HomeIntroducing CybercultureBook ReviewsCourses in CybercultureEvents and ConferencesFeatured LinksAbout RCCS

View All Books

Cybertext: Perspectives on Ergodic Literature

Author: Espen J. Aarseth
Publisher: Baltimore, MD: John Hopkins University Press, 1997
Review Published: January 2004

 REVIEW 1: Vika Zafrin

How do you review a book of fundamental importance to its field, six years after its publication? Espen Aarseth's Cybertext: Perspectives on Ergodic Literature appears on many new media and humanities computing reading lists, and is referenced at least a few dozen times at any given conference. Most every student and practitioner of humanistic informatics has read it -- or else they are ashamed to admit that they haven't, and are making mental notes to finally get to it even as they lie to your face. Any academic book that provokes interest this widespread deserves many reviews.

Yet, the reviews haven't been many. Besides the (by now well-known) review and ensuing discussion at ebr, I have only been able to locate one other, a review Dickie Selfe wrote for a graduate course he took several years ago. Granted, I did not search my local library, but went straight to the Internet. However, scant search results in the MLA bibliography indicate that there cannot have been many more print reviews of this work, and even if there were some, and if we throw in another, more general response to the theory contained in the book, the total would still be unimpressive. Here is another one, then, which after all this time cannot help but be a response itself.

Scrutinizing academic minds will have noticed that in the beginning paragraph of this review I used three terms almost interchangeably: new media, humanities computing, and humanistic informatics. In doing so, I hope not to provoke wrath but instead to call attention to the use of terms. Aarseth's study is of interest to fields distinct but intertwined, nowadays perhaps interdependent. For consistency's sake, I will refer to these fields as humanities computing. Included in its meaning will be the theory and practice of new media arts, as well as any "traditional" humanistic study (literature, history, sociology) which applies computing algorithms to data in order to accomplish its academic goals.

Aarseth is careful to reiterate his term definitions often. This would appear to lend an apologetic tone to the book, but is a rhetorical necessity. From the beginning, the reader is asked to return to the very basic concepts -- literature and text on one hand, cyber- and nonlinear on the other -- and re-evaluate them in light of the [text and game] writing which has emerged in the last few decades. Old habits die hard, we have grown to take certain terms for granted; and so even the nth reminder that texts are any semiotic sequences "meant to be read" (149) makes logical sense as part of an ongoing argument which attempts to reprogram our basic academic instincts. In this, Cybertext is similar to a well-constructed conference paper: recognizing the probable background(s) of his readership, Aarseth makes an obvious effort not to lose them to the complexity of the argument.

Omitting a preface and placing his introduction inside the first chapter, Aarseth cleverly encourages his reader to not skip any part of this study out of habit. The tactic works -- and it's a good thing, too, since this chapter really is fundamental to the rest of the book. The middle section (chapters on hypertext aesthetics, the adventure games, automated poetics, MUDs, and even the role of the reader) may be mixed and matched at will; but skip the first chapter, and the rest of the book will make only nebulous theoretical sense. This may still be useful to a reader; however, the author clearly favors comprehensive academic discussion rooted in concrete examples, and so reading the introductory chapter is essential to understanding the perspective he presents.

The complexity takes its toll already in the beginning. One must read carefully in order to catch the use of the term cybertext to denote at least three different concepts. On one hand, it is a "broad textual media category" (5, emphasis mine), on the other -- "not a 'new,' 'revolutionary' form of text . . . [but] is a perspective on all forms of textuality, a way to expand the scope of literary studies to include phenomena that today are perceived as outside of, or marginalized by, the field of literature" (18). In addition to being a category and a perspective, a single artifact may be referred to as "a cybertext" -- the term appears to be at least as versatile than even text is. In itself, this versatility of meaning is not problematic, but becomes confusing when it is not systematized in a concluding paragraph. This lack of emphasis gives a false impression of the term's cohesiveness, and adds an unnecessary level of obscurity to an already complex work.

