1991; Mateas, 2001), post-classical
narratology (Ryan, 1999, 2001), African oral traditions (Jennings, 1996;
Introduction 5
Harrell, 2007) and French post-structuralism (Montfort, 2003). In recent
years, scholars have started to look beyond narratology to understand the
changes in narrative modalities afforded by IDN. Also, particular aspects,
such as spatiality, have come to the forefront of analytical work. The editors
introduce this section with an overview of these earlier approaches before
the book’s second part presents a range of current theoretical perspectives.
First, Gabriele Ferri proposes a common ground for narrative theory of IDN
by reexamining the similarities and differences with unilinear storytelling.
Hartmut Koenitz argues for a theoretical approach that is specic for IDN
based on cognitive science and cybernetics, while Marie-Laure Ryan dis-
cusses spatial representations as a key topic in interactive narratives. Janet
Murray analyses dynamics of relationships in literature and discusses their
application as a schema for IDN, before Nicolas Szilas in the nal theoreti-
cal chapter offers a critical perspective on the role of Articial Intelligence
in developing a future, better form of digital narrations.
The book’s third part is concerned with practice. When a new medium
appears, early practitioners often engage with it rst by extending existing
practices. In this way, early lm was used to show theatrical performances.
Eventually these modes of extension lead to distinct practices. As the writ-
ten text became more than a collection of printed pages in the form of the
novel, and lm became more than a theatrical performance through mon-
tage, it is no longer adequate to relegate IDN practices to the fringes of a
perspective centred on narrative in long-established media forms. The third
part of the book is intended to examine the wide range of current practices
and the emergence of IDN as a distinct phenomenon. Ulrike Spierling begins
this part of the book with a chapter that emphasises the importance of user
interface design for the IDN experience, as well as its implementation in
practice. Scott Rettberg describes current practices in electronic literature,
while Sandy Louchart, John Truesdale, Neil Suttie and Ruth Aylett report
on research and implementations of emergent narrative, based on autono-
mous intelligent virtual characters. Andreea Molnar and Patty Kostkova ask
how story-based learning is transformed by the encounter with truly mal-
leable narrative. Mads Haahr analyses examples of location-based games
that position digital narrative elements in the real world, and Didem Sezen
examines video game poetry. Martin Rieser puts the spotlight on distinctly
artistic uses of IDN, while Tonguç Sezen’s chapter on remakes alerts us to
the fact that IDN has already reached a self-reective state.
In addition to reading the book’s three parts in a linear order, the reader
can also follow specic trajectories across the whole volume, for example
on IDN and the human-computer interface (Knoller and Ben-Arie, Szilas,
Spierling, Haahr, Rieser), on literary aspects (Rettberg, Murray, Didem
Sezen), transformation of existing elds and self-reective practices (Hales,
Rettberg, Molnar and Kostkova, Tonguç Sezen), novel theoretical approaches
(Ferri, Koenitz), spatial aspects (Ryan, Haahr, Rieser) and critical/practical
perspectives on the role of articial intelligence (Szilas, Louchart et al.).
6 Hartmut Koenitz et al.,
As the development of procedural media progresses, the powers and
abilities of readers as interactors and authors as procedural creators are con-
stantly being shifted and rebalanced. Since we are aware of the continuous
advances in the IDN eld, a companion website at www.gamesandnarra- will provide a space for further discussion. IDN enhances
the experiential dimensions of human expression, with multimodal manifes-
tations, procedural generation and novel structures. Furthermore, technical
and artistic advances in interactive narratives open epistemological ques-
tions that require constant theoretical attention. As this volume attests, the
development—in every dimension—is continuous and shows no signs of
slowing down. And herein might lie the lasting attraction of the eld—to
further human expression by applying a range of human faculties, from the
invention of digital technology to the continuous development of hardware
and software to artistic treatment and critical reection.
1. Cervantes’ Don Quixote (1605) can be seen as the foundation of the modern
novel (Riley, 1962).
Aarseth, E. (1997) Cybertext: Perspectives on Ergodic Literature. Baltimore: Johns
Hopkins University Press.
Aarseth, E. (2001) Computer game studies, year one. Game Studies. 1(1).
Aarseth, E. (2004) Genre trouble. Electronic Book Review. 3.
Aarseth, E. (2012) A narrative theory of games. In Proceedings of the International
Conference on the Foundations of Digital Games. New York: ACM.
Blank, M. and Lebling, D. (1980) Zork I. [Interactive ction]. Cambridge, MA:
Bruni, L. E., and Baceviciute, S. (2013) Narrative intelligibility and closure in inter-
active systems. In: Koenitz, H., Sezen, T. I., Ferri, G., Haahr, M., Sezen, D. and
Çatak, G. (eds.) Interactive Storytelling: 6th International Conference, ICIDS
2013. Heidelberg: Springer.
Crawford, C. (2004) Chris Crawford on Interactive Storytelling. New Jersey: Pear-
son Education.
Eisenstein, S. (1949) Film Form: Essays in Film Theory. New York: Harcourt Brace.
Eskelinen, M. (2001) The gaming situation. Game Studies. 1(1).
Eskelinen, M. (2012) Cybertext Poetics: The ritical Landscape of New Media Liter-
ary Theory. New York: Continuum. C.
Ferri, G. (2013) Satire, propaganda, play, storytelling, notes on critical interactive
digital narratives. In: Koenitz, H., Sezen, T. I., Ferri, G., Haahr, M., Sezen, D. and
Çatak, G. (eds.) Interactive Storytelling: 6th International Conference, ICIDS
2013. Heidelberg: Springer.
Introduction 7
Frasca, G. (2003a) Ludologists love storties, too: Notes from a debate that never
took place. In Proceedings of the DIGRA 2003 Conference.
Frasca, G. (2003b) Simulation versus narrative. In: Wolf, M. J. P. and Perron, B. (eds.)
The Video Game Theory Reader. New York: Routledge.
Harrell, D. F. (2007) GRIOT’s tales of haints and seraphs: A computational narrative
generation system. In: Harrigan, P. and Wardrip-Fruin, N. (eds.) Second Person:
Role-Playing and Story in Games and Playable Media. Cambridge, MA: MIT
Jenkins, H. (2004) Game design as narrative architecture. Electronic Book Review.3.
Jennings, P. (1996) Narrative structures for new media: Towards a new denition.
Leonardo. 29(5). pp. 345–350.
Juul, J. (1999) A Clash Between Game and Narrative: a Thesis on Computer Games
and Interactive Fiction. MA Thesis. Institute for Nordic Languages and Litera-
ture, University of Copenhagen.
Juul, J. (2001) Games telling stories. Game Studies, 1(1), 45.
Juul, J. (2011) Half-real: Video Games Between Real Rules and Fictional Worlds.
Cambridge, MA: MIT Press.
Koenitz, H., Haahr, M., Ferri, G., and Sezen, T. (2013a) First steps towards a unied
theory for interactive digital narrative. In: Pan, Z., Cheok, A. D., Mueller, W., Iurgel, I.,
Petta, P., and Urban, B. (eds.) Transactions on Edutainment X. Heidelberg:
Koenitz, H., Haahr, M., Ferri, G., Sezen, T., and Sezen, D (2013b) Mapping the
evolving space of interactive digital narrative—From artifacts to categorizations.
In: Koenitz, H., Sezen, T. I., Ferri, G., Haahr, M., Sezen, D., and Çatak, G. (eds.)
Interactive Storytelling: 6th International Conference, ICIDS 2013. Heidelberg:
Koenitz, H. (2010) Towards a theoretical framework for interactive digital narrative.
In: Aylett, R., Lim, M. Y., Louchart, S., Petta, P., and Riedl, M. (eds.) Interactive
Storytelling: Third Joint Conference on Interactive Digital Storytelling, ICIDS
2010. Heidelberg: Springer.
Koster, R. (2012) Narrative Is Not a Game Mechanic. Available at: http://www.
Laurel, B. (1986) Toward the Design of a Computer-Based Interactive Fantasy
System. PhD Thesis. Drama Department, Ohio State University.
Laurel, B. (1991) Computers as Theater. Reading, MA: Addison-Wesley.
Mason, S. (2013) On games and links: Extending the vocabulary of agency and
immersion in interactive narratives. In: Koenitz, H., Sezen, T. I., Ferri, G., Haahr,
M., Sezen, D., and Çatak, G. (eds.) Interactive Storytelling: 6th International
Conference, ICIDS 2013. Heidelberg: Springer.
Mateas, M. (2001) A preliminary poetics for interactive drama and games. Digital
Creativity. 12(3). pp. 140–152.
Montfort, N. (2003) Twisty Little Passages: An Approach to Interactive Fiction.
