On-paper printing allows for great variation in textuality, and is as capable of producing complex experimental texts that subvert the fixity of meaning as is electronic/digital hypertext. Yet recent claims by Peter Lurie in "Why the Web Will Win Culture Wars for the Left" make the claim that hypertext encourages a mode of reading that is culturally revolutionary (see CTheory:a125). Lurie argues that hypertext exploits Saussurian arbitrary signification more than any other linguistic media, but does not consider the history of experimental literature that has perhaps already accomplished some of this. Despite the fact that digital publishing is creating a mode of reading that is original and is thus starting a second printing revolution, hypertext itself is not as revolutionary as its proponents claim -- nor as damaging to traditional print as its opponents fear. Hypertext is a fascinating textual form, and one that I think offers several innovations in printing, but it borrows so much from traditional print that its formal qualities cannot represent a break from on-paper print. In Hypertext 2.0, George Landow offers a brilliant theory of hypertext, but his theory has limits in its audacity. Consider the following claim, for instance: "hypertext offers the possibility of presenting a text as a dispersed field of variants and not as a falsely unitary entity." Yes, but so does on-paper print; hypertext is not the first and only form of text to challenge print's unitary tendency even if it is the most fruitful. This quote is one of many I could choose to indicate that Landow overextends his arguments. Thus, this set of lexias is as much a response to Landow as it is a demonstration of the ability of an on-paper text to embrace its own dispersal.
But this text -- the one which you are reading -- might not be a hypertext. Actually, you will find that it is not a hypertext, and yet it is similarly decentered in all of the ways that Lurie valorzies. I am attempting to treat the differences between on-paper printing and electronic/digital hypertext with a set of lexias that each treats a different aspect of this difference. For my purposes, I have decided that this format best allows me to display the capacity of on-paper print to de-center itself in the hypertextual manner. As Jerome McGann writes in The Textual Condition, "theory of texts cannot move long or far from specific examples and case studies." In addition to discussing Laurence Sterne's eighteenth-century novel Tristram Shandy, I will use this set of lexias itself to demonstrate that the printed text shares much with the hypertext and is not so much being replaced as emulated by hypertext.
Please, reader, remember that this text is not a hypertext. It aims to show that any printed text can be de-centered and dispersed through its own design. (Its own closed design, as it is.)
The text is limited by the number of pages it uses, the number of words, the words its author chooses, and the order of those words. The text is further limited by its physicality, be it ink on a sheet of paper or the specters of words on a computer screen. No matter how many copies of a text are being read, each copy can only reveal certain things about itself before the duration of reading ends. Unlike the scribal text, the printed text show only minor differences among its copies -- a lightly-printed page, a missing page number in one volume, and such. The printed text is otherwise a closed set of words arranged into chapters, lexia or stanzas. No variations can be made without physically altering the text. The creative text only pursues this closure further, and also seeks relative permanence. As Jerome McGann writes, "literary works are distinct from other linguistic forms in their pursuit of extreme concrete particularity." Unlike a conversation, a text is fixed against time.
Print also places spatial limits on words by bringing them into physical being. They must sit in neat rows, obey grammatical laws, and not spill off the page. Few printed texts have words carelessly arranged on a page with no care to appearance and meaning. As Walter Ong writes, "print situates words in space more relentlessly than writing ever did. Writing moves words from the sound world to a world of visual space, but print locks words into position in space. Control of position is everything in print." A printed word cannot be moved, nor can it be altered in print. At best one can resort to inscription to cover it with more ink, or with a correction fluid to bury it under more empty space. There is little room for a reader to change the words without removing them completely. One is correct to write that print is "intolerant of physical incompleteness." Yet print on paper affords the possibility to put words in order in a way that is not possible orally or scribally. The closure a text seeks brings with it meaning, poetics, and ambiguity. The closed text might exclude certain elements of language, but it cannot become seamless. Both word meanings and the physical properties of print -- like typefaces, font sizes, and arrangement -- work against the completion of a text.
A hypertext is as prone to seeking closure as any other text. Hypertextual linking affords connectivity between closed documents, thus opening them up to endless extension, but linking cannot alter the composition of hypertext. It is, after all, neatly arranged words arranged into neat lexias. Hypertext might utilize new printing technology to defer its closure, but it is still afflicted with the textual condition. Hypertext is still a group of words frozen -- perhaps momentarily, unlike on paper -- in time and space where the very fragility of that freeze is readily apparent. Just like other print, hypertext undermines its essentially closed nature by pointing it out. The hypertext shows its seams in an attempt to loosen them. This is in keeping with the literary traditions of on-paper printing, as described by Jerome McGann in The Textual Condition: " 'literary' work, in its textual condition, is not meant for transparency, is not designed to carry messages." A literary text is not supposed to be a seamless message lacking paradox, ambiguity and contradiction. The literary text can even be dispersed, like the hypertext.
