The palpable sense of handling -- as distinct physical beads -- the crown jewels: that's what it felt like to prepare for this review by transcribing (by hand, using Graffiti handwriting software on a Newton) and then manipulating exact quotes from this continuous song of a book by Michael Joyce. Though surely there are in this amazing book some points to which one might object, I find myself standing in wonder at the breadth of riches it contains. A definitive collection of essays dating back to 1987 by the man universally acknowledged as the father of hypertext fiction, this book is cause for celebration by anyone with an interest in literary hypertext. The book is organized into three parts: 1 Of Two Minds: Hypertext Contexts; 2 Siren Shapes: Hypertext Pedagogy; and 3 Contours: Hypertext Poetics; but that list hardly begins to suggest the breadth of topics covered. I found my list of saved quotations aggregating around the topics Programming, Polemic, The Sheer Joyous Images, Exploratory vs. Constructive Hypertext, Contour, Pedagogy, Spatiality, Meaning, Reality (Virtual and Otherwise), Time, Mind -- and even this list seems to me hideously incomplete. An entire essay (or several!) could easily be devoted to discussing just the topic of exploratory vs. constructive hypertext, which is an important recurring theme in most of these essays: "Scriptors use constructive hypertexts to develop a body of information that they map according to their needs, their interests, and the transformations they discover as they invent, gather, and act upon that information. More than with exploratory hypertexts, constructive hypertexts require a capability to act to create, change, and recover particular encounters within the developing body of knowledge." [emphasis mine].

I was extremely intrigued by the section on pedagogy. Michael Joyce's credentials as a postmodernist are impeccable, and yet here you will find detailed protocols for the use of hypertext in the classroom to teach what many in the literary profession would shudder to have to do: remedial composition. If anyone believes hypertext writing is just for eggheads after reading part 2 of this book then I give up, such people are not reachable. That is not to say this section of the book is any the less literary:

"We redeem history when we put structure under question in the ways that narrative, hypertext, and teaching each do in their essence. Narrative is the series of individual questions that marginalize accepted order and thus enact history. Hypertext links are no less than the trace of such questions, a conversation with structure. So too, the networked classroom is a place of 'making do' as constructive action. All three -- narrative, hypertext, and classroom -- are authentically concerned with consciousness rather than information, with creating knowledge rather than the mere ordering or inventory of the known. The value produced by the readers of hypertexts or by our colearners is constrained by systems that refuse them the centrality of their authorship. What is at risk is both mind and history."

There are a few places in this book where I find myself taking distinct objection. In places the degree of polemic does reach the level of being an unguided missile: "The book is an obscure pleasure like the opera or cigarettes. The book is dead -- long live the book." I know many hypertext writers feel this way, but I happen to feel we are not doing our cause any good by this kind of talk. (It is worth noting that the "obscure pleasure" of this book extends to the binding: it's sewn!) More serious is the issue of programming -- or to be more accurate, the absence of a serious consideration of programming. We hear frequently that "Hypertexts are read where they are written and written as they are read." True enough -- but what about programming? To this reviewer, omitting the programmer from a discussion of writer/reader duality is to miss the essential fact that hypertext involves a triad of reader/writer/programmer; to exclude the programmer from the reader/writer rhetoric is as serious a problem as excluding the reader-as-writer from a discussion of reading. Example: "How, in the landscape of the city of text, can the reader know that what she builds will move the course of the river?" She could do worse than to start by reading the code. (If the code is not accessible then that itself is an important issue for hypertext rhetoric.) Or: "Hypertext programming -- in the double sense of instructed machine behavior and information content: the thing that, like situation comedy, weather map, or docudrama, is shown on the screen -- is privileged, centralized, and self-sufficient." Privileged? Self-sufficient? How incongruous this sounds amidst the many pleas for constructive over exploratory hypertexts. The degree to which hypertext software systems should be extensible -- precisely so the programming is not "privileged" -- is contentious; suffice it to say that this reviewer believes that systems must be extensible.

Surely the most challenging enterprise in reading this book -- and the one most likely of highest priority for those in the hypertext research community -- is deciphering Joyce's absolutely essential concept of contour. (One notes with a wry smile that at one point Joyce himself pleads guilty to writing with obscurity on this subject.) The contour is: An intersection in pathways through the many lexia / An outline, a memory, a resonance in the replacement of one writing by another / A density from following multiple links / A contour line, an altitude, an ache of criss-cross as we stretch the end-to-end link iteration from one heart-foot left behind -- these are guesses on my part. The contour is: a multiplicity. As an extremely young poet, I, like many others of my generation faced a kind of rite of passage in poetics in wrestling Charles Olson's famous essay "Projective Verse" into some form of ready raw-material congeniality; it was something no one could do for you. So too to understand Michael Joyce -- another breather in that being-sigh of reception from Olson, the archeologist of morning -- when he speaks about the contour is to undertake, not to be handed a cute package. And yet, one is convinced -- haunted perhaps -- this is indeed the heart of the matter: the topographic after-image in the face of the constructive act formed of so many individual bits of interactive doing.

Those of us who write interactive literature believe, down to our very bones, that a literary hypertext is what it is through the interactive choices of the reader; if a few simple sentences strung end to end would serve just as well then we would put those into play instead. We put in play text interlaced with interactive structure operations because we believe nothing else will serve the purpose. Given that, it is worth asking whether we can expect that something as complex as the mechanism by which the reader arrives at a resonance shape in how a hypertext means could possibly be formulated in a few simple sentences of prose. A hypertext is a journeying system; in asking an author to do more than set out beaconing launch points toward our understanding of how the energy in a hypertext works is perhaps asking too much. To ask the reader to make do with a neatly formulated few simple sentences of guidance is perhaps asking too little.

I close where I began, with the last and first word about Michael Joyce: that he writes so excruciatingly well, those of us who follow can only read such words as, "We are the children of the aleatory convergence. Our longing for multiplicity and simultaneity seems upon reflection an ancient one, the sole center of the whirlwind, the one silence" and murmur. We nod: oh yes, oh yes

Jim Rosenberg's home page o Poetics