Cambridge, MA: MIT Press.
Moulthrop, S. (2003) Stuart Moulthrop’s response. Electronic Book Review. 3.
Murray, J. (1997) Hamlet on the Holodeck: The Future of Narrative in Cyberspaace.
New York: The Free Press.
Murray, J. (2005) The last word on ludology v narratology in game studies. In Pro-
ceedings of the DiGRA 2005 Conference.
8 Hartmut Koenitz et al.,
Myers, B. A. (1998) A brief history of human-computer interaction technology.
Interactions, 5(2), 44–54.
Nitsche, M. (2008) Video Game Spaces. Inage, Play, and Structure in 3D Worlds.
Cambridge, MA: MIT Press.
Riley, E. C. (1962) Cervantes’s Theory of the Novel. Oxford: Clarendon Press.
Ryan, Marie-Laure(1999) Immersion vs. interactivity: Virtual reality and literary
theory. SubStance, 28(2). pp. 110–137.
Ryan, Marie-Laure(2001) Narrative as Virtual Reality: Immersion and Interactivity
in Literature and Electronic Media. Baltimore: Johns Hopkins University Press.
Ryan, Marie-Laure(2006) Avatars of Story. Narrative Modes in Old and New Media.
Minneapolis: University of Minnesota Press.
Ryan, Marie-Laure (2001) Beyond myth and metaphor—The case of narrative in
digital media. Game Studies. 1(1).
Spector, W. (2013) Holodeck: Holy grail or hollow promise? Gamesindustry
International. [online] August 31. Available at:
Section I
IDN History
A Concise History of Interactive
Hartmut Koenitz, Gabriele Ferri, Mads Haahr,
Didem Sezen and Tonguç brahim Sezen
The rst part of the book is concerned with the history of Interactive
Digital Narrative (IDN). Its intention is to serve as a concise historical
account of the development of IDN from its beginnings to recent works
by means of representative and inuential examples. The identication of
distinct historical phases is problematic, given the many parallel develop-
ments in the eld, for example of hypertext ction and graphical adven-
ture games. Therefore, we identify trajectories based on form—in the sense
of particular visual and physical manifestations. The three evolutionary
trajectories identied here—text-based, cinematic/performative and ludic/
experimental—represent major facets of IDNs. The trajectories traced here
are not meant to be mutually exclusive the same artifact might easily be
related to several of them.
Text-based examples constitute the rst trajectory, from the very rst
IDN artefact originating in the 1960s, to Interactive Fiction games in the
late 1970s and Hypertext Fiction in the early 1990s, leading to their recent
resurrection in the Versu platform in 2013. The second trajectory adds an
audio-visual dimension that partly remediates aspects of cinema and per-
formance, and examples in this group range from interactive movies over
multi-linear TV shows to experimental art installations. This trajectory
also shows the strong interests of avant-garde artists in the expressive use
of interactive technologies. Finally, the third trajectory encompasses video
games and experimental forms that feature complex narrative design. This
last trajectory traces examples that benetted most from recent advances in
technology—better visual representation, more advanced AI and increased
storage capacity.
The beginnings of IDN can be traced back to the computer program Eliza,
created as an experiment in articial intelligence (AI) in 1966 by Joseph
Weizenbaum. Eliza took the form of a program that emulates a Rogerian
therapist; it responds to a user’s textual input by adopting simple but effective
techniques of parsing and pattern matching. For example, Eliza could reply
12 Hartmut Koenitz et al.,
to sentences like “I’m depressed much of the time” with “I am sorry to hear
you are depressed” (Weizenbaum, 1966). Eliza’s ability to sometimes sustain
surprisingly compelling dialogues marks a signicant milestone for the use of
computers as an expressive narrative medium. Elizas considerable impact at
the time (Murray, 1997, pp. 69–70) was also due to the still largely unchal-
lenged belief in the abilities of AI in 1966, and therefore users interacting with
Eliza were more disposed to accept the premise of a computer program as an
intelligent therapist. With this work, Weizenbaum became the rst successful
author of an IDN experience by nding the right balance between procedur-
ality (the rules behind Eliza’s responses), agency (allowing natural language
input) and scenario/role (therapy session and patient) that played into the
belief system of his contemporaries (AI as capable of intelligent conversations).
Adventure (Crowther, 1976) is the next seminal piece in the IDN tradi-
tion that marks the beginning of the Interactive Fiction (IF) genre. Adventure
allowed players to explore a ctional world set in a large cave that is rendered
to the players in the form of textual descriptions and subject to interaction
through the entry of textual commands such as “go north, “pick up sword”
or ght troll with sword. The basic mechanics of Adventure consisted
of problem solving, combining objects, dialogues and spatial exploration.
Adventure’s considerable success would reach into the commercial realm, as
the American company Infocom famously expanded this framework in the
following years. Their rst product, Zork I (Blank and Lebling, 1980), broke
new technical ground as programmers applied techniques like object orienta-
tion, demons and states to create a dynamic ctional universe (Murray, 1997,
p. 78). IF successfully integrated complex narrative with puzzles and riddles
that not only control the revelation of the narrative (Montfort, 2003a, p. 3)
but also generates narrative through the players’ typing of words.
Michael Joyce’s Afternoon, A Storyrst shown in 1987 and subse-
quently published in 1991—constitutes one of the earliest pieces of Hyper-
text Fiction (HF, sometimes also abbreviated as Hyperction), another
text-based IDN subgenre that was particularly active until the mid-1990s.
Michael Joyce and Jay Bolter, cocreators (with John B. Smith) of the HF
authoring tool Storyspace, clearly position HF as a new form of highbrow
literature in contrast to IF:
Interactive ction has already existed for some time in the form of
computerized adventure games. … Admittedly the text of the current
games is simple-minded, but the method of presentation is not. This
method of presentation can now be applied to serious ction.
(Bolter and Joyce, 1987)
Whilst the other examples discussed so far originated within research labs in
computer science, HF works from the very beginning were created by authors
like Michael Joyce and Douglas Cooper who had already published tradi-
tional books before picking up HF. These creators aimed at overcoming the
Introduction 13
limitations of the printed book by embracing digital media and turning readers
into participants, which Murray terms interactors (Murray, 1997). The interac-
tor of HF as envisioned by Bolter and Joyce is no longer the passive consumer
of a nished work but instead is given an active role in constructing meaning.
HF relies on the principles of segmentation and linking, as authors pro-
duce screen-sized segments, or lexias, and connect them with different types
of hyperlinks. Interactors traverse the story by selecting links, unveiling new
lexias, or returning to the ones already visited. Such repeated visits—called
multivalence by hypertext theorist Mark Bernstein (2000)—constitute a
design strategy specic to HF, where the meaning of particular lexias change
upon revisitation, as the interactor gains additional insights. The success of
this strategy depends on the complexity and depth of the particular narrative.
In Michael Joyce’s Afternoon, A Story (1991), multivalence is particularly
successful, as the interactor slowly gains a better understanding of the unreli-
able narrator’s narrative by traversing more than 500 lexias connected by
over 900 links regarding the life-changing event of witnessing a car accident,
the protagonist’s failure to provide help and his consequent psychosis.
Another design strategy in HF is in the equivalence between content and
structure: for example, a fragmented narrative like Afternoon is presented
in fragmented pieces and the associative connections as links. In Shelly
Jackson’s Patchwork Girl (1995), the protagonist herself is literally patched
together from body parts of deceased women. This narrative strategy sets
the stage for a fragmented narrative, exploring the main character as well as
the lives of the donors.
After years of relative obscurity, Interactive Fiction recently seems poised
for a return to the spotlight in the form of Versu (2013), a project that origi-
nated with Linden Labs, the developer of Second Life. Versu merged text-
based narrative with advanced articial intelligence methods and expressive
graphics. The project aims to create a platform for procedural textual nar-
rative and hopes to attract authors by offering a specic authoring tool and
a business model for distribution. The rst examples—narratives set in a
Jane Austen-inspired Regency era and the Roman Empire, respectively—by
renowned interactive ction writer Emily Short received positive reviews.
Interactive Cinema i s a n u mb rel la term for wor ks and e xperim ents combining
cinematic experiences and interactivity, dating back to the 1967 experiment
Kinoautomat created by Radúz Çinçera for the Czechoslovakian pavilion at
the Montreal World Fair. The movie One Man And His World was stopped
at several points during the presentation, and the audience was asked to
make a decision. Depending on the answer, the projectionist exchanged the
lens cap between two synchronised lm projectors (see Naimark, 1998).
