In this lecture we will be looking at the important concept of hypertext. We will look at its history; and at various notions of hypertextuality. We will consider its effect on our cultural understanding of narrative, linearity, authorship, textual closure and artistry. Our theoretical discussions will then be placed in context by looking at some of the work being produced by text artists working with hypertext. We will then consider the importance of hypertext to notions of knowledge, in particular the pedagogical role of hypertext within universities and schools. The lecture will conclude with a creative workshop in which the group will produce a small collaborative hypertext.
The hypertext theorist George P. Landow describes hypertext as:
"Texts composed of blocks of text and the electronic links between them" (Landow 2001 pg 100)
However, it is important to note that Landow then extends this simple description by defining another term hypermedia, which includes images, sound, animation and other forms of data. In doing so, he recognises that these elements are also part of the hypertextuality. As he notes:
"Hypertext, which links a passage of verbal discourse to images, maps, diagrams, and sound as easily as to another verbal passage, expands the notion of text beyond the solely verbal." (Landow 2001 pg 100)
It is this structure of elements (text, image, animation etc) and their links that will be discussed in this lecture.
Landow thinks that hypertext is highly significant because:
"Hypertext, which permits readers to choose their own paths through a set of possibilities dissolves the fundamental fixity that provides the foundation of our critical theory and practice"
This lecture will explore why he makes this claim, and some implications of his theory.
Firstly, it is important to clarify the use of the term hypertext being used in this lecture as 'hypertext' is used to identify and a number of related issues. The novelists and critic Umberto Eco notes that there are at least three definitions of the term.
1. Hypertext as a system (analogous to a dictionary and a grammar). E.G. The World Wide Web
2. Hypertext as text within that system E.G. The hypertext novel Victory Garden by Stuart Moulthrop
3. The notion of a hypertext as an unlimited and infinite 'writing space'.
This lecture will consider all three, describing 1) as a hypertext system, 2) as a hypertext text and 3) as a specific theoretical example of 2.
Hypertext enthusiasts often use the printed book to define the very opposite of hypertext, describing reading printed text as a linear, fixed reading experience. They often do this to highlight the value of writing on the screen. In fact, a sizable section of early criticism of electronic writing was written in this 'book vs screen' vein (see 'Writing Space' by Bolter and 'The Gutenberg Elegies' by Sven Birketts). However, a number of critics and authors have come recognise the hypertextual quality of certain types of printed text. In particular, the academic article with its use of apparatus such as footnotes and end notes which allow the reader to link from one place in the text to another. As Landow notes:
"The standard scholarly article in the humanities or physical sciences perfectly embodies the underlying notions of hypertext as mulitisequentially read text" (Landow 2001 pg 100)
Also certain printed works of art, such as Italo Calvino's If on a winter's night a traveller, have also been recognised as having a hypertextual quality.
In his introduction to Reading Digital Culture David Trend notes that 'Digital Culture did not emerge from a vacuum.'(Trend 2001 pg3). This is particular true of digital hypertext systems such as the World Wide Web. Historians have identified a number of systems, stretching back over the last fifty years that have helped to develop our contemporary notions of hypertext. It is worthwhile studying aims and ambitions of these precursor systems in order to understand the current implementations of hypertextuality.
Trend begins his account of hypertext with Vannevar Bush's 1945 essay 'As We May Think'. In this essay Bush describes a mechanical microfilm system designed to address the problems of the 1940s researcher trying to handle the burgeoning amounts of published work. Bush identified that the paper-based systems of his day had a number of problems, in particular in that they relied on artificial indexing systems (e.g. alphabetic filing). As he notes, these kinds of system were hard to use because the human mind works by association not by catalogue and index. Bush describes this process of mind as follows:
"With one item in its grasp, it snaps instantly to the next that is suggested by the association of thoughts, in accordance with some intricate web of trails.." (Bush pg10)
This notion of associative indexing led Bush to identify two of the chief characteristics of hypertext systems.
1) The ability to make and follow links. As Bush notes 'the tying of two items together is the important thing' (Bush pg12)
2) The ability of authors to record their mental trails for others to use.