In Hypertext 2.0, George Landow finds traditional print to be doomed to only creating fixed texts with no polycentricity. He understands that "without fixity one cannot have a unitary text" without delving into the history of textual fixity in print. Landow ignores countless tendencies of printing that work to disclose the structure of closure that makes texts possible. These tendencies are resisted by publishers who do want to sell seamless messages, but they still prevail occasionally. Jerome McGann is astute in his discussion of the book designed by William Morris, which reveals its closed physicality at the same time that its author uses it to produce a beautiful, polycentric text that employs literary devices contingent upon mechanical print itself. In other words, McGann acknowledges that texts by Morris and others simultaneously rely upon and resist mechanical printing, and are openly posed as products of technology: "there is a sense in which any device that calls the reader's attention to the constructed nature of texts contains an implicit or possible 'mode of resistance to the literary work of art'." Hypertext provides similar tools, which Landow praises, but hypertext is as much a technological product as printed books are. If the "mode of resistance" of Morris's books was through typography, than the corresponding mode for hypertext is the link; both modes allow the text to disclose its limits and then attempt to transcend them.
Landow observes that "we find ourselves, for the first time in centuries, able to see the book as unnatural, as a near-miraculous technological innovation and not as something intrinsically and inevitably human." He is correct to state that the closed-off nature of texts has been addressed by recent critical theory. Yet he should note that hypertext itself is as much an unnatural thing as is on-paper printing. Landow defends hypertext for being the same thing that on-paper print is: a technology-dependent version of textuality. Yet both forms allow writers to write texts that are closed-off from the external world while remaining inwardly open to different readerly interpretations. Hypertext retains the drive to enclose the text within certain boundaries. Landow's theory of hypertext needs to heed the words of Walter Ong: "despite their cultivated air of spontaneity, these [electronic] media are totally dominated by a sense of closure which is the heritage of print."
Is hypertext really an original form of textuality? George Landow claims for hypertext an essential originality although he does note that its predecessors introduced some of its features. Still he announces hypertextuality as if it is a break with all other textual forms, a very difficult claim to believe: "Conventional notions of completion and a finished product do not apply to hypertext, whose essential novelty makes difficult defining and describing it in older terms, since they derive from another educational and information technology and have hidden assumptions inappropriate to hypertext." Surely new terms are needed to precisely articulate a theory of hypertext, but most of the old vocabulary of textuality is still applicable. Landow is still caught on questions of technology, the reader, the author, origins and meaning. These are the same concerns raised throughout the history of textual studies and seem not to represent a big break at all, but a new application. Unlike the first incunabula, the first hypertexts are not introducing a new form of textuality.
It's not clear that on-paper textuality has exhausted its creative possibilities. After all, it is not yet a thousand years old. As Elizabeth Eisenstein points out, on-paper printing is not a unifying technology at all: "Concepts relating to uniformity and to diversity -- to the typical and to the unique -- are interdependent, they represent two sides of the same coin. ... The more standardized the type, indeed, the more compelling the sense of an idiosyncratic personal self." The on-paper text has a lot in common with hypertext in that it allows for a diverse range of practices within a set of fixed conventions. Hypertext allows for a new textuality in which some conventions -- the paper page, the author -- are replaced with other, original ones. Perhaps hypertext is a new set of textual conventions and not a new textual form.
There is no doubt that the best hypertexts are completely devoid of a monocentric or even concentric structure; their links allow for simultaneous access to the text that create so many paths through the words that not every reading will encounter every word, or even words encountered on previous readings. The readings are possibly always new encounters with the text. The one center to text -- the main path through which all others must pass -- is gone. As Landow notes, "hypertext creates an almost embarrassingly literal reification or embodiment of a principle that had seemed particularly abstract and difficult when read from the vantage point of print." The de-centered text with its dispersed meanings is wholly realized in some hypertexts. Yet not every hypertext disperses its meanings, and none forsakes center formation. Each hypertext creates at least one center because the one reading always calls into being its own excursion. Many hypertexts then allow other readings to make other excursions, so that different centers are possible. No hypertext that creates meaning is completely de-centered.