14 Hartmut Koenitz et al.,
The Kinoautomat therefore required a human intermediary to execute the
audience’s choices, and direct interaction between an interactor and a cin-
ematic experience was not possible until the late 1970s when MCA/Phillips,
Pioneer and RCA introduced the laser disc system which allowed random,
direct access to every point in a video via a computer interface. With this
technology, the Architecture Machine Group at MIT created the Aspen
Moviemap (1978), which enabled an interactor to virtually explore the
town of Aspen in Colorado, USA by using a touch screen interface to con-
trol a running video of a drive through the town. The interactor could click
on the facades of houses along the way to access additional material, such as
interior shots, historical images, menus of restaurants and video interviews
with inhabitants. In the following years, many other applications combining
video and interactivity were explored. Of these, A City in Transition: New
Orleans 1983–86 (Davenport, 1987), a multimedia experience providing
access to narrative video and other content, stands out as a particularly
rened piece. Glorianna Davenport, a pioneer in the area of interactive doc-
umentaries, focused her work on a massive urban development effort on a
strip of New Orleans Mississippi river embankment in connection with the
1984 world fair.
The terms Interactive Movie and Interactive TV have also become asso-
ciated with experiments in interactive lms for the cinema and television,
respectively. In 1991, Oliver Hirschbiegel created Mörderische Entscheidung
(Murderous Decision), a crime story broadcast on two TV channels simulta-
neously, each one presenting the same story from the perspective of a differ-
ent character and allowing the audience to interact by zapping between the
channels with an ordinary remote control. Hirschbiegel experimented with
several narrative strategies to adapt his story for interactivity—for example,
cueing interactors to switch channels by reducing the amount of informa-
tion given (Weiberg, 2002), but also making sure that information essen-
tial for understanding the story was given on both channels. An empirical
study about the experiment (Kirchmann, 1994) suggested that the narrative
“worked best when both versions showed the same information from differ-
ent points of view” (Weiberg, 2002), for example when both main characters
were present in the same space and their views were represented similarly.
Conversely, the moments in which the representation diverged (for instance,
when one of the two characters was depicted as intoxicated) proved more
problematic for the audience.
In the following years, the same concept was reelaborated in the Danish
experiment D-Dag (Kragh-Jacobsen, Levring, Vinterberg and von Trier,
2000), showing four different narratives on separate channels plus additional
channels presenting the directors’ commentary, for a total of seven options.
The framing narrative for D-Dag was a bank robbery on NewYear’s Eve of
the new millennium in which the noise from the celebratory reworks was
used to mask the explosion needed to break into the bank.
Interactive video installation pieces combine video segments with
algorithmic rules and a level of interactive control by the audience or a live
Introduction 15
performer. Historically, the majority of artists came to this eld through a
gradual process, often by starting to use computers as control devices for
noninteractive work before exploring the potential of user participation. For
example, the artist Toni Dove started using computers to synchronise slide
shows in her 1990 work Mesmer: Secret of the Human Frame. Her rst inter-
active piece Archeology of a Mother Tongue (1993) is a virtual reality murder
mystery (Dove, n.d.) that combines interactive computer graphics, laserdisc
video, and slides with interactive sound (Dove, n.d.). The interactor controls
the environment by using a small camera to look around a virtual reality envi-
ronment and a data glove to touch virtual objects. As an untrained interactor
might be overwhelmed by the technology involved, Dove often uses a trained
tutor to interact with her pieces (Bonin, 2001). What Dove explores in her art
pieces is the sensation of walking around in a movie, of actually being inside of
a narrative space (Jennings, 1995) and also the powerful experience of a physi-
cal action [that] produces a response in video and audio (Jennings, 1995).
A particularly interesting piece is Wheel of Life (1993), jointly directed
by Glorianna Davenport and Larry Friedlander. The large-scale installation
was created around the idea of representational spaces for the different ele-
ments of water, earth, re and space as symbols for both the circle of life
and the evolution of life on Earth and beyond. Each space contained video
screens and projectors, a sound system, light installations and interactive
objects. What sets Wheel of Life apart from other examples is its three-way
interaction: the piece augments the usual interaction between a computer
program and a human interactor by including a second interactor. Indi-
vidual spaces were designed for interaction between a guide controlling the
space on a computer display from the outside and an explorer experiencing
the space from within:
Together they had to discover how to navigate through a world that
responded mysteriously to their actions; the explorer’s task was to
decipher the rules and narratives governing each area, while the guide
sought to help the explorer by using the computer to manipulate the
images, lights and sounds in the area.
(Davenport and Friedlander 1995)
Toni Dove’s latest work, Lucid Possessions (2013), is a contemporary ghost
story centred on a programmer whose advanced skills make her a target for
being possessed by ghosts. Beyond this narrative frame, the piece explores
the state of identity in a dichotomous relationship between the virtual and
the real on the backdrop of our ever-increasing presence in the virtual world
of social networks. In technical terms, the work combines actors and musi-
cians, robots, custom computer hardware and real-time motion-tracking
technology in an on-stage performance. The artist becomes an interactor in
the dynamic presentation by means of a gesture-based interface, controlling
the presentation of video clips. Lucid Possessions is a compelling example
of the expressive potential of IDN in the hands of an artist.
16 Hartmut Koenitz et al.,
We conclude this overview of the historical evolution of IDN by noting sev-
eral important examples of narrative video games and experimental forms.
Following in the footsteps of the early text-based narrative games, graphic
adventure games helped to diffuse and popularise IDN, pioneered by the
King’s Quest series, the rst of which appeared in 1984.
The Monkey Island series of games (1990–2010) is a paradigmatic exam-
ple in its combination of narrative and game elements. The game places the
interactor in the role of a hapless pirate who has to overcome many difcult
challenges to prove himself to the pirate establishment while winning the
heart of his love interest, the governess of a pirate municipality. The series
featured a rich narrative and became famous for its humour and irony. The
ow of the narrative is intertwined with puzzles requiring funny, improb-
able solutions, such as freeing a prisoner by melting the iron bars of his cell
with a very potent drink. Advancement in the narrative becomes the reward
for puzzle solving, often in the form of noninteractive cut scenes that follow
major accomplishments. The Monkey Island series is exemplary for keeping
a balance between puzzle-solving and narrative development—establishing
a consistent style and setting many canonical conventions still in use today.
Myst (1993) continues the adventure game tradition in terms of spatial
exploration and puzzle-solving, but it also introduces a highly atmospheric
visual representation seen from a rst person perspective. The interactor
nds her/himself on an island that contains abandoned buildings and mys-
terious machinery and is left to explore a highly detailed and evocative
world. The game was not only an aesthetic milestone but also a convincing
example of embedded narrative (Jenkins, 2004) through narrative-infused
encounters. More recently, Gone Home (2013) applies the same underlying
strategy. The game places the interactor in the role of a student returning
from an exchange year abroad, only to nd the family home empty and the
parents and sibling gone. By exploring the house and its contents—the fur-
niture, notes from the inhabitants, audio cassettes and other personal items,
the interactor patches together the narrative of her sister’s disappearance
and the parent’s attempt to resurrect their marriage.
Although not a commercial success, The Last Express (Mechner, 1997) is
remarkably innovative for its integration of narrative, game and exploration.
This game casts the interactor in the role of a passenger investigating a
murder aboard the Orient Express from Paris to Constantinople on the eve
of World War I. Space and time play a central role in this piece, as many
events take place simultaneously and provide a variety of narrative paths.
By moving through the train, the interactor assembles a particular narrative
composed by the conversations he/she overhears and the events he/she wit-
nesses. The game runs in 6x accelerated real time and presents the player’s
current location on a map. This use of temporality and location helps to
Introduction 17
enhance the player’s sense of immersion and precludes exhausting the limited
amount of possible narratives easily, as the interactor can only be in one loca-
tion at any given time. The train’s stops in stations provide a natural means
to structure the narrative into chapters, which make the amount of possible
combinations more manageable by folding back to a shared backstory.
The second half of the 1990s saw an important technical development
in the advent of 3D representations in video games. The 3D game engines
gave interactors the ability to roam free in the designed spaces, but they
also removed a measure of control from the creators, which has important
implications for narrative design.
The inuential game Blade Runner (1997) uses a 3D depiction and is
set in the same world of the movie of the same name by Ridley Scott. The
interactor is cast in the role of a police ofcer whose job it is to nd and kill
replicants, illegal synthetic humans that are so much like their natural coun-
terparts that they are almost impossible to distinguish. The game confronts
the interactor with strong moral choices that affect the outcome of the nar-
rative. For example, the protagonist can decide to go over to the outlaw
side and ght alongside with the replicants; try to restore his reputation by
hunting down synthetic humans; or simply leave the city and the ghting
behind. These decisions eventually lead to thirteen different endings, varia-
tions of the three main outcomes based on the interactor’s earlier choices.