One dangerously brave description of cybertext is found in the end of the first chapter. While admitting that distinguishing between a text and its readings is as necessary as it is impossible -- an ideal (20) -- Aarseth places cybertext squarely on a timeline. The text is not; it "takes place" (21) somewhere at the meeting point of operator, verbal sign, and medium (see Figure 1.1, The Textual Machine, on p. 21). Placing the focus on "the cybernetic intercourse between the various part(icipant)s in the textual machine" (22) gives it a necessary temporal, ephemeral dimension which places undeniable importance on the immediate process of perception and re(inter)action. The author indirectly reaffirms this in chapter nine. We viewers are denied co-creatorship of the images we perceive ("[t]he coded images are objectively there . . . and we can no more influence their appearance than we can influence a Rembrandt or a Van Gogh," 180). However, by approaching certain works in specific ways "[w]hat we have gained is a perspective, a mode of perception" (180). Here, Aarseth almost precisely echoes his early assertion of cybertext as a perspective; at least in this, of its many meanings, the term is heavily dependent on in-the-moment perception and reaction.

Such would seem to be the case with MUDs, which, while they certainly qualify as texts meant to be read, are also in practice read almost exclusively by their participant writers, as they are being written. Transcripts of MUD sessions are possible, of course, and sometimes useful as logs of particularly interesting conversations, or even of awards ceremonies (see ifMUD and the Xyzzy interactive fiction awards, for example). But for the most part, the MUD cybertexts are read as they are written, and the transcripts are echoes, a wholly different type of text, finite and for the most part not subject to editing by the reader. Oddly, Aarseth does not explicitly make the distinction between MUD sessions and their transcripts, although he does quote Elizabeth Reid to that effect (147). The rest of the chapter, however, is a wonderful discussion of the complexity and speed of MUD interaction. Here (158-161) Aarseth touches on netiquette and its influence on discourse; this section would have possibly benefited from a deeper comparison of MUDs to IRC, BBSs, and newsgroups, which would have also augmented the generic span of the study.

But what of the earlier chapter, titled "Paradigms and Perspectives," in which the distinction between a temporal experience of an object and the object itself is emphasized as necessary? "A piece of writing on paper or a computer screen should not be confused with the act of reading it," Aarseth writes (46). This would seem thoroughly inconsistent with other parts of the book, but instead underscores the multiplicity of meanings of cybertext. Here, again, the textual machine itself, a cybertext, is separated from the act of playing with it, which produces cybertext. This, in turn, reinforces what was described only a few pages before as "the unique dual materiality of the cybernetic sign process" (40). Admitting that "the internal, coded, level can only be fully experienced by way of the external, expressive level" (40), Aarseth nevertheless draws a distinction between the two and places importance on both. The level of importance placed upon code invisible to the user is surprising, if not paradoxical, in a book so focused on "the player's pleasure of influence" (4). On the other hand, Cybertext's reader is reminded here that the code is as much part of the textual machine as the content, and indeed, we are focusing not on "what [is] being read" but on "what [is] being read from" (3, emphasis author's).

The usefulness of the typology in chapter three is yet to be fully realized. This is the crux of the book, a tool for other scholars to comparatively study any texts in relation to any other texts. Relatively few such comparative elements have explicitly found their way into scholarship to date, which is a shame: a complex typology such as this must be tested on a large scale and in a variety of contexts if it is to evolve. One can only hope for this unique tool to be more practically applied as time progresses.

But the chapter's usefulness is not limited by the typology alone; it lies also in the very clear definitions of text. Aarseth's genius for clarity shines here: he describes text simply as "any object with the primary function to relay verbal information [which] cannot operate independently of some material medium [and] is not equal to the information it transmits" (62). Lengthier explanations follow, but the simplicity with which such an elaborate concept is worded is an achievement in itself.