Also, no hypertext accomplishes an embodiment of a principle that is impossible in print. Printed texts have polycentric structures as well, although their technological limits often preclude elaborate experiments. Landow errs in suggesting that polycentric printed texts are difficult to write -- he might instead observe that few writers are willing to write them. Print technology left its innovation stage centuries ago and now seems to resist the production of polycentric texts. Clarity and closure are valued by many writers and readers, although not all. It is likely that hypertext writers will begin to adapt to emerging conventions that the new print technology will introduce. Already, the constraints of the popular WWW are changing hypertextual writing practices.
These conventions will be similar to those with which traditional print has been shackled; they are the work of users and not of the technology itself. Print can accommodate elaborate experiments should one want to make them. Even the blandest, apparently closed text can be read to elicit new meanings. New paths are possible even when an author and other readers are working to preclude that possibility. As Jerome McGann notes, "every text has variants of itself screaming to get out, or antithetical texts waiting to make themselves known." There is little that is final in a printed text, even when there is much that is closed. A closed set can still yield an inestimable number of permutations, or reading centers. One closed text still has to meet millions of readers before it realizes its potential for making meaning. Even if the text will not willingly yield many different readings, the chance is great that one readers will find it where no one else was reading. McGann's "radial" reader empowers a text to contain a new reading path with each reading by each reader.
This is much like hypertext, of course. Landow notes that "electronic text always has variation, for no one state or version is ever final." The hypertext can allow its reader to become a writer. That is difficult with printed text, but not impossible. Again, the difference in the possibilities lies in the technology used to publish: the printed text can provide space for response in reading, writing marginalia or filling in forms, and in creating meaning; the hypertext can allow the reader to add to the printed text, an authoritative act that undermines the unity of the text more visibly than anything print allows. Both printed texts and hypertexts are never final, because all of their instances are copies. Differences in technology are responsible for traditional print's seeming more final and unified than hypertext. Yet traditional print is not inherently monocentric, nor is it necessarily final. Hypertext simply uses digital and electronic technologies to attempt to surpass the polycentric inertia of paper printing.
The reader of the hypertext is, according to George Landow and others, a "wreader" who is as much a producer of texts as a consumer of them. In this new reader, both production and consumption of texts is combined into one process that is self-contained. The new reader navigates through lexias to find threads of connected meaning where no author placed them. This reader has made the author only a little more authoritative a source on texts than a stocker of the shelves in a library. This new reader is reminiscent of the old reader who has always decoded texts and made new meanings with them, prowling them for paths that go toward new textual centers and make new experiences. These readers are what Jerome McGann terms "radial" readers, which means that they read texts in an open-ended search for meaning. McGann humbly describes their reading: "radial reading involves decoding one or more of the contexts that interpenetrate the scripted and physical text." The radial reader finds a lexia and then tracks it back through the text and into others to find different paths of reading, to test the resistance of a text to a new center. Radial readers are those who find hypertextual links too restrictive because they limit the range of lexia that one might read into and out of a given text. Nonetheless, they are almost exactly the sort of readers that are reading hypertext -- yet they have been around for years. In fact, McGann writes about them primarily to discuss how they read on-paper texts. Even Sven Birkerts, who is hostile to electronic and digital publishing, notes that the mode of reading is more important to the making of meaning than is the text being read (or its publication method): "we are, for the duration of our reading, different, and difference has more to do with the process than the temporary object." Birkerts can rest assured that hypertext is only enhancing radial reading's cultural impact, although for how long is not certain. New hypertexts might adapt to Internet technology's image-based meaning systems and render Landow's new reader archaic.
Printed texts have introduced many of the elements that George Landow claims are unique to an electronic or digital hypertext. In fact, hypertext offers few narrative inventions at all when one looks at the feats of writers like Calvino, Borges, Silko and Joyce. The few innovations hypertext relies upon are those that are made possible through new technologies, but are not properly textual features. Landow writes that "Hypertext, which challenges narrative and all literary form based on linearity, calls into question ideas of plot and story current since Aristotle." Yet hypertext only continues this questioning, which has been ongoing since writing began. Landow obscures the difference between electronic/digital publishing and hypertext so that the two seem intertwined. His attribution of revolutionary narrative abilities to hypertext is an overwrought celebration of the new publishing's possible enabling of what has been a counter-tradition -- the nonlinear, fragmented text -- to become a dominant tradition.
