So what does it mean to read? Do we really know anything about the process? This is Birkerts on the interiority of the relation between the reader and literary language: "Isn't this the most elusive and private of all conditions, that of the self suspended in the medium of language, the particles of identity wavering in the magnetic current of another's expression? How are we to talk about it?" (Birkerts, p. 78).

Although many claims are made about reading, there is no independently valid account of reading, its varieties or its difference on different occasions (for example, when I read Milton for the narrative in contrast to studying his relationship to Virgil). One might say, claims Landow, that hypertext linking embodies "the way one actually experiences texts in the act of reading; but if so, the act of reading has in some way gotten much closer to the electronic embodiment of text and in so doing has begun to change its nature" (Landow, p. 82). But since we don't know much about reading of any kind, this claim is quite speculative. This comment also contains a normative assumption: that reading should become that modelled by hypertext. The same assumption occurs later: "Unlike users of 'most information systems, hypermedia users must be mentally active while interacting with the information'" (cited, Landow, p. 220). This begs several questions about the nature of reading, such as the receptive phase suggested by Birkerts's "shadow life" of reading.

A hypertext system is assumed to efface the novice/expert distinction, that is, that the reading of an expert differs only on instrumental grounds. An electronic Bible, for example, Landow suggests, "by making some of the scholar's procedures easily available to almost any reader, this electronic Bible might demystify a text that possesses a talismanic power for many in its intended audience" (Landow, p. 84). But the drive to desacrilize (evident generally in the attack on the canon) does not simultaneously deliver the procedures, experience, or wisdom of the scholar.

Demystification also flows from treating text as data in a computer. It is not clear how seriously this proposal is intended by Landow: "A data base search . . . permits the active reader to enter the author's text at any point and not at the point the author chose as the beginning" (Landow, p. 94). Text, of course, is more than "data." A text that has been marked up and made available for text analysis is sometimes called a "textbase": but this makes no presuppositions about the act of reading. Note the implications of Landow's statement for narrative.

The nature of reading is said to change according to the medium (McLuhan again), argues Birkerts: "by moving from the order of print to the electronic, we risk the loss of the sense of obstacle as well as the feel of the particular that have characterized our experience over millenia. We are poised at the brink of what may prove to be a kind of species mutation. We had better consider carefully what this means" (Birkerts, p. 31). This is another example of reifying the medium: compare reading (say) Paradise Lost on the page with reading on the screen. In what way has the screen dimished the "sense of obstacle"? While a hypertext set of links to annotations might ease some of the problems of understanding, this is equally possible in a book with footnotes (cf. Landow's system for reading Milton, pp. 80-82); but such links are surely only a part of the solution to reading Milton, and a minor part at that. The poem cannot be dissolved into a set of links.

Birkerts suggests a different perspective, which cannot be comprised within an information processing model, the commitment of the self to the act of reading: "Reading is not on a continuum with the other bodily or cognitive acts. It instigates a shift, change of state . . . Reading a novel, then, is not simply a matter of making a connection to another person's expression. Over and above the linguistic connection, the process makes a change in the whole complex of the self. We are, for the duration of our reading different . . . the interior rhythms are modified in untold ways" (Birkerts, pp. 80-81). Among the effects of reading on the self, says Birkerts, "We do not learn so much from the novel itself, the lessons of its situations, as we do from having stayed free of our customary boundaries. On return, those boundaries seem more articulated, more our own; we understand their degree of permeability, and this is a vital kind of knowing" (Birkerts, p. 93).

In contrast to Bolter's constantly "active reader," Birkerts points to a phase of reading that he called its shadow life, which suggests the continuity of our engagement with it: "In some ways I am reading the novel as I walk, or nap, or drive to the store for milk. When I am away from the book it lives its shadow life, its afterlife, and that, as the believers have always insisted, is the only life that matters" (Birkerts, p. 95).

Hypertext is also made the occasion for attacking the supposed linear nature of book reading. "To read multiply is to resist the temptation to close off possible courses of action; it is to keep open multiple explanations for the same event or character. It is an almost impossible task for the reader to remain open in a medium as perfected as that of print" (Bolter, pp. 142-3). The reordering that changes the self of the reader is not at issue here: "The computer frees the writer from the now tired artifice of linear writing, but the price of this new freedom for the writer is that the writer must allow the reader to intervene in the writing space" (Bolter, p. 146). But note that in referring to "event or character" Bolter only addresses the level of plot. Does understanding of plot develop in linear fashion? And what about other interpretive processes that cut across plot, and which may deal with alternative perspectives during the course of reading? How linear is reading, really?

And it is worth noting that feeling is not considered by Bolter in relation to literature: it is identified mainly with television. Becoming weary we try to cut through the network of signs, says Bolter: "On the level of popular culture, this longing expresses itself as an obsession with 'feelings,' a denial of the value of semiotic thought." And this leads us to watch television (Bolter, p. 225). Bolter here cuts himself off from considering the contribution of feeling to reading.