As we shall see both of these features are core aspects of the World Wide Web.
The term 'hypertext' was coined by Ted Nelson in the 1960s. Nelson was very influential in the development of the Internet and devised his own hypertext system called Xanadu. In many respect Nelson was ahead of his time and his ideas were not taken up with the development of the Internet in the 1990s. This is probably a pity as many of Nelson's supporters regard Xanadu as being superior to the World Wide Web in that it solves many of the problems of current web usage such as 'broken links'.
The Xanadu web site notes the following features of Xanadu.
In their work, Remediation, Bolter and Grusin argue that the new media are characterised by dominance of the visual arts. In making this case they are following in the footsteps of Marshall McLuhan who argued that the modern era (in his case that meant the 1960s) was dominated by film and television.
However, the World Wide Web is full of graphically-rendered text. In addition, the systems that support the web are designed to support textual way of thinking. In fact it has been argued by the hypertext theorist and novelist Michael Joyce that text is the dominant mode in the new media. He notes that:
"Hypertext has been called the revenge of the text on television since under its sway the screen image becomes subject to the laws of syntax, allusion and association, which characterise written language" ( Joyce 1995 pg 23)
You might be interested to know that the story has not ended with the World Wide Web and IE6. At the recent Hypertext '03 conference Ted Nelson described the currrent www as being " inherently flawed". Nelson then presented his vision for the future of hypertext - Zig Zag project. He notes that it is:
"a new structure for building anything-- crossed lists in many dimensions.... ZigZag structure may greatly simplify casual database, casual programming, user-reconfigurable sofware, and casual visualization (made easier by the availability of multiple dimensions)." ( See Zig Zag website)
It will be interesting to see whether any of his ideas will be adopted by W3C.
The rest of this lecture is going to look at the at the work of certain hypertext artists who have begun to explore the artistic potential of hypertext and create works of art that highlight certain characteristics of the medium. It will do this to investiagte the claims made for hypertext by Lanodw ( quoted earlier).
The hypertext writer and critic Carolyn Guertin noted that there have been three generations of artists working with hypertextuality. The first generation were programmer-artists working with hand coding or with the earliest version of Apple's HyperCard. This generation drew the first maps of electronic texts and explored the concept of automated links. Everglade by Tim Willmot, for example, is a long poem in hypertext, each node of which offers a self-contained stanza of blank verse. It was written in Willmot's own DOS-based hypertext environment, Orpheus. The poem uses the metaphor of 'doors' to introduce links between nodes of the poem. Interestingly, Willmot's program hid these links, (they could be found by clicking the right mouse button) so that the poem read in a transparent fashion like a printed text (A full description of Willmot's work can be found on the Electronic Labyrinth site). Most text now highlight their links, providing an example of a switch from immediacy to hypermediacy.
The second generation began with Michael Joyce's Afternoon. This is the seminal hypertext work and the text most often discussed by literary critics such as George Landow. This hypertext 'novel' comes in a CD-ROM format and is supported by an Eastgate Systems program called Storyspace. This work is regarded as a classic as it realises a number of features of the genre, such as the artistic use of links, the uncertainty of the narrative and the importance of the decisions of the reader as they navigate the text. Other examples of hypertext include Victory Garden by Stuart Moulthrop. More recently, Eastgate systems and others have made some of their hypertext available over the World Wide Web.
Guertin argues that the current crop of multimedia work, that use text, graphics, sound etc represent the third generation of hypertext. Others have argued that the increased use of conditional programming to create complex event handling and user interaction represents a new departure. Aarseth for example has characterised these kinds of text, cybertext ( see discussion below). Terminology in this area is hotly disputed and different critics often use different terms to highlight different features of the same work.