Any other attribution of narrative innovation to hypertext is unjustified, as Laurence Sterne's novel Tristram Shandy -- which Landow credits as a precursor to hypertext -- illustrates. Shandy is a novel that undermines its narrative attempts to establish a unity of time, location, voice and action. Nothing certain lasts here; the only thing that is fixed is the language that is used to build the novel. Sterne himself is nearly kicked out as author at the novel's start. Many intertextual elements appear: a marbleized page; an approximation of a tombstone; and the narrator Tristram Shandy's sketches of his own meandering narrative paths. Yet the most effective device to undermine the unity of the novel is a very simple component of on-paper printing: the dash.
In Tristram Shandy, the dash is a prominent item of punctuation that serves to do many things in the novel. Sterne obviously is not alone among early eighteenth century writers in widely using dashes to make shifts in sentences, and by no means is every dash in Tristram Shandy significant in and of itself (a point which will be covered shortly). Yet within this novel, the dash is on nearly every page of text, in every character's speech, and in every chapter. The dash here serves multiple functions: it binds disjoined phrases in sentences, it begins paragraphs, it stands for omitted letters in prudently censored passages. Tristram Shandy is pervaded with dashes, but the effect is not exclusively to articulate Sterne's (and Tristram's) rambling style. Here, the abundant dashes make Tristram's storytelling possible through their adhesive properties. The dash serves to tie Tristram Shandy's disparate parts into a cohesive, associational whole -- it serves as a literary approximation of the point of contact between thoughts that drives the human mind's associations.
In Tristram Shandy's syntax, the dash not only binds together related phrases, but it serves to connect seemingly unrelated phrases and words to sentences in which the meaning Tristram already sees is made clear to the reader. The first paragraph of the "Author's Preface" illustrates how the dash joins the related into the whole of a paragraph so that Tristram's words create two meanings -- their literal meaning as phrases and the meaning that their paragraph has: "No, I'll not say a word about it -- here it is; -- in publishing it -- I have appealed to the world -- and to the world I leave it; -- it must speak for itself." Most of the phrases are related to each other and would almost be a sentence without the dashes. But individually, each phrase's meaning is different from the next: "here it is" does not even closely mean "in publishing it," and so forth. The dash -- the moment of connection -- literally joins phrases and makes the total meaning unambiguous.
Of course, what is seen on the page represents what happened in Tristram's mind as he originated and associated the thoughts that made the paragraph. Since there is association with joining throughout Tristram Shandy, there is consistent meaning in thought and word. There are places in the novel where a disjoined phrase or thought will arrive as its own paragraph, which alone does not accommodate any connection to preceding paragraph's meaning. Yet the dash visually and syntagmatically joins the paragraph with the afterthought. Consider this passage:
It is for this reason, an' please your Reverences, That key-holes are the
occasions of more sin and wickedness, than all other holes in this world put
-- which leads me to my uncle Toby's amours. 
Tristram is returning to his intended story of Toby after diverting his thoughts to a story about his father. When his thoughts return to Toby, the dash lets him close the gap between the thoughts; the gap is that transition between thoughts that is the moment of association. This dash allows Sterne a literary device to capture a highly abstract, highly vital component of psychology. Tristram Shandy becomes a more striking approximation of an individual mind because it contains, in print, the associations as well as the actual moments of associative linking.
Even Sven Birkerts realizes that the text of lexias is a valuable literary form. In fact, Birkerts argues that it might be the only form that captures the experience of perceiving in this (post)modern world. In this assertion, he goes farther than a lot of the theorists he criticizes, who tend to extol hypertextual innovation but accept the conventions of on-paper printing as inherent properties. Not so for Birkerts, though, who writes:
"Not only must [the contemporary novelist] figure out to do with the flatness of quotidian experience, but he must also deal with the fact that the greater part of human activities . . . now take place on many tracks at once, with the individual in a state of distracted absorption."
Only the multivalent text will do in these times; anything less is destined to fail to use print's capacity to reify the tendencies of this age. Of course, Birkerts is fearful of hypertext and admits no positive interest in it. His call for a multivalent, polycentric textual experience implies that this will take place in an on-paper page written by Kathy Acker or Italo Calvino.
Still, one must note the coincidence of his argument with that of Landow. Landow also finds the multivalent text to be the most advanced text for this age, and that is why he sees hypertext as the ultimate textual experience. Yet his principles are really not dependent on electronic or digital print technology, but describe a textual form:
"The presence of multiple reading paths, which shift the balance between reader and writer, thereby creating Barthes' writerly text, also creates a text that exists far less independently of commentary, analogues, and traditions than does printed text."