Blade Runner eclipses many other narrative games in the variety of narra-
tive paths that lead to alternative endings, in contrast to the singular suc-
cessful completion of Monkey Island. A key narrative element that enables
enhanced variety and deep engagement with the character is the ability to
switch sides, as the police ofcer may turn into an outlaw and gain an obvi-
ously different perspective. In this way, the game invites the interactor to
explore moral ambiguities in the Blade Runner narrative, for example the
corruption of police ofcers and the ethics behind killing replicants.
Fahrenheit (2005), also known as Indigo Prophecy in North America,
contains the narrative of ritualistic murders in New York City in an imag-
ined year 2009 and combines a 3D real-time rendered gameworld with cin-
ematic elements in the form of screen montage and transitions. In addition
to a well-formed multilinear narrative, Fahrenheit is especially relevant for
the unusually high degree of narrative control given to the interactor over
three different characters in a single game session. This interesting mechanic
results in novel narrative experiences, for example when two user-controlled
characters work against the third who is also managed by the interactor.
The critically acclaimed The Walking Dead (2012) is an adventure
game in the setting of the TV series of the same name, which depicts a
post- apocalyptic world after a zombie outbreak has befallen the United
States. The narrative design requires the interactor to make difcult, mor-
ally ambiguous choices, such as which of the other characters to save. These
decisions are coupled with a feedback system that succeeds in making the
choices meaningful and memorable for the interactor. In tandem with a rich
18 Hartmut Koenitz et al.,
narrative world, the game creates a compelling IDN experience, which is
indicative of the development in the adventure game genre. The focus on
narrative feedback is a productive direction for future work.
The Last of Us (2013) pairs the player in the role of a middle-aged man
with a 14-year-old girl on a journey through a dark, postapocalyptic world
that is full of deadly enemies. The narrative design is carefully crafted so
that the interactor builds a connection with the teenage sidekick through
many dialogues and a slow change in behaviour, signalling growing trust.
This emotional connection helps give the narrative depth and creates an
immersive experience to the point that some players felt that the ghting
scenes represented a distraction. The Last of Us is an outstanding example
of a satisfying narrative experience that not only enhances the shooter genre
but also works as a stand-alone design.
Finally, we discuss IDN experiments in the form of interactive drama and
hybrid forms. This combination of drama and interactivity has been intro-
duced to the digital realm in the late 1980s by the OZ project at Carnegie
Mellon University. Inuenced by Brenda Laurels neo-Aristotelian approach
to interactive narratives (Laurel, 1986, 1991), research in the OZ group
focused on related Articial Intelligence (AI) techniques and their concrete
implementation. Interactive drama was conceptualised as a combination of
presentation, virtual characters and a drama manager component to preserve
coherence and advance the narrative. Based on this conceptual framework,
the OZ group produced two implementations: Lyotard (Bates, 1992), a text-
based experiment that simulated a house cat, and Edge of Intention (Loyall
and Bates, 1993), a graphical experiment that contained an animated ava-
tar (called Woggle) plus other autonomous agents in different roles. Michael
Mateas later continued the research on interactive drama with his collabora-
tor Andrew Stern by working on Façade (Mateas and Stern, 2003, 2005a,
2005b). This interactive experience cast the interactor in the uncomfortable
position of witnessing a couple on the verge of a break-up, in a “dramati-
cally interesting, real-time 3D virtual world inhabited by computer-controlled
characters. (Mateas and Stern 2003) The narrative development in the work
is dynamic and leads to various consequences. Stuck between two arguing
partners, the interactor of Façade faces an implicit choice between helping
them to face their issues and stay together or siding with either of the char-
acters and potentially leading to a break-up. Façade adopts several strategies
to engage the interactor in the unfolding narrative: a nite space (the couple’s
small apartment), the uncomfortable but familiar situation of a ghting
couple, the continuous real-time ow of events and the audible answers of the
virtual characters work together to create immersion. Façade exemplies the
kaleidoscopic nature of interactive narratives (Murray, 1997), as the system
can produce a wide variety of different narrative paths leading to more or less
satisfying nales; in the best version, interactors may experience a powerful
sense of agency by saving the couples marriage. Unfortunately, this ending
is hard to reach for a variety of reasons, some due to technical limitations
but also because of particular design decisions, especially the advancement
Introduction 19
in real-time, which affords the interactor very little time to make decisions.
Nevertheless, Façade remains the most complete interactive drama in the tra-
dition of the OZ project—Mateas and Stern’s piece has received considerable
acclaim and was praised as a critical breakthrough. Mateas has, since then,
continued his work in Prom Week (2012), an IDN piece that recreates the
social situations in a high school class in the week leading up to the prom.
As interactive media matures, some authors have consciously begun to
cross over conventions, applying foreign design strategies to produce ambig-
uous artefacts. Adam Cadre’s IF Photopia (1998) is a work that presents
interleaving narrative strands of the events leading up to a car accident, the
exploration of an alien planet and a surreal world in which the interactor can
y. However, this IF presents spatial and textual exploration and emerges as
a segmented narrative with hyperlinks substituting standard IF commands
and creates an interesting hybrid between IF and HF. Natalie Bookchin’s
The Intruder (1999) turns a short story by Jorge Luis Borges into an interac-
tive experience by requiring the interactor to play several rudimentary video
games in the style of classic titles like Pong or Space Invaders. Both Cadre’s
and Bookchins pieces offer a glimpse of possible future design directions that
transgress traditional boundaries within Interactive Digital Narrative and are
a testament to the vitality of this eld for future experiments. A case in point
is Device 6 (2013), a work that continues in the same vein of genre crossover
by mixing textual presentation reminiscent of HF with animated audio-visual
elements. In this interactive thriller, the screen text is arranged spatially and
serves as both narrative manifestation and map. Save the Date (2013) pushes
the boundaries of narrative on a different level. The premise of the game ini-
tially seems straightforward, as the interactor is given a range of options on
how to arrange a dinner date; however, the task quickly becomes problematic
because the love interest invariably gets killed. In addition, the game changes
the epistemological dimension by representing the interactor’s memory in
successive sessions. Yet, even with the added knowledge of previous failed
attempts, a satisfying narrative ending remains elusive and the interactor ulti-
mately faces a choice between two unsatisfying alternatives—either to give up
on the date or to end the game before disaster strikes. In that sense, Save the
Date presents a considerable challenge to our established sense of narrative.
Interactive Digital Narrative, in its manifold forms, has come a long way
since Eliza and Adventure. Works like The Walking Dead, Gone Home, but
also Ludic Processions and Device 6, are testimony of an established eld.
And yet, there are no signs of stagnation on this creative frontier for narra-
tive expressions. Great things are yet to come.
1. See for example
20 Hartmut Koenitz et al.,
Adams, E. (2014) Fundamentals of Adventure Game Design. San Francisco: Peachpit
Bates, J. (1992) The Nature of Characters in Interactive Worlds and the Oz Project.
Technical Report CMU-CS-92-200, Department of Computer Science. Pittsburgh,
PA: Carnegie Mellon University.
Bates, J., Loyall, A. B. and Reilly, W. S. (1992) Integrating reactivity, goals, and emo-
tion in a broad agent. In Proceedings of the Fourteenth Annual Conference of the
Cognitive Science Society. New York: Lawrence Erlbaum.
Baudrillard, J. (1983) Simulations. New York: Semiotext(e).
Bernstein, M. (1992) Contours of constructive hypertexts. In Proceedings of the
1992 Hypertext Conference. New York: ACM.
Blank, M. and Lebling, D. (1980) Zork I. [Interactive ction] Cambridge, MA:
Bolter, J. D. and Joyce, M. (1987) Hypertext and creative writing. In Proceedings of
the 1987 Hypertext Conference. New York: ACM.
Bonin, V. (2001) Toni Dove. [Online] Available at: http://www.fondation-langlois.
Bookchin, N. (1999) The Intruder. [Interactive narrative] Available at: http://bookchin.
Cyan, (1993) Myst. [video game] Brøderbund.
Coover, R, (1992) The end of books. New York Times Book Review. June 21. 1.
Crowther, W. (1976) Adventure. [Interactive ction].
Davenport, G. and Friedlander, L. (1995) Interactive transformational environ-
ments: wheel of life. In: Barrett, E. and Redmond, M. (eds.) Contextual Media:
Multimedia and Interpretation. Cambridge, MA: MIT Press.
Davenport, G. (1987) New Orleans in transition, 1983–1986: The interactive deliv-
ery of a cinematic case study. In Proceedings of the International Congress for
Design Planning and Theory, Education Group Conference. Boston, MA.