While Dungeons and Dragons may be oral cybertext, and its successor the adventure game is certainly written, programmed cybertext (98-99), I will leave to the professionals the debate of whether interactive fiction is a flawed term. Whatever we wish to call the adventure game, it is always reassuring to read a separation between it and the novel. Cybertext goes further, and explicitly states that "[t]he adventure game is an artistic genre of its own, a unique aesthetic field of possibilities, which must be judged on its own terms" (107). To this end, I very much look forward to the release of IF Theory and Twisty Little Passages.

Introducing us to cyborg literature in chapter six, Aarseth points out that "the computer will never become a good traditional author, if only because it cannot criticize or appreciate its own work" (131). From this it is logical to conclude, as he does in the following pages, that traditional literary aesthetics -- indeed, traditional systems of poetics, such as Aristotle's -- are inadequate critical tools for the study of cybernetically produced literature. Given these, the analysis of an Eliza-like responsive bot program, a story generator, and a set of theoretical principles for interactive drama is well executed. More importantly, it needs make no claims for an objective aesthetic "greatness" of these works, only for their being functionally interesting: a refreshing approach, because it is both unapologetic and academically sound. Finally, the chapter is simply interesting to read, excited as it is about "failures" of cyberliterature over its "successes"; and this makes it all the more unfortunate to witness the argument fall into two inconsistency traps. First, Aarseth refers to "narrative media" (141), which is odd if one remembers narrative being referred to as a "formal category" earlier (85). (If most of Cybertext refuses to associate any medium with any specific formal category, why start being lax with terms now, even in passing?) In the very next paragraph there is a call for computer-generated literature to "focus on the computer as a literary instrument: a machine for cybertext and ergodic literature" (85, emphasis mine). This, after a (quite rightful) invective in the introduction against "the essentialist idea of 'the computer medium' as a singular structure of well-defined properties of communication" (19). These are not disastrous faux pas in the carefully choreographed waltz of academic terminology, but they struck me as significant in the context of the present study.

"My main effort is," Aarseth wrote in his introduction, ". . . to show what the functional differences and similarities among the various textual media imply about the theories and practices of literature" (17). Still, it is only in chapter eight that Cybertext touches on scholarly hypertext. The Perseus Project and The In Memoriam Web are the two primary examples brought in by the author. Even when they are brought in, these texts are discussed almost exclusively in a political -- not functional -- context. Whether individual examples of scholarly hypertext are cybertexts may be debatable; on the other hand, the book does include an entire chapter dedicated to hypertext aesthetics, which in fact alludes to, but does not discuss, "translating codex texts into hypertexts" (76). In that chapter, Aarseth discusses in depth Michael Joyce's afternoon and Stuart Moulthrop's Hegirascope as examples of hyperfiction, but does not make the obvious leap to a discussion of hypertext construction as literary study, and the impact of material reconfiguration on both the study and the reading of literary texts originally written on paper or presented orally. Perhaps this would have been too big a can of worms to open in this investigation. However, no discussion at all of literary hypertext's scholarly uses makes for an omission too large in a book which looks at similarities and differences among textual media. Including it, on the other hand, in the chapter on democracy and politics in social electronic spaces charges the book with an unexpected, jarring political twist and promotes more divisiveness than dialogue.

Other textual media, particularly Web-based, have seen a lot of development in the six years since the publication of Cybertext: weblogs and various forms of news services, to name only two examples. Aarseth aimed to "construct a model of textual communication that will accommodate any type of text" (18) early on in the life of the Web, and although the study's media scope was certainly not limited by things electronic, the Web has allowed cybertextual communication to evolve in rich and interesting ways. Cybertext is an excellent beginning in constructing a typology of textual machinery, solid groundwork which was never intended to be completist, as the author himself states in the concluding pages. Now and for the foreseeable future, Aarseth's cybertext typology deserves a re-evaluation and evolution of its own.

Vika Zafrin:
Vika Zafrin is a doctoral student in humanities computing at Brown University. When not thinking about intercultural transmission pathways, she serves on the editorial boards for The Decameron Web and Heliotropia.  <vika@wordsend.org>

©1996-2007 RCCS         ONLINE SINCE: 1996         SITE LAST UPDATED: 12.10.2009