A number of critics and hypertext authors have described hypertext art works as being non-linear. They note that the reader does not follow the prescribed order of words set out by the author. This has been described as being in contrast to the printed book. In this kind of traditional text the author (with the help of her editor and reader) selects the precise word order of a text. For example, these notes you are reading now conform quite closely to a traditional notion of linearity despite being a web-based document. They have a beginning (which frames and contexts the following text), a middle (that comprises of a series of logical points - one following on from the other), and it has an end (that acts to conclude and close the text that precedes it). A reader could start with the first word and progress through it word by word to the end and find a coherent reading. As such these lecture notes reflect a mode of thought that theorists have closely associated with book-based literacy.
It is important to note this reading is not the only kind of reading one could make from these notes. One could jump down to the bibliography or select a link. It is also important to note that this does not mean that every reader experiences, simply that text from which they draw their reading is fixed.
There are many paper based texts that one reads in a non-linear fashion ( e.g. the telephone directory). However, traditional narrative techniques have been closely allied to writing that is based on the model of a beginning, middle and end.
Hypertext, however, do not have a fixed word-order. The user makes choices that effects the texts available for reading. In fact, sections can be read twice and in many cases, whole sections of the text will not be read (because they were available as links that were not selected during the reading).
The classic model of hypertext, as represented by a text like What Fits by Adrienne Elsen (available on the www in the Eastgate Reading Room - see link below) comprises of a series of textual lexia ( small bits of text) with a series of author-determined links. They can be imagined to be like a series of flash cards. It is important to note that the author has predetermined the text of each individual lexia and also predetermined the links. The order in which they are read, however, is not prescribed.
Many proponents of hypertext have described this property of hypertext as being non-linear. But what exactly do they mean by non-linear?
The whole notion of non-linearity in hypertexts (and by association linearity) has been surrounded a cloud of confusion, mainly as result of wildly different usages of the same term. In his essay, 'Wiggenstein, Genette and the Reader's Narrative in Hypertext' Gunnar Liestøl notes the importance of distinguishing what exactly we mean by non-linearity. As Liestøl points out:
"One must distinguish between non-linearity in time and non-linearity in space. Non-linearity in time is imaginary; it is a fundamental contradiction of terms and necessarily impossible. Time is linear, at least the time that is required to read and write hypertexts. Reading and writing are linear phenomena, they are sequential and chronological, conditioned by the durative ordering of time, although their positions as stored and in space may have a non-linear organization" (Liestøl 1994 pg 106)
This is important, as hypertext enthusiasts have often used the 'non-linearity' of their texts as a way of distinguishing them from page based works. However, as Liestøl points out:
"nonlinearity is never actually experienced directly. It exists only as a logical negation and at distance on both time and space..the moment one reads into non-linearity by clicking on one of the icons… one reduces non-linearity to linearity" (Liestøl 1994 pg 106)
In this hypertext is simply carrying on a recent literary tradition as there has been considerable interest in non-linear writing amongst contemporary authors. There have been a number of formal experiments. 'Happy Endings', for example, a short work of prose by the Canadian Author Margaret Atwood is laid out in a linear fashion on the page. However it is structure into a series of choices ( labelled A, B ,C etc.) The text encourages the reader to treat the text like a simple computer program (see hand -out).
Computer-mediation, however, does provide a unique opportunity to work in this kind of non-linearity. The cybertext theorist Aarseth argues that there are four degrees of non-linearity in computer mediated works:
"1) the simple nonlinear text, whose textons are totally static, open and explorable by the user;
2) the discontinuous nonlinear text, or hypertext, which may be traversed by jumps ( explicit links) between textons;
3) the determinate "cybertext" in which the behaviour of the textons is predictable but conditional and with the element of role playing ; and
4) the indeterminate cybertext in which textons are dynamic and unpredictable." (Aarseth 1994 )
Clearly, these kinds of non-linear structures will have serious implications for the art of narrative. The author is no longer able to specify a 'one and only' order of the text.
Writing techniques such as 'the surprise', 'the cliff-hanger', and the 'twist-ending' all depend upon a close control of word-order. Does this mean that we will not be able to narrate stories in the new media? Carolyn Guertin argues that in successful hypertext the narrative is an emergent property that is brought into being by the acts of reading and selecting links.