Dove, T. (1990) Mesmer: Secret of the Human Frame. [interactive installation].
Dove, T. (1993) Archaeology of a Mother Tongue. [Interactive installation].
Dove, T. (2013) Lucid Posessions. [Interactive installation].
Dove, T. (n. d.). [Project description of Archeology of a mother tongue”]. Retrieved
March 12, 2009 from
Evans, R. and Short, E. (2013) Versu. [Interactive Narrative Platform].
Fauth, J. (1995) Poles in your face: The promises and pitfalls of hyperction.
Mississippi Review. 6.
Fullbright Company (2013) Gone Home. [Video game] Fullbright Company.
Jackson, S. (1995) Patchwork Girl. [Interactive ction] Watertown: Eastgate Systems.
Jenkins, H. (2004) Game design as narrative architecture. In: Harrigan, P., Wardrip-
Fruin, N. (eds.) First Person: New Media as Story, Performance, and Game.
Cambridge, MA: MIT Press.
Jennings, P. (1995) Interpretation on the electronic landscape: a conversation with
Toni Dove. Felix: a Journal of Media Arts and Communication. 2(10).
Joyce, M. (1991) Afternoon, a Story. [Interactive ction] Watertown: Eastgate
Introduction 21
Kirchmann, Kay (1994) Umschalten erwünscht? Wenn ja, von wem? Ergebnisse
einer Studie zu Ästhetik und Rezeption des ersten interaktiven. TV-Spiels des
Deutschen Fernsehens Arbeitshefte Bildschirmmedien. 48. pp. 23–60.
Kragh-Jacobsen, S., Levring, K., Vinterberg, T., von Trier, L. and Ehrhardt, B. (2000).
D-Dag. [Interactive movie].
Laurel, B. (1986) Toward the Design of a Computer-Based Interactive Fantasy
System. Doctoral Thesis. Drama Department, Ohio State University.
Laurel, B. (1991) Computers as Theatre. New York: Addison Wesley.
Loyall, A. B., and Bates, J. (1993) Real-time control of animated broad agents. In
Proceedings of the Fifteenth Annual Conference of the Cognitive Science Society.
Boulder CO: Lawrence Erlbaum.
Mateas, M., and Stern, A. (2003) Integrating plot, character and natural language
processing in the interactive drama Façade. In Proceedings of the 1st Interna-
tional Conference on Technologies for Interactive Digital Storytelling and Enter-
tainment (TIDSE-03).
Mateas, M. and Stern, A. (2005a) Structuring content in the façade interactive drama
architecture. In: Young, R. M. and Laird, J. (eds.) Proceedings of the First Arti-
cial Intelligence and Interactive Digital Entertainment Conference (AIIDE 2005).
Menlo Park, CA: AAAI Press.
Mateas, M. and Stern, A. (2005b) Façade. [Software] Available at: http://www.
Mateas, M. (1997) An Oz-Centric Review of Interactive Drama and Believable
Agents. Technical Report CMU-CS-97-156, Department of Computer Science.
Pittsburgh, PA: Carnegie Mellon University.
Mechner, J. (1997) The Last Express. [Video game] Brøderbund.
Montfort, N. (2003a) Twisty Little Passages: An Approach to Interactive Fiction.
Cambridge, MA: MIT Press.
Montfort, N. (2003b) Toward a Theory of Interactive Fiction. Available at: http://
Moulthrop, S. (1992) Victory Garden. [Interactive ction] Watertown: Eastgate
Murray, J. (1997) Hamlet on the Holodeck: The Future of Narrative in Cyberspace.
New York: The Free Press.
Naimark, M. (1998) Interactive art—Maybe it’s a bad idea. In: Leopoldseder, H.,
and Schšpf, C. (eds.) Cyber Arts 1998 International Compendium Prix. Ars Elec-
tronica. Berlin, Heidelberg: Springer.
Naughty Dog (2013) The Last of Us. [Video game] Sony Computer Entertainment.
Paper Dino Software (2013) Save the Date. [Video game] Paper Dino Software.
Quantic Dream (2005) Farenheit. [Video Game] Atari.
Rand, A. (1971) Night of January 16th. New York: New American Library.
Simogo (2013) Device 6. [Video game] Simogo.
Telltale Games (2012) The Walking Dead. [Video game] Telltale Games.
Weiberg, B. (2002) Beyond Interactive Cinema. Available at:
Weizenbaum, J. (1966) ELIZA—A computer program for the study of natural lan-
guage communication between man and machine. Communications of the ACM.
9(1). pp. 36–45.
Westwood Studios (1997) Blade Runner. [Video game] London: Virgin Interactive.
2 The American Hypertext Novel, and
Whatever Became of It?
Scott Rettberg
The 1990 Eastgate publication of Michael Joyce’s Afternoon, A Story
earned hypertext ction a place within institutionalised literary culture.
Robert Coover’s 1992 essay “The End of Books” announced hypertext c-
tion as a challenge to traditional conceptions such as narrative linearity,
the sense of closure, and the desire for coherence. While some theorists,
such as George Landow, praised hypertext for instantiating poststructural-
ist theory, others, such as Sven Birkerts in The Gutenberg Elegies (1994),
regarded it with strong concern. The publication of more hypertext ctions
such as Stuart Moulthrop’s Victory Garden (1991) and Shelley Jackson’s
Patchwork Girl (1995) resulted in a small, dedicated interest community.
However, no paradigm-shifting rise in interest took place. The indepen-
dent publication of hypertext novels on the World Wide Web such as Rob-
ert Arellano’s Sunshine ’69 (1996), Mark Amerikas Grammatron (1997),
and William Gillespie, Frank Marquardt, Scott Rettberg and Dirk Strat-
ton’s The Unknown (1998) briey revitalised the networked ctional form
before it was eclipsed in the rst decade of the 21st century by a range of
other digital narrative forms.
During the 1990s, as the personal computer became a xture of everyday
life in ofces and homes, writers and academics became interested in explor-
ing the potential of hypertext as a genre for narrative ction. During the
early 1990s in particular, the form seemed to offer great promise as a logical
literary follow-on both to late 20
-century poststructuralist theory and to
general shifts in late 20
-century ction writing, especially in the United
States, towards postmodern narrative structures that contested prior con-
ventions of linearity, closure and immersive reading.
For what was in retrospect a quite brief period during the late 1980s and
early 1990s, what Robert Coover (1999) referred to as the Golden Age of
literary hypertext, writers working in dedicated hypertext systems such as
Hypercard, the Intermedia system at Brown University and, most impor-
tantly, Storyspace, explored the potential of hypertext for narrative ction.
The American Hypertext Novel, and Whatever Became of It? 23
Enough writers were working in the form during this period, and their works
garnered enough critical attention that we can speak of these hypertext c-
tions as a group—what some of have called ‘The Eastgate School’—named
after the publisher responsible for publishing and distributing most of the
early works of hypertext ction.
Hypertext was rst conceptualised by Theodore Holm Nelson (1965)
in his paper A File Structure for the Complex, the Changing, and the
Indeterminate” when he introduced the term “to mean a body of written
or pictorial material interconnected in such a complex way that it could
not conveniently be presented or represented on paper. In his “No More
Teachers’ Dirty Looks” (1970), Nelson followed up with a more expan-
sive denition of hyper-media as “branching or performing presentations
which respond to user actions, systems of prearranged words and pictures
(for example) which may be explored freely or queried in stylized ways.
Among the types of hypertexts he discusses in that essay, which focused
on the potential uses of hypermedia in new systems that could potentially
revolutionise education, are discrete hypertexts which “consist of separate
pieces of text connected by links. The majority of hypertext ctions pub-
lished during the 1980s and 1990s would t within this rubric though, as
Noah Wardrip-Fruin (2004) argued, this conception of hypertext as chunk-
style linked nodes is somewhat narrower than hypermedia as Nelson origi-
nally envisioned it. The postmillennial turn in electronic literature towards
a broader use of media-rich texts, more complex uses of generativity and
other computational processes, and deeper engagement with network-
specic communication technologies and styles of writing better represents
the broader range of hypertext and hypermedia as conceptualised by Nelson
than either the rst-generation hypertext ctions published by Eastgate or
the second-generation works published on the Web.
During the 1980s, digital writing experiments, including long-form
hypertext ctions such as Judy Malloy’s pioneering hyperctional narrative
database Uncle Roger (1986),
were distributed on bulletin boards such as
the WELL.
