1. Ted Nelson, the inventor of the term, said "By 'hypertext' I mean non-sequential writing. Familiar text is a linear sequence of parts. Those parts can be of varying sizes and can be nested within one another (sentences, paragraphs, sections, chapters, and so on) but they are presented in a linear order, one after another. It is true that books are not perfectly linear; they may have footnotes, cross-references, and indexes. Yet although the parts of the text may refer to one another non-sequentially and we may choose to read by skipping about in the book, the text comes to us in a preferred order. Authors work hard to create a clear and convincing sequence of narrative, exposition, or argument. It is not easy to decide how to order the text, and the best rhetorical order may not be the order of dependence of the ideas involved. A hypertext, by contrast, is a web of pieces of text. The individual units of the web may be sentences or chapters, but what is crucial is that they do not relate to one another in any unique sequence. The web is like a map or a landscape: many different routes emerge. The author cannot control which links the reader will follow, and in some systems the reader can create new units and links so that the web changes and grows. (from Socrates In the Labyrinth, Eastgate Systems)
2. The structures of ideas are not sequential. They tie together every which-way. And when we write, we are always trying to tie things together in non-sequential ways. . . . people keep pretending they can make things hierarchical, categorizable and sequential when they can't. . . . In an important sense there are no 'subjects' at all; there is only knowledge, since the cross-connections among the myriad topics of this world cannot be divided up neatly. Hypertext at last offers the possibility of representing and exploring it all without carving it up destructively. (from Ted Nelson, Computer Lib/Dream Machines. Seattle: Microsoft Press, 1987.)
3. What is it that makes hypertext different from linear text? Abstractly speaking, in hypertext the discrete units of text with multiple connecting links compose a differently connected textual space. Hypertext can be visualized as a web or network or space whose multiply connected points, related by quanta of distance (one link, two links, three away) allow varying paths connecting the points in the space. Multiple paths can be followed along the links, though not every point is connected to every other (if they were there would be no information value to the links). Points can be near on one path, far apart on another. There may be varying degrees of hierarchical order imposed on the space; a group of items could form a highly connected island with only a few links to other islands; there could be clusters of clusters of clusters of such islands. Links can be one-way or allow travel in both directions. Paths through the space can be linear, looping, branching, or intersecting. (from Socrates In the Labyrinth)
4. There is no reason to think that there is one essential or best way to use hypertext, any more than there is one essential or best way to use paper. Linear writing can produce laundry lists, thank-you notes, scholarly treatises, disjointed jottings on the backs of train schedules, love letters, instruction manuals for VCRs, romance novels, and countless other texts whose forms and purposes have little in common. We can imagine how hypertext might help or alter activities that we already perform with ordinary text, but it is harder to imagine what new activities or uses might develop. (Although we already know one--a new form of cooperative writing.) (from Socrates In the Labyrinth)
5. Hypertext . . . the nonlinear or not sequential space made possible by the computer . . . multiple paths between text segments . . . webs . . . networks of alternate routes (as opposed to print's fixed unidirectional page-turning) . . . favoring a plurality of discourses over definitive utterance, and freeing the reader from domination by the author. Hypertext reader and writer are said to become co-learners or co-writers, as it were, fellow-travelers in the mapping and remapping of textual (and visual, kinetic and aural) components, not all of which are provided by what used to be called the author. . . . .Much of the novel's power is embedded in the line, that compulsory author-directed movement from the beginning of a sentence to its period, from the top of the page to the bottom, from the first page to the last. Of course, through print's long history, there have been countless strategies to counter the line's power . . . true freedom from the tyranny of the line is perceived as only really possible now at last with the advent of hypertext . . . where the line in fact does not exist unless one invents and implants it in the text. . . . No fixed centers, for starters--and no edges either, no ends or boundaries. The . . . line vanishes into a geographical landscape or exitless maze, with beginnings, middles, ends being no longer part of the immediate display. Instead: branching optional menus, link makers and mapped networks. There are no hierarchies in these topless (and bottomless) networks [of] evenly empowered and equally ephemeral window-sized blocks of text and graphics. (Robert Coover, "The End of Books." New York Times Book Review , June 21, 1992.)
6. No, hypertext is not about lack of linearity. For me, hypertext is about the necessary combination of nonsequential and linear. There is never a lack or complete absence of linearity. (Carolyn Guyer, commenting during an on-line conference at PMC-Moo, October 1993.)