A number of recent literary theories have been invoked to describe the peculiar properties of hypertext. Perhaps the most famous theorist whose work has been cited in these discussions of hypertext is Roland Barthes. Barthes is famous for creating the idea of the 'readerly' text in which the reader is free from the meaning imposed on the text by the Author-God. Barthes once famously declared that:
"the birth of the reader must be at the cost of the death of the Author"( Barthes 1988 pg 172)
Proponents of hypertext fiction argue, that hypertext frees the reader from the author's control, As such they herald it as an example of the 'Death of the Author'. However, it is important to note that it also imposes duties on the reader such as the duty of navigating the text. Is this really freedom?
As the critic and hypertext author J. Yellowlees Douglas notes:
"Like so many extensions of our sensibilities, hypertext is an odd creature, one that simultaneously promises more autonomy to its readers while offering authors a degree of control unthinkable with more conventional materials - seemingly as two-faced as any politician's promises." (Douglas 2000 pg 125)
Is it really possible to dispose of the Author? Is it desirable?
Michel Foucault, the French philosopher and historian, notes that the author plays a vitally important role in defining a work of art. He states that:
"The author's name is not simply an element in a discourse; it performs a certain role with regard to narrative discourse, assuming a classificatory function" (Foucault 1989)
He continues:
"He (sic) is a certain functional principle by which, in our culture, one limits, excludes, and chooses; in short, by which one impedes the free circulation, the free manipulation, the free composition, decomposition and recomposition of fiction." (Foucault 1989)
Hypertext challenges this notion of the author, and the ability of authorship to define or limit a work. In the classic hypertext the user is also an active agent in the construction of the text.
With the creation of network-able and interactive art even greater challenges to the traditional notion of the author are realised. In recent works, such as Jane Prophet's Swarm, the artist invites the users to add their own texts to her site. This aesthetic is sometimes described as mosaic or bricolage. Prophet argues that her work is an example of 'hive mind' (as discussed in a previous lecture). These kinds of work raise a number of interesting questions. Is this a single text with multi-authors? Who is in control of the text? Who gets access to the database? Who gets to edit the work? Is this a singular work of art or an aggregation? What makes it so? Why has the site only got Prophet's name on it?
A number of critics have noted that hypertexts are like labyrinths. They twist and turn. They provide endless choice of paths, and in doing so they appear to constantly bar the reader's progress . A number of critics have also noted that it is almost impossible to find the end of a hypertext novel like Victory Garden. It just keeps on coming, proving new links, new texts and insights into the story. Does this mean it is not possible to find closure when reading a hypertext? What does it mean when we say I have finished reading Victory Garden?
It has been argued that a book-based cultural bias has led us to conflate the end of a codex book (with its blank pages and back cover) with the end or 'closure' of a reading. However, this is not the only kind of closure available to a reader even when reading a printed book. For example, one does not look up a phone number and then read to the end of a telephone book.
This kind of selective 'spot' reading highlights the importance of the reader's involvement in creating an end to their reading. As hypertext theorist and writer J.Yellowlees Douglas notes:
"Once we dispense with closure as an entity that is always determined by the author and always consumed by the reader… we can clear ourselves a bit of neutral ground to examine what defines an ending beyond the blank space accompanying it."(Douglas 2000 pg 88)
Douglas then goes on to suggest a strategy for reading contemporary hypertext fiction. She suggests that:
"Our sense of arriving at closure is satisfied when we manage to resolve narrative tensions and minimize ambiguities, to explain puzzles, and to incorporate as many of the narrative elements as possible into a coherent pattern… Unlike most print narratives , however, hypertext fiction invites us to return to it again and again, its openness and indeterminanacy making our sense of closure simply one "ending" among many possible." (Douglas 2000 pg 88)
The properties of hypertext discussed above have raised a number of issues for scholars and teachers interested in using it for teaching purposes. In particular, they see the use of hypertext in the classroom as a marked departure from the book-based practices of conventional pedagogy. Kathleen Duguay, for example, notes that in producing an academic essay:
"Traditional writing emphasizes the presentation of a clear, coherent thesis supported by examples and illustrations leading to a definitive resolution or closure" (Duguay 1999 pg15)
Duguay argues that hypertext document stands in mark contrast to this. She notes that:
"Instead of a single thesis, a hypertext can offer multiple positions and viewpoints privileging none of them" (Duguay 1999 pg15)
Is this desirable in the classroom situation? In what situations?
Beyond Hypertext
Hypertext theory attempts to explain computer-mediated textuality in terms of links between fixed sections of alphabetic texts called lexia. However, in many contemporary computer-mediated textual works it is hard to identify such lexia. For example, The Virtual Disappearance of Miriam, by Martyn Bedford and Andy Campbell the reader is presented with a richly-layered, coloured animation that continues to loop until a link is selected (Bedford and Campbell 2001). The lowest layer of this animation is an enigmatic dark-green murk in which it is hard to pick out much more than idea of lines and texture and overlaying this is a ‘searchlight’ animation and two rotating wire-frame images of a human head. There is also a loop of electronic music. Clearly, the alphabetic text is only forming a small part of the overall semiosis of this work.
The new media and hypertextuality offer the creative artist a number of exciting new opportunities to explore. However, it is worth noting that some of these new media works of art will probably have a short life span. The cyber-poet John Cayley sounded a mournful note recently when he noted at the end of a conference speech in 2002 that most of the works he had produced in the 1980s no longer ran on any system being used today. A number of his important works have effectively been lost (until someone takes the time to port them over to a newer system).Only time will tell whether these works have been truly lost and whether that lost matters. However as the poet Peter Howard notes in a recent article:-
"Anyone who uses computers as part of their art has to allow for or accept that what is ground-breaking today may be buried tomorrow."
It is intersting to speculate what effect this sense of uncertain future has on the works that we create.
Aarseth, Espen J (1997) Cybertext: Perspectives on Ergodic Literature Baltimore and London The John Hopkins University Press
Atwood, Margaret (1998) 'Happy Endings' pg 1059 -1061 in the Riverside Anthology of Short Fiction: Convention and Innovation edited by Dean Baldwin Boston and New York Houghton Mifflin
Bush, Vannevar (2001) ' As We May Think' pg 9-13 in Reading Digital Culture edited by David Trend Malden, MA and Oxford Blackwell
Calvino, Italo (1981) If on a winter's night a traveller translated by William Weaver London. Secker & Warburg.
Douglas, J. Yellowlees (2000) The End of Books - Or Books without End; reading interactive narratives Ann Arbor The University of Michigan Press
Duguay, Kathleen (1999) 'State of Conflict: The Challenges of Hypertextualizing Composition in the College Writing Class' in Contexts, Intertexts and Hypertexts Cresskill, NJ Mampton Press
Eastgate Systems are available at http://www.eastgate.com/
Electronic Labyrinth is available at http://www.eserver.org/elab/elab.htm
Elsen,. Adrienne What Fits is available at http://www.eastgate.com/ReadingRoom/WhatFits/WhatFits.htm ( 28/11/2002)
Foucault, Michel (1988) 'What is an Author' in Modern Criticism and Theory edited by David Lodge London and New York Longman
Howard, Peter 'Flash Poetry' is available on
Nelson, Ted -Ted's Homepage is available at http://ted.hyperland.com/
Landow, George P 'Hypertext and Critical Theory' in Digital Culture edited by David Trend Oxford and Malden, MA Blackwell
Liestøl , Gunnar (1994) 'Wiggenstein, Genette and the Reader's Narrative in Hypertext' in Hypertext Theory edited by George P Landow Baltimore and London John Hopkins University Press
Moulthrop, Stuart (1991) Victory Garden Watertown, MA Eastgate Systems
Project Xanadu is available at http://www.xanadu.net/
Trend, David 'Introduction' in reading digital culture edited by David Trend Oxford and Malden, MA Blackwell
Zig Zag deatils of Ted Nelson's latest hypertext project can be found at http://xanadu.com/zigzag/
Tomorrow http://diffie.luton.ac.uk/~gs/