This text could very well be the new novel Birkerts envisions, or it could be Tristram Shandy. Landow shares with Birkerts a confusion of textual and technological properties: both ascribe their fear or wishes to properties that are technological, but they both value a set of textual characteristics that are in accord. Birkerts and Landow disagree on the properties of printing technology without realizing that their arguments about reading and writing are actually arguments about textual form that are not as far apart as their rhetorical stances would indicate.
There is no doubt that authorship is undergoing a fundamental change due to the proliferation of electronic and digital publishing. There is also no doubt that authorship as a writer's stance is largely a particular element of print culture that altered scribal and oral modes of attribution. Authorship as most readers and writers know it emerged from the print revolution and will not survive the digital revolution intact. The author as a solitary authorizer (or authority) of the reading experience is undergoing a change that Roland Barthes aptly called a "death." This death does not mean the death of on-paper print, just as the birth of the author did not mean that oral and scribal attribution were gone forever. This death means a shift in the dominant form of textual attribution, just as the birth of the author did. As Elizabeth Eisenstein writes in The Printing Press as an Agent of Change, "The wish to see one's work in print (fixed forever with one's name in card files and anthologies) is different from the desire to pen lines that could never get fixed in a permanent form, might be lost forever, altered by copying, or -- if truly memorable -- be carried by oral transmission and assigned ultimately to 'anon'." The electronic or digital text that is infinitely alterable and infinitely linked has no one authorizer but many collaborators who build the text for the reader, who also builds the text through linking and choosing reading content and sequence. George Landow states that "all writing becomes collaborative writing" with hypertext. He is largely right, except that this shift has more to do with electronic and digital media than with a specific textual form (which has more theoretical importance than immediate social impact).
Yet Landow's conflation of technological and textual modes leads to his admission that it is indeed technology that is leading to the dispersal of authorship and the demise of the old "author." As he writes, "conceptions of authorship are a matter of convention, and they relate to whatever information technology currently prevails." Authorship is a function of the technology and the social use of that technology. Or, as Jerome McGann defines it, "authorship is a social and not a solitary act or set of acts." This is as apparent in the Bible, a supposedly authoritative text, as it is on the texts that compose the online publication Salon. Factor in the editors and designers and one sees that the author is a very specialized role that cannot bring about a text alone. The author today only recognizes him or herself as a writer, whereas before he or she claimed a solitary role. The writer today acknowledges many collaborators, especially the reader. Yet this is not a hypertextual innovation but an erosion of old printing conventions through textual innovations that Laurence Sterne and others set into motion centuries ago. The result is that the author has relinquished any claim to being the originator of the text. Sven Birkerts is wrong when he writes that "in the case of hypertext it is not God who is gone, but the author, the traditional originator of structure and engineer of meanings." The function of the author remains, modified to accommodate all of the voices that compose a text. "The author" is indeed gone, but meaning and structure remain.
Looking at the dash on a macro level in Sterne's Tristram Shandy, its role in organization of the entire novel rejects Tristram's formal plan for Book VII, which resembles one column-long dash. Although that Book turns out to be another recall of associations, it is important that Tristram begins it with the long dash form in his mind. He presents it half-seriously, claiming it as his course and then ridiculing the "straight and narrow" metaphors for human behavior moralists employ. For example, one way the line is used to delimit proper behavior: "This right line, -- the path-way for Christians to walk in! say divines --." Tristram is not using the line of a dash to limit himself, so this consideration calls into doubt his plan to write a straight line. Here, Tristram reaffirms his use of the straight line to order thoughts but not to symbolize the order of the novel itself; it is not possible then for the Book VII plan to be an enlargement of the dash. After all, the novel is trying to get beyond the proscriptions of fixity to a more authentic particularity, and the small dash -- not the long straight line -- reveals this complex area better. If used, the straight line model would supersede the dash and replace the integrated whole of each rambling Book with one whole that could not be broken into parts -- something that hardly corresponds with the view of the human mind Sterne constructs in the novel. The dash would be of no use in such a plot model, as the role of the dash is to order the disorder of Tristram's mind.
Conversely, the disorder of the different thoughts and segments would be meaningless without the dash to organize them. If the dash here stands for the points of contact that organize the vast human mind, at least for those ways in which humans associate memories and mental images, its role in the organization of Tristram Shandy clearly demonstrates that role. Using Tristram's models of four of his books , one can consider what the novel would be like without its dashes:
These broken lines compare to the actual lines:
Clearly, the loops and dips of the mind are not cohesive without points of contact. Without the dashes arising between fragments (they cannot be simply added later), Tristram's mind would never be revealed to the reader in a way that allows the reader to respond to the text. If the above fragments indicated what Tristram related to the reader of his life and thoughts, the reader would not be able to understand Tristram Shandy nor Tristram Shandy. This novel tries to capture those free-flowing abstractions that compose the individual human mind and thus has to integrate its own parts. But it never loses sight of the dual whole the parts must make: the mind as represented by the novel. Just as the dash is not a very significant organizing force in a linear narrative, it makes all the difference in the novel of Lockean psychology. The dash in Tristram Shandy makes the parts of Tristram Shandy's mind into a whole mind and a whole novel by literally becoming the figurative moment of association. It is Landow's "link," albeit one that holds together a unified text by linking what might be mere fragments.
There are textual elements in the novel which work against the unifying force of the dash, namely the use of grouped asterisks (the blank and marbleized pages being accounted for in the text). At times, the asterisks only omit letters in words, names, and sentences that are bawdy or Tristram chooses not to have the reader know. But occasionally, the asterisks stand for music or other elements that cannot be put into words -- but must still be related to those parts that have been approximated by text. So the dash is used to join words to sounds too. When Uncle Toby sings "Lillabullero" in Book IX, Chapter 20, both instances of his singing are represented by spaced asterisks -- and both are began and closed with dashes. Associations in Tristram Shandy go beyond connecting remembered incidents (stories that can be effectively written), and sounds -- Toby's repeated singing, swearing -- are commonly recalled. Even characters' speaking is mostly denoted by dashes rather than quotation marks. Sterne does not fail to use the dash to connect all sensual memories to each other, and hence is able to make his novel a literary approximation of the mind rather than a literary reduction of the mind -- and again, the dash makes possible a linear narrative where each sentence is nonetheless the site of dispersal.
Extolling the virtues of hypertext, George Landow suggests that hypertext might be the form of text that brings about a complete de-centering of power. He discusses the ways in which hypertext has eroded the notion of author as producer and reader as consumer, replacing it with a more fluid and shared power to produce between the two. The reader cannot be overpowered by reading hypertext, or so Landow argues. He writes that
"I contend that the history of information technology from writing to hypertext reveals an increasing democratization or dissemination of power. Writing begins this process, for by exteriorizing memory it converts knowledge from the possession of one to the possession of more than one."
The reader becomes a possessor of information and then a producer of it, or at least can recall information as authoritatively as any author. Landow discredits on-paper printing's previous de-centering of power, but he most glaringly omits the ways in which electronic and digital publishing has been manipulated by advertisers, censors and the like. Textuality is empowering, just as Landow writes, but print technology itself is as much an enabler of manipulation as it is of dissemination of power. Access in itself is not empowering if one's voice cannot interact with the text.
 Landow, George. Hypertext 2.0: The Convergence of Contemporary Critical Theory and Technology (Baltimore: Johns Hopkins UP, 1997), p. 68.
 McGann, Jerome. The Textual Condition (Princeton, N.J.: Princeton UP, 1991), p. 177.
 Ibid., 182.
 Ong, Walter. "Print, Space and Closure" (The Press of Ideas; Julie Bates Dock, ed; Boston: Bedford, 1996. 52-64.), p. 54
 Ibid., p. 59.
 McGann, p. 76.
 Landow, p. 54.
 McGann, p. 83.
 Landow, p. 25.
 Ong, 63.
 Landow, p. 79.
 Eisenstein, Elizabeth. The Printing Press as an Agent of Change (Cambridge: Cambridge UP, 1979), p. 84.
 Landow, p. 65.
 McGann, p. 10.
 Landow, p. 64.
 McGann, p. 119.
 Birkerts, Sven. The Gutenberg Elegies: The Fate of Reading in an Electronic Age (New York: Fawcett Columbine, 1994.), p. 81.
 Landow, p. 181.
 Sterne, Laurence. The Life and Opinions of Tristram Shandy, Gentleman (New York: Quality Paperback Book Club, 1992.), p. 153.
 Ibid., p. 488.
 Birkerts, p. 206.
 Landow, p. 25.
 Eisenstein, p. 121.
 Landow, p. 104.
 Ibid., p. 300.
 McGann, p. 64.
 Birkerts, p. 161.
 Sterne, p. 382.
 Ibid., p. 381.
 Ibid., p. 506-7.
 Landow, p. 277.