7. A philosophical argument (just as a mathematical proof)Ê cannot be a cloud of disjointed statements. Hence the philosophical line cannot be dissolved in the way some have dreamed of dissolving the narrative line. And thus philosophical hypertext would have to respect the line by making arguments the units of presentation (from Socrates In the Labyrinth)
8. Hypertext does make total closed works impossible--new links can move in and reuse pieces of my writing. But whatever form the author gives the writing remains available. It cannot dominate the hypertext space and it cannot claim to totally determine the meanings and roles of the individual lexias, but its presence remains. Otherwise the hypertext would have no structure save as a cloud of individual lexias. (from Socrates In the Labyrinth)
9. The key to using hypertext to show logical structure is to be willing to mix presentation of structure within a single writing space, and hypertext links to other writing spaces. Let single writing spaces present passages of argument, diagrammed or analyzed however is convenient. Let hypertext links take the reader to the arguments that support the parts of the argument presented on the page. Don't try to show all the structure either on the page or in the hypertext links. The single writing space would show fine structure where individual propositions relate to one another over a relatively small span that does not tax the reader's short-term memory. The hypertext links would show how larger blocks of argument relate to one another. Such a mixed presentation would take advantage of the power of prose paragraphs to present fine structure, and also of hypertext's ability to lay out typed relations between blocks of prose. By not using hypertext to model every move, we would assist perception rather than burdening it. (from Socrates In the Labyrinth)
10. Hypertext needs to discover ways to enact complex interactions that go beyond simple implications or 'topic and comment' connections. (from Socrates In the Labyrinth)
11. We should distinguish the general edgelessness of hypertext, and its ability to mix genres, from what might be possible through new figures of linkage and the resultant new modes of reading. If hypertext can do something different that linear prose it will be in the linkages and the way the reader co-constitutes of the text. There are ways of using hypertext that emphasize that, and other ways (indeed the predominant ways, if one were to count all those informational data bases) that do not. The same would be true of philosophical hypertext. (from Socrates Apology)
12. Can hypertext display, for instance, the process of first putting up an idea as independent or immediately given, then showing how it is involved in complex relations that qualify its independence and constitute a larger unity that both keeps and overcomes the seeming independence of the first stages? Though not a linear argument, this process demands links that do multiple duty and should be traversed from a beginning. For the reader to intercept only a part of this movement would be to see as independent something that is not. To leave the first links behind would be to miss an aspect of the larger unit. To lose the movement in a cloud of links would be to diffuse the investigation into comments that presuppose too much independence in the individual moments of the discourse. (from Socrates In the Labyrinth)
13. As Kant pointed out, philosophy and mathematics are crucially different in that mathematics starts with definitions while philosophy ends with them (Critique of Pure Reason, B 741ff). Philosophy discusses alternative first principles and disputes definitions and modes of argument. It tries to find a rigorous--or probable, or persuasive--way of disclosing the conditions that make rigorous argument possible. (from Socrates In the Labyrinth)
14. We see another model of philosophical writing by noting that philosophy's line finds itself constantly surrounded by supplements which it both desires and rejects: marginalia (as in medieval manuscripts), parallel columns of text (as in Kant's antinomies), and material in parentheses or footnotes that provides self-critical comments, amplification, methodological reflections, objections and replies, ironic juxtapositions, historical precedents and deviations, references to other texts, quotations, and so on. (from Socrates In the Labyrinth)
15. The argumentative line is surrounded by a fluid discourse in which there are no fixed primacies and no firm meta-levels, because it is in that discourse that such dependencies get established. The fluid discourse gives argument its bearings. Hypertext seems to be a medium in which this fluid discourse would flourish. (from Socrates In the Labyrinth)
16. Linear argumentative critical thinking is only one mode of thought's responsibility for itself--a responsibility which is sometimes a responsiveness to itself rather than a guard mounted over itself. There are more skills of thought than argument, more kinds of connection than logical necessity, more ways of self-response than self-reflection, more spaces than we know how to move within. (from Socrates In the Labyrinth)
17 There is the danger that creating a hypertext web would be the functional equivalent of writing without self-discipline: publishing drafts and jottings, self-indulgently exfoliating ideas without taking any position. This would infect the philosophical work with well-known diseases: wandering commentary, endless qualifications, fruitless self-reflection, unnecessary contentiousness, the deadweight of meta-level upon meta-level. This could both stem from and result in intellectual laziness. It could also cater to an uncritical audience that wanted to be titillated by the passage of ideas, but not to be challenged in its beliefs or values. But perhaps working up one's drafts and jottings into good philosophical writing is precisely a matter of the critical mastery and judgment that demand and produce a line. What would "thinking" mean if it were not providing form and focus, critical judgment, beginnings, middles, and ends, and preventing the indefinite accumulation of words and images? This is an open question. The openness tells us something important. (from Socrates In the Labyrinth)