Interactive ction developed separately from hypertext ction,
emerging from a different cultural context. The text-parser-based form of
interactive digital writing that began with Colossal Cave Adventure by Will
Crowther and Don Woods (1976) saw commercial success during the 1980s
with Infocom titles such as the Zork series before graphic computer games
swallowed the market in the late 1980s. As Nick Montfort documents in
his Twisty Little Passages (2003), an amateur community developed around
the form shortly thereafter, enabled by group effort and the release of free
programming languages, authoring and reading environments such as TADS
(in 1987) and Inform (in 1993). Though there were substantial literary and
aesthetic achievements in interactive ction, outside of a few outliers,
it was
not until the publisher and software developer Eastgate Systems began pub-
lishing hypertext ction in 1987 that literary critics began to signicantly
engage with electronic literature.
24 Scott Rettberg
The novelist Robert Coover played an important role in bringing these
experimental ctions to the notice of a broader public, most notably in his
1992 New York Times Book Review essay “The End of Books, which intro-
duced hypertext ction to a broader literary audience. The essay describes
the ways that the hypertext form poses challenges for writers, and readers
accustomed to conventional narrative forms, including assumptions about
linearity, closure and the division of agency between the writer and reader.
Coover focused on the disruptions that hypertext caused to narrative linear-
ity, its interruption of “the tyranny of the line” (1992). Coover noted that in
hypertext, the narrative structures of ctions are foregrounded and become
a predominant concern: “The most radical new element that comes to the
fore in hypertext is the system of multidirectional and often labyrinthine
linkages we are invited or obliged to create.
Coover identied a core tension for ction writers working in hypertex-
tual forms in the “conict between the reader’s desire for coherence and
closure and the text’s desire for continuance, its fear of death. He asked: “If
the author is free to take a story anywhere at any time and in as many direc-
tions as she or he wishes, does that not become the obligation to do so?”
Coover’s essay on hypertext remains a key summation of the affordances
and opportunity costs of hypertext for ction writers. Forms such as the
short story and particularly the novel have a deep, if not intrinsic, relation-
ship to the limits and constraints of print. Whatever belief one has of the
function of ction, most stories or novels strive to reduce, distil and make
comprehensible an experience or perception of the world. The role of the
ction writer is both to provide a particular perspective on human experi-
ence and to select and enliven some narrative possibilities while eliminating
others. The form of the book—such that each page requires more paper
and each word more ink—pulls the writer towards economy, restraint, and
limitation while the form of hypertext, offering multiple pathways, limitless
text and multiple medial modalities, pulls authors towards exploration and
The potential uses of hypertext links in a narrative ction are multifari-
ous, ranging from offering the reader conscious plot choices in a choose-
your-own-adventure style, to establishing multiple lines of narrative or
multiple character perspectives on the same set of events, to serving as any
kind of footnote-style reference, to poetic or linguistic play between words
or scenes in a narrative.
In Michael Joyce’s Afternoon, A Story (1989) we saw the hypertext link
used in all of these various ways in the story of a man in a state of break-
down after a car accident in which his son may or may not have been killed.
The fragmented and disrupted nature of the narrator’s psychological and
emotional state nds an objective correlative in the hypertextual form of the
narrative. The narrative of Afternoon centres on the various overlapping
relations and conicts between the main narrator Peter, his ex-wife Lisa, her
lover Wert (Werther), Lolly and Nausicaa.
The American Hypertext Novel, and Whatever Became of It? 25
There is no need to provide a close reading of Afternoon here—as it is the
most cited work of electronic literature,
and ample discussions of the work
are available elsewhere. However, we can highlight a few formal aspects of
the work. The rst is that the work is playing a great deal with indetermi-
nacy and a sense of loss and confusion. The reader can follow a default path,
but there are also typically several links on a given node of the story. Joyce,
however, chose not to make those links visible to the reader. Instead, he
encouraged readers to seek out ‘words that yield’—that is, to click on given
words that seem to be particularly evocative, and to see if the system will
respond. This indeterminacy extended to the narrative voice. Though the
main perspective is that of Peter, selecting some nodes will pull the narrative
into the perspective of one of the other characters. These switches are not
always clearly marked in a way that would make the perspective immedi-
ately clear to the reader. Another notable aspect of the work is its very clear
use of modernist techniques and tropes. The work is thick with intertextual
references: to the Odyssey, to James Joyce’s Ulysses, to Goethe’s Werther,
to the Grimm brothers, to Tolstoy and to Pynchon. Formally, the text is
also diverse. Following links takes the reader from interior monologues and
musings to dialogues to poems and lists. The fact that the text as a whole
echoes the breakdown state of the main character is yet another modernist
gesture. We encounter Peter’s interior state through the form of the work in
a similar way to encountering the stream-of-consciousness narration in, for
example, Virginia Woolfs To the Lighthouse.
Michael Joyce’s Afternoon is, in short, a ‘writerly’ work of modernist
ction developed in Storyspace software. While the material form of the
work is digital, its DNA is clearly a strand of the experimental writing tradi-
tions of the 20th century. We can see very clear relationships between the
work and antecedent print works. While it is not accidental that the work
is presented in hypertext form, Joyce’s gravitation towards experimenting
with hypertext was driven by the fact that the fragmented, nodal style of
Storyspace software, the tentative, indeterminate, searching way that the
reader encountered the interface, and the various models of interconnectiv-
ity available in Storyspace were particularly well-suited to the form of the
story he wanted to tell.
Joyce’s Afternoon is an exclusively text-based hypertext—it included
no images or other multimedia assets. Compared to many contemporary
works, it also had very limited use of computation, including links, some
limited ‘yes/no’ text parsing, and guard elds that limited access to some
nodes until other nodes had been visited. Writing on the rst hypertext
experimentalists in his Literary Hypertext: The Passing of the Golden Age,
Coover (1999) notes that:
[E]arly experimental writers of the time worked almost exclusively in
text, as did the students in our pioneer hypertext workshops at Brown
University, partly by choice (they were print writers moving tentatively
26 Scott Rettberg
into this radically new domain and carrying into it what they knew
best), but largely because the very limited capacities of computers and
diskettes in those days dictated it.
Coover noted that these constraints were also empowering. The writers
working with hypertext in the 1980s and 1990s were not primarily focused
on manipulating images or animations or complex programming tasks, but
were instead mostly working with words, lines, texts and scenes, just as
any other writer of the time would have been. They were working within a
computational environment that offered new ways of remedying narrative
techniques largely derived from the canon of modernist and postmodernist
ction that preceded them. Though the majority of the hypertext systems
available at the time did allow for some use of visual media and other media
assets, text was clearly the dominant mode of expression.
The majority of the hypertext ctions published by Eastgate during
the late 1980s and early 1990s similarly emerged from a clearly liter-
ary heritage. The two other most-frequently cited works of this period,
Patchwork Girl by Shelley Jackson (1995) and Victory Garden by Stuart
Moulthrop (1991), were each novel-length works of deeply intertextual
Patchwork Girl, Jackson’s inventive retelling of the Frankenstein story
from the perspective of the female monster, wore its postmodernism on
its sleeve. In the tradition of American literary postmodernism, Patchwork
Girl is a self-conscious text, by turns intertextual, polyvocal and expressly
concerned with poststructuralist questions of identity. Five sections of the
hypertext novel—the journal, story, graveyard, crazy quilt, and the body of
the text—each used different material structures and stylistic conventions.
Amore visual thinker than Joyce, Jackson both used some limited imagery—
woodcut images of a female body and a man’s skull—and took advantage
of the visual layout features of Storyspace software to produce visual user
interfaces. The reader of Patchwork Girl participated in navigating the work
in acts of pastiche. Assembly of different types of texts occurs through-
out the work. In the ‘crazy quilt’ section of the text, for example, Jackson
stitches together quotations from Jacques Derrida’s Disseminations, Donna
Haraway’s A Cyborg Manifesto, Mary Shelley’s Frankenstein, L. Frank
Baum’s The Patchwork Girl of Oz, Barbara Maria Stafford’s Body Criticism
and the Storyspace User’s Manual. The rst draft of Patchwork Girl was
originally produced for one of theorist George Landow’s courses at Brown
University, and throughout the novel, we can see how Jackson was testing
the waters, using both ction and a new writing technology she was encoun-
tering for the rst time to represent and deconstruct theories of identity.
