Form, Medium and Function: the Rhetorics and Poetics of Text and Hypertext in Humanities Publishing

by Tim Bulkeley (Published in International Journal of the Book 1, 2003, 317-327)


Books are a highly evolved, convenient and aesthetically pleasing means of conveying ideas and information from authors to readers. In scholarship, a number of different genres of text have evolved to fill clearly defined roles. Some among these are well suited to print media. Others, however, contain information or ideas that "leap off the page".
This paper will explore some of the differences between two such genres: monograph (a book length text, arguing one or more theses) and commentary (a text explicating another prior text) and show why one is well suited to text presentation while features of the other - even in print medium – seems uneasily restricted by this medium.
A second section will explore some of the changes introduced by an electronic hypertext format and their impact on the rhetorics and poetics of commentary writing. This section will draw on the experience of generating such a hypertext and planning for a collaborative publication project to comment on the whole Bible in such a format.


Discussion of “the future of the book” has often either focussed on the possibilities offered by new technologies of writing, or on the concept of “book” as an idealised cultural icon. As such these discussions have often degenerated into tit-for-tat competing absolute claims.[1] So Annie Proulx absurd claim “no one is going to read a novel on a twitchy little screen. Ever.”[2] is often quoted; even though it focused on the contemporary limitations of the technology at the expense of envisaging even its then current possibilities. On the other side, Raymond Kurzweil’s claim that “the printed book, like any other technology, will not live forever”[3] has equally been too frequently cited, despite ample evidence for the continued survival and flourishing of the technology of print, not to mention older technologies like writing with ‘lead’ pencils.

Where more nuanced discussion takes place, and it has since at least the conference on the future of the book whose proceedings were published in 1996,[4] it has usually taken the form of extolling the advantages of print as a cultural icon, or the material advantages of the competing media, less attention has been paid to the different rhetorical effects of text and hypertext presentation or of the different needs of different genres of communication.

This paper argues that discussions of the future of the book should pay closer attention these differences. As Chartier argued we must take “into account the effects of meaning produced by the material aspects of writing”.[5] These meaning effects do indeed differ according to the features of the technology. Yet they differ even more between text (however presented) and hypertext. The question of the goals or purposes of the writing under consideration also needs to be taken into account. Different genres seek to achieve different ends, and a consideration of these will suggest whether a particular writing is more suited to text codex or on-screen hypertext.

So, unlike Duguid’s plea that we avoid a dualism of information and technology but consider information incarnate in technology,[6] my plea is that we not overlook the different rhetorical possibilities of differing forms of writing, quite aside from consideration of their technological containers. That is we need to separate, discussion of the effects of the technologies used for the transmission of text, from discussion of the differing rhetorics of text and hypertext.

This paper will address this issue of different genres of writing, in the light of the different rhetorics and poetics of text and hypertext, with respect to two common genres of scholarly writing in the humanities. While the examples are drawn from the field of biblical studies similar principles apply across a range of disciplines. And the rule that form should follow function applies to all writing.

Text and codex

As we begin this examination it is helpful to remind ourselves that “the book” is not a stable concept that has always pointed to one technology. The object that we think of as a book - often in contrast with other forms of written and printed material such as brochures, leaflets and letters - can more precisely be called a codex. These collections of leaves with writing on both faces, joined at one edge, were a huge technological advance in the early centuries of the Christian era.[7]

Traditional books were scrolls, long collections of leaves fixed edge to edge, usually written on only one face of the leaf. These rolls took up more storage space, were less easy to transport and only allowed sequential access. The technology of the codex developed from “notebooks” of several wooden “pages” either painted white for writing with ink or charcoal, or hollowed and filled with wax for inscribing with a stylus. The use of papyrus or parchment for the pages allowed much larger works to be collected this way. However, the codex was not at first seen as a “book” - that is as a medium appropriate for literary or scholarly communication - until it had won popularity as a means of collecting and transmitting Christian writings.[8]

Martial, a Roman poet of the late first century, mentions codices containing his own works, as well as those of better known authors such as Homer, Virgil, Cicero, Titus Livius and Ovid.[9] Since Martial refers to them as novelty gifts, Gamble calls them “inexpensive pocket-editions”.[10] Although by the second and third centuries the codex had become the standard medium for Christian writings, it was not until the late third and fourth centuries that it attained parity with the scroll for non-Christian literature.[11] The development thus traced is somewhat conjectural, the third century is a poorly documented period, however the statistic that 99% of texts surviving written in Greek in the second century are scrolls, while for the fifth century 90% are in codex form is striking.[12]

Compared with the scroll the codex was cheaper,[13] more portable, and easier to use; it allowed the collection of larger quantities of writing in one document; and permitted non sequential access.[14] It may have been the first advantages which encouraged the adoption of the codex; it was the last which had an enduring impact on the nature of writing. The ability to access material out of order, together with the possibility of longer documents, encouraged the adoption of systems of headings and subheadings. Many of the differences between writing and speaking, which we take for granted, are directly or indirectly the result of the impact of the codex on our writing habits, particularly of developments in the Middle Ages.[15]

This excursion into the history of book technology serves to remind us that the codex is not the book, and the book is not the codex. We must distinguish the bundle of poetic and rhetorical effects of a particular kind of text from the technology in which we imagine them incarnate. Yet we cannot ignore the effects of the technology in which the form of writing developed.

The codex and academic writing

Most genres of academic writing are now highly adapted to the technology of the print codex. Systems of headings and subheadings, page numbering and the presence of an index make these works easier to use. Different genres of academic writing have adapted in only slightly different ways to the medium. Considering the genres on which this paper focuses - the monograph and the commentary - elements common to all academic writing include: a table of contents, headings and subheadings, foot or endnotes, a sequential presentation of material, and one or more indices.[16]

Elements which differentiate their form are mainly confined to the more formalized organization of the commentary with an opening introduction situating the work commented, and then a series of sections providing different kinds of comment on sections of the work, together often with bibliographical material often separated according to the sections into which the text commented is divided. The commentary, as a text explicating another prior text, usually also has more cross-references than does the monograph.

Both the features of the broad genre of academic writing, and the distinguishing features of the two sub-genres have become sharper and more pronounced as time has past. Commentary increasingly includes sections divided according to the approach taken. The headings differ according to the specialist interest of the series so The Forms of Old Testament Literature (a form-critical commentary) includes the headings: text, structure, genre, setting, intention, and bibliography; while The NIV Application Commentary (a series aimed at preaching) has: original meaning, bridging contexts, and contemporary significance. So, in the twentieth century a variety of distinct styles and forms of commentary can be recognized.

Monographs and commentaries as text and hypertext

The genre monograph suits print well. A monograph is required to present a coherent argument presenting a particular case, or thesis, over an extended time. To do this there is little need for the reader to be required to jump backwards and forwards, the occasional footnote and even more occasional excursus is all that is needed by way of non-sequential access. In a monograph the non-textual elements – figures or images – are rare, and often only needed at one point in the text.

O’Donnell argues that

“the traditional monograph, with its sustained linear argument, its extraordinary high costs of publication and distribution, and its numerous inefficiencies of access only partly retrieved by the assignment of an LC call number, is beginning to look more and more like a great lumbering dinosaur, feeling a bit poorly and looking around for a place to lie down.” [17]

Yet his argument is based solely on the economic and access limitations of print, rather than a consideration of the rhetorical needs of the genre monograph and the different workings of text and hypertext. The future of the book looked different in 1993 from how it seems in 2003!

By contrast with the monograph, another common form of scholarly communication in the humanities, the commentary, has a different goal; its objective is to explain another text. Since different readers have different prior knowledge and interests, and since a variety of different approaches are used to any text, commentaries naturally contain a number of different sections all referring to the same section of text. Commentaries also contain frequent cross references to other sections of the work where the same issue has been treated. In the late twentieth century this is led to the genre commentary becoming increasingly ill at ease in the print medium.


Most definitions of hypertext derive from Ted Nelson's classic: "non-sequential writing, text that branches and allows choices to the reader ".[18] A wider definition of hypertext includes not only non-sequential writing, but writing that is comprised of a number of textual elements (or "lexia"[19]) and requires jumps from one lexia to another, whatever the technology that enables it.[20]

Hypertext and the Rabbinic Bibles

Since commentary is text referring to another text, its very nature is hypertextual. The early Jewish printed editions of the Bible known as Rabbinic Bibles (or in Hebrew as Mikra'ot Gedolot “Great Bibles”) illustrate this. The first of these Great Bibles was prepared by Felix Pratensis and printed by Daniel Bomberg in 1516-17. It used large format pages to offer:

   ●    Hebrew text

   ●    variant readings from the manuscript evidence available (often handled in footnotes in modern print commentaries)

   ●    Masoretic notes on the text designed to assist scribes in accurate copying, but also offering information on issues such as rare word usage

   ●    Aramaic translations or Targums of the text

   ●    commentary by various authors surround the biblical text commented

Figure 3

Torah page from Interactive Rabbinic Bible

On the Torah page shown,[21] the biblical text appears at the top right of the double page spread; the Masoretic notes are associated with the text.  To the left of the text across the double spread are a band of Aramaic translations (Targumim) with the Toledot Aharon - a system of cross-references to other Jewish writings mainly the Babylonian Talmud (but occasionally other well know material).

Below these are commentaries, including those by Rashi (Rabbi Soloman ben Isaac was a particularly authoritative interpreter from the second half of the eleventh century, with a strong commitment to the plain meaning of the text); Ramban (Rabbi Moses ben Nahman, Nahmanides in Latin guise) who often discusses both Rashi and Ibn Ezra's interpretations, which are themselves found on the left leaf and the Ba'al Ha-Turim to the Torah (the name is interesting as it refers to the author of these playful comments on the wording of the text by reference to his best known work - a Jewish law code in 4 columns ('arba'a turim).

Recent print commentaries

However, features of print technology have usually operated to restrict the hypertextual nature of commentaries. Watching a user read a printed commentary illustrates both of these tendencies. Fingers and eyes move backward and forward between text and comment, between the material dealing with a section of text to one dealing with a particular word or sentence.

Recent commentaries often present different sorts of material in sections marked off typographically one from the other. One needs to refer back to a collection of material about the book as a whole, separated from the comment on particular passages, in a section labelled "introduction". Other material treating some question in depth may be relegated to an "excursus". One recent commentary series even advertises itself as "as close to multimedia as you can get in print" while claiming to be "a