18. What kind of responsibility does the text take for itself when it does not have complete control, when there is no single total transparent reading? This makes the questions about hypertext figures not so distant from current ideas about any text and any reading. (from Socrates In the Labyrinth)
19. There is a "knowing how to go on" which is: knowing how to read, how to continue the series, how to use the tool. This is a matter of finding yourself, of accepting descriptions, of knowing what sort of place you are in. This is not established by argument; it is presupposed in being able to follow an argument. (from Socrates In the Labyrinth)
20. Emerson's finding of founding as finding, say the transfiguration of philosophical grounding as lasting, could not have presented itself as a stable philosophical proposal before the configuration of philosophy established by the work of the later Heidegger and the later Wittgenstein, call this the establishing of thinking as knowing how to go on, being on the way, onward and onward. At each step, or level, explanation comes to an end; there is no level to which all explanations come, at which all end. (Stanley Cavell, This New Yet Unapproachable America. Chicago: University of Chicago Press, 1989, 116)
21. The line tends to become encrusted with its preconditions, with meta-comments and amplifications. Suppose it dissolved into them? Imagine a text that became nothing but footnotes and marginalia referring one to another. This would say many things at once without any primacy; could it still be philosophy? It sometimes seems that contemporary deconstructive texts aspire to this status. (from Socrates In the Labyrinth)
22. Maneuvers that reveal the necessarily slippery existence of anything that claims to be an essence or final unity. These maneuvers let the essence or boundary be, but show that in order for it to do what it claims it must be also undermining its own closure. It exists in a space it cannot dominate, and its claims must be revisited. (from Socrates In the Labyrinth)
23. No matter how extensive, a hypertext web remains an artifact with defined components and links. The unlimited economy or general textuality cannot be made present, for the act of making-present is the creation of a limited economy. The general economy can only be indicated indirectly, because while it is a condition of the possibility of any defined signifier or network, it does not exist in the usual sense as an encounterable signifier or network. (from Socrates In the Labyrinth)
24. Some writers, inspired by deconstruction, want to liberate us from limitations by instituting the general economy in writing or in community, and they then seek institutions or ways of writing that will do the job. But I take deconstruction to be as much about the impossibility of such an instantiation as about the impossibility of a completely closed text or economy. (from Socrates In the Labyrinth)
25. Question: Despite the insistent questioning of forms and modes of argument in your essay, the philosopher-as-subject in your essay seems rather conventional, even Cartesian/classical-- consciously knowing, in total control of the discourse and its effects. In what ways might postmodern conceptions of the subject-position of the philosopher in discourse serve to change or reinforce your opinions about the possibility of a hypertextual philosophy? Or reversing the question: in what ways might hypertextual writing change the postmodern conception of the death of the author or the disappearing subject? Derrida's critique of the philosophical tradition continually refers to the attempts by philosophy to define its own borders by inclusion/exclusion (Tympan comes to mind) and thus ensure the philosopher's possession of his/her work; does the possibility of a hypertextual philosophy call into question this time-honored assumption in the philosophical tradition? (from the Seulemonde editors)
Response: Does the possibility of a hypertextual philosophy call into question this time-honored assumption [of a subject in total control] in the philosophical tradition? This question doesn't seem clear: if philosophy is in the old self-possessed style, then doing it in hypertext does not of itself call the assumption into question. If it is in a new style then philosophy has already called the assumption into question, and hypertext is not the crucial factor, though it may amplify the effect. It's clear that hypertext can't be a necessary condition of questioning the assumption (of possession of the work), since that assumption has already been questioned in the linear texts that the questioners are appealing to. It's also clear from the questioners' very questioning of Socrates that neither is hypertext a sufficient condition of such questioning. (Unless they think that Socrates isn't really hypertext, but that's a dubious essentialism.) So it seems that the possibility of a hypertextual philosophy does not by itself call into question the time-honored assumption. (DK's response, in Socrates Apology)
26. We, all of us, have been at one time or another guilty of hymning the virtues of schizophrenia and media addiction, the ecstasy of self-liquidation, and the utter loss of boundaries. But have we lost sight of the dialectical fact that intertextuality, ambiguity, indeterminacy often enough themselves lie on the side of dominant ideological discourses? (K. Michael Hays, "Architecture Theory, Media, and the Question of Architecture," Assemblage 27, Fall 1995,45)
27. Part of this task is discovering how to run meta-dialogues and critical discussion in a medium where there is no privileged spot reserved for the Critic (hypertext, but also the Internet and all that it presages). We still have not adapted our thinking to a world where hierarchies are at most local. What does it mean to be critical when the critic cannot speak for the Center or the Universal? Which is not to say the critic speaks only for a local prejudice--but putting together these two insights is no trivial task. Hypertext is only part of an answer. (from Socrates In the Labyrinth)
Last modified 13 October 1997 by David Kolb