Both Afternoon and Patchwork Girl were extensive hypertexts, includ-
ing a good deal of writing—according to Raine Koskimaa, 539 lexia (text
spaces) were in Afternoon and 323 were in Patchwork Girl. Though After-
noon includes A Story in its somewhat humble title, both are clearly of
The American Hypertext Novel, and Whatever Became of It? 27
such length and narrative development that they are properly considered
novels. Stuart Moulthrop’s 1991 Victory Garden included 933 lexia. The
most Borgesian of the hypertexts published by Eastgate Systems, Victory
Garden extensively experimented with the multilinear possibility space of
hypertext ction. Taking its cue from Borges’ story “Garden of Forking
Paths,Victory Garden uses hypertext to explore narrative structures: mul-
tiple potential outcomes of given storylines, changes in perspective on given
events, chronological jumps and ashbacks, and different forms of read-
ing cycles (for instance, narrative loops that vary on recursion). Set during
the rst Gulf War and centred on a group of characters living in a univer-
sity town, the novel in a broad sense is about the reception of war ltered
through contemporary academic, popular and media culture. In one lexia
of the work,All of the Above, Moulthrop summarised Borges’ story: “In
all ctional works, each time a man is confronted with several alternatives,
he chooses one and eliminates the others; in the ction of Ts’ui Pen, he
chooses—simultaneously—all of them. Victory Garden is perhaps the sin-
gle experiment in the hypertext canon that attempted to do just that. As
Koskimaa notes, “ Victory Garden is clearly pointing towards the kind of
hypertext ction which, because of its size, is theoretically and practically,
inexhaustible. Just as an event such as a blitzkrieg war unfolding both in
the lived experiences of soldiers and civilians abroad and on multitudes of
cable television screens and other networked media outlets is in some sense
fundamentally unknowable as a totality, Moulthrop was providing us with
a textual analogue in his ction. As in the other two examples, Moulthrop’s
novel included a great variety of different types of textual materials and
was intertextual in relation to several works of print literature—Borges’s
ction and Thomas Pynchon’s novels serve as touchstones and make cameo
appearances. Moulthrop also integrated images and maps as navigational
apparatuses to aid movement through the text.
In discussing the early hypertext ctions, I have cited the three most
referenced works, all of which were produced in the Storyspace platform.
It is important to emphasise that during this time Eastgate also published
works in other platforms, such as John McDaid’s Uncle Buddy’s Phantom
Funhouse (1992), authored in Hypercard, and later M.D. Coverley’s Calia
(2002), produced in Toolbook. The degree to which platform is important in
determining categories or genres of electronic literature is open for debate.
It is clear that particular constraints and affordances of operating systems
have aesthetic effects. While the stories and styles of Afternoon, Patchwork
Girl and Victory Garden were radically different in content and in many
qualities of style, the writing and reading platform of Storyspace set com-
mon material limits and possibility spaces for the authors using it and the
readers encountering the works in the platform.
The other important aspect to consider is that, as a publisher, Eastgate
was participating in an evolving literary culture that was trying to estab-
lish itself. Eastgate advertised itself as the publisher of serious hypertext
28 Scott Rettberg
ction—perhaps in an attempt to differentiate itself from ‘text adventure’
publishing enterprises or the emerging market of games. The social and criti-
cal apparatus is important. Publishing both software and literary works,
Eastgate was partially modelled after the type of serious small press pub-
lishers that popularised modernist literature during the 20th century, and
partly modelled on contemporary software companies. Theorists such as
George Landow, Terry Harpold, Jane Yellowlees Douglas and others were
foregrounding the connections between hypertext narratives and postmod-
ern theory, just as later theorists such as N. Katherine Hayles emphasised the
materiality of these text-machines in their media-specic context and their
relation to the idea of the posthuman.
For a brief period, there was a concerted if in retrospect somewhat des-
perate attempt to carve a niche for hypertext ction within the boundaries
established by the print literary context and traditions—and for a while this
made sense. At the time, most software was sold and distributed as distinct
packaged objects, which could be sold by mail order or in stores. Thus, it
was sensible for a publisher of literary works for the computer to think
along the lines of a traditional publishing model, in which programs could
be treated like individual books, and the author/publisher relationship situ-
ated in a similar fashion to that of print literature. There were some real
benets to this sort of apparatus. The early hypertext movement had a sort
of spiritual home at Brown University and its digital writing workshops,
a house publisher in Eastgate, and a school of critics gathering around its
creations. Though their works were diverse in terms of content, the partici-
pating authors had similar literary heritage and stylistic concerns and were
similarly steeped in modernist and postmodernist 20
-century American
print literature. The digital writing workshops at Brown emerged from a
long tradition of embracing literary experimentation, and Robert Coover
himself is of course one of the pivotal gures in postmodern ction and an
avid experimenter with narrative form who had long played with expan-
sions and perversions of the storytelling apparatus.
By the mid to late 1990s, new models of publishing, communicating and
interacting in network culture were already beginning to evolve, and this
small press model of selling and distributing hypertext ction would become
obsolete before it was ever widely adopted. The World Wide Web came along
and brought with it signicant changes. The market benets of selling com-
puter programs on oppy discs or CD-ROMs were soon trumped by the
practical benets of distributing work instantaneously to the global network.
Sadly, though most of the Eastgate hypertext ctions are still “in print, in
the sense that they can be bought on physical CD-ROM, few of them can
be read on contemporary operating systems. The technical challenges of
updating individual works to make them operable with new versions of the
Mac OS, for example, have proven to be too much for Eastgate to handle.
In other cases, such as that of Uncle Buddy’s Phantom Funhouse, the whole
platform (Apple’s Hypercard) is no longer supported. After teaching the
The American Hypertext Novel, and Whatever Became of It? 29
Eastgate hypertexts for more than a decade, I can no longer use them in
the curriculum of my electronic literature classes, except as a past reference.
They no longer work on any of the computers I own or those available
in our university labs. Short of setting up a dedicated lab with old com-
puter equipment (as Dene Grigar at Washington State University and Lori
Emerson at University of Colorado Boulder have done), the classic hyper-
texts are now difcult to access and operate. While critics and archivists
have provided valuable references that to an extent document and describe
these works, it is not sufcient to read about works of e-lit. Students need to
be able to access the primary texts themselves.
The short history of electronic literature has proven that works produced
and distributed in nonproprietary, open-access and/or open source plat-
forms, and using open standards, have tended to have a longer shelf life
than works produced using proprietary software. If a software developer is
purchased by another company or goes out of business, the platform itself
can cease to be operable. If a publisher holds on to copyright but ceases to
release updated versions of the work to operate with contemporary oper-
ating systems, then even the authors themselves cannot release their work
in an operable format. One hopes that the list of hypertexts published by
Eastgate Systems will soon be made available again in a more accessible and
durable way, so that their important list of early hypertext ctions will not
be lost to contemporary and future readers.
After Tim Berners-Lee’s development of the World Wide Web in 1989–91,
the release of the Mosaic Web browser in 1992 and its commercial cousin
Netscape in 1994, the World Wide Web was by the mid-1990s emerging as
a new playground, distribution platform, and online community for hyper-
text authors. Though the proprietary Storyspace platform actually offered
more hypertext authoring features than did simple HTML, the extensibility
of the Web (allowing for other standards and platforms to plug in to its
architecture) and its capacity as an open global distribution network had
an irresistible pull for hypertext authors eager to reach an audience beyond
the community that had developed around Storyspace and other propri-
etary platforms. During the late 1990s a number of substantial novel-length
works of hypertext ction were published on the Web, where they could
encounter new audiences and for a time, some notoriety, if never main-
stream commercial success.
Judy Malloy moved her Uncle Roger and other ‘narrabase’ writing proj-
ects to the Web in 1995. A wave of hypertext ctions written for the Web
followed for the next ve years; they were hypertexts written and distributed
on the Web in a network-specic context. One could consider this period,
however brief it was, as a second wave or renaissance of hypertext ction.
30 Scott Rettberg
At the time of its initial release in 1996, Robert Arrellanos Sunshine ‘69
was one of the rst hypertext novels written specically for the Web, and
it remains one of the best attempts to tell a coherent but multilinear and
polyvocal story in the distractive environment of the network. Sunshine
‘69 is a historical hypertext novel that attempts to encapsulate the zeitgeist
of the 1960s by tracking events in the lives of nine characters from June
through December 1969, concluding with the Altamont festival held on
6 December 1969. Arellano based his ction loosely on historical fact. At
different points in the novel, the reader can encounter a bird’s-eye view of
the historical events in 1969, and the work includes a bibliography of the
nonction sources that Arellano sampled from. While the novel references
historical fact, Arellano uses that context as a background for a largely
metaphoric tale of corrupted visions. In the novel, Mick Jagger makes a
deal with Lucifer that results in the tragedy at Altamont, and LSD is trans-
formed from a substance for utopian mind-expansion into a sinister market
The characters of Sunshine ‘69 are sketched very quickly as cartoonish
types—in one section of the hypertext, a page including a cartoon drawing
of a suit and a character sketch represents each individual character. The
atness of the characters helps Arellano to avoid the problem of slowly
developing characters in a novel that could be read in thousands of pos-
sible orders. The drawings don’t include any faces—as if to underscore that
these characters should be understood not as individual human beings, but
as stereotypes, ctional personalities representative of the cultural forces at
play in the novel.
The cast of characters of Sunshine ‘69 includes Alan Passoro, a Hell’s
Angel, hired for security at Altamont; Lucifer; the Glimmer Twins, Mick
Jagger and Keith Richards of the Rolling Stones; Ali a.k.a. Ronald Stark,
a shady agent provocateur with connections to the CIA; Meredith Hunter,
a young African-American hipster from South Berkeley; Orange Sunshine,
alternatively a hippie girl-next-door and a brand of LSD; Norm Cavettesa,
a discharged Vietnam veteran; and Timothy Leary, one of the leading advo-
cates of experimentation with hallucinogenic drugs.
Arellano freely mixed real people with ctional characters. Leary, Jagger
and Richards, real people in the world outside the novel, are also icons who
represent cultural movements. Hunter and Passoro are real people who did
not become icons: Hunter was the 18-year-old murdered at Altamont, and
Passoro was the Hell’s Angel who stabbed him. The other characters were
presumably ctional.
Real events, adapted from nonction texts, are juxtaposed with imagined
events to create an alternative history and to underscore the idea that all
histories are narrative constructs. In the context of the novel, the characters
who really lived through the events, or who died as a result of them, are
no more or less real than those imagined by the author to represent ower
children, government spies or allegorical evil.
The American Hypertext Novel, and Whatever Became of It? 31
A particular innovation of Sunshine ‘69 was the diverse range of navi-
gational options it presented to the reader, in addition to the in-text hyper-
text link. In the absence of the reading conventions of the book, authors of
nonlinear ction can provide readers with other navigational tools to guide
them. These tools can be as simple as the alternate reading order that Julio
Cortazar provides the reader of his Hopscotch (1966), or they can make
more elaborate use of the multimedia capabilities of the computer. Once the
readers get past the animated Flash introduction to the work, each screen
of the novel has four buttons linking to ‘Calendar,’ ‘People,’ ‘8-Track’ and
‘Map.Each button links to a different navigational apparatus, so that readers
can navigate by character, chronologically, according to musical selections, or
by a map.
Mark Amerika’s Grammatron (1997) was another expansive work of
hypertext and hypermedia narrative. Emerging from the context of the Alt-X
online publishing network and Mark Amerika’s then-popular Amerika
Online” column,
Grammatron retells the Golam myth in digital form.
The work centres on Abe Golam, a pioneering Net artist who creates the
Grammatron, a writing machine. The creature becomes a kind of combina-
tory monster, wherein all texts recombine. Throughout the work, Golam
searches for his second half—a programmer, Cynthia Kitchen, who could
provide the missing link to another dimension of digital being.
Grammatron included more than 1,000 text elements (some of them
scripted and some randomised), thousands of cross-links between nodes of
the text, many still and animated images, a background soundtrack and
spoken word audio. Grammatron was pushing toward a Gesamtkunstwerk
mode of hypertext writing and was as much a philosophical exploration of
network consciousness as it was a novel. It was also situated specically as
Net Art. Amerika made a conscious move to position himself in an art world
context in this and in later work. Grammatron was received as one of the
rst signicant works of Net Art and was embraced within the art world
context, having been exhibited at the 2000 Whitney Biennial.
With Grammatron, we can see several strands that would become more
marked in later years, including the move away from a specically literary
audience and an openness towards other cultural contexts, such as concep-
tual art and performance. Grammatron also marked a shift from hypertext
per se towards hypermedia, in which text is one of many media elements.
This multimodal shift became even more pronounced in subsequent years,
particularly with the rise of Flash as an authoring platform in the late 1990s
and early 2000s.
William Gillepsie, Frank Marquardt, Scott Rettberg and Dirk Stratton’s
The Unknown was the co-winner
of the 1999 trAce/ AltX competition
(Gillespie et al. 1999). A comic novel, it begins with the premise that the hyper-
text novel is itself a promotional stunt for a printed book, an anthology of
experimental poetry and ction. The hypertext is the story of the eponymous
authors’ book tour, which takes on the character andexcessesofarocktour.
32 Scott Rettberg
As The Unknown authors tour venues across the USA and abroad ranging
from small used book stores to the Hollywood Bowl, they have encounters
with literary and cultural celebrities ranging from Newt Gingrich to William
Gaddis, from Marjorie Perloff to John Barth, from Terry Gilliam to Lou
Reed. Complications develop, as one of the protagonists becomes a cult
leader before becoming a human sacrice; another becomes a mean and
withdrawn social outcast; and another becomes a heroin addict enamoured
of celebrity and its excesses. As their fame reaches its apex and a Hollywood
blockbuster is made of their hypertext ction, things generally fall apart.
Apicaresque novel with classic elements of a road trip novel, The Unknown
freely mixes writing styles and forms ranging from prose to poetry, credit
card statements to freshman composition writing assignments, pastorals to
corporate typing tests. Many scenes of The Unknown are parodies or trib-
utes to other writers: Scenes are written in the style of Jack Kerouac, Edgar
Alan Poe, Cormac McCarthy, Nelson Algren, Kathy Acker and many other
notable American authors.
Like many of the other hypertexts discussed, The Unknown tended
toward expansiveness and embraced excess. As the project was written and
distributed, the authors kept writing and adding new material for several
years after the novel was rst published on the Web and announced. While
the main component of The Unknown is a ctional narrative, a ‘sicken-
ingly decadent hypertext novel,the work also included several other lines of
content including documentary material, ‘metactional bullshit,’ correspon-
dence, art projects, documentation of live readings and a press kit (Gillespie
et al. 1999).
As the project progressed and after it had won an award, the authors
toured in person to a variety of venues and performed interactive readings
of the work in jacket and tie, ringing a call bell every time a link appeared
on the page, encouraging readers to interrupt and shout out a link to fol-
low whenever they encountered one they found particularly toothsome.
The majority of these readings were recorded in audio and/or video, and
those recordings integrated into the given page of the hypertext, so that
readers could listen to the authors reading the text. These travels also pro-
vided further material and settings for writing. The Unknown was thus
both a novel and a work of performance writing and in some respects also
a constraint-driven writing game. Like Sunshine ‘69, The Unknown made
extensive use of hypertext links to cross-link scenes of the novel and pro-
vided other indices and apparatuses for navigation. The links were used in
a variety of different ways. Sometimes they were used to guide the reader to
the next section of a narrative sequence, sometimes to provide further ref-
erential information and other times according to a more whimsical logic:
for example every time the word ‘beer’ appears it is a link taking the reader
to another scene in which beer is mentioned. Readers can follow the links
into the spiralling web of stories, or they can navigate via a “people” index
of celebrities and literary gures in the novel, a list of bookstores in which
The American Hypertext Novel, and Whatever Became of It? 33
reading scenes took place, and a map of the USA, each providing links to
episodes based on location. The different lines also each have their own
index. By including a series of web documentaries about the making of The
Unknown, documentation of readings and performances, a press kit of links
to popular media reviews of and scholarly articles about The Unknown as
well as correspondence between the authors, The Unknown gestures toward
a totalising encyclopaedic hypertext form. It is a novel that attempts to fully
integrate its own publishing and critical apparatuses. As the authors’ state-
ment published in the Electronic Literature Collection, Volume Two (only
half-jokingly) attests, The Unknown “attempts to destroy the contemporary
literary culture by making institutions such as publishing houses, publicists,
book reviews and literary critics completely obsolete” (Rettberg et al. 2011).
As was the case for the majority of the other hypertexts mentioned, The
Unknown had a number of metactional characteristics, including a num-
ber of asides on writing in the form itself, which are included in the “meta-
ctional bullshit” nodes of the text. In the node “Hypertext is/are Electronic
Space” William Gillespie, for example, mused in Deleuzian fashion:
Hypertext, to put it clearly, is a mapping of a text onto a four-
dimensional ‘space.Normal grammars, then, do not apply, and become
branching structures anew. Fragments, branches, links … The word is
glowing and on a screen. It is electronic and cannot be touched. It has
been copied over thousands of times and reverberates through virtual
(Gillespie et al. 1999)
Reection on the form of hypertext seems an almost inevitable outcome of
writing a hypertext novel. As writers rst encountered digital environments,
the works they produced were narratives or poems and expressions of ideas,
but they are also always explorations and experiments. Hypertext ctions
are often eld notes as much as they are ction.
One of the virtues of the World Wide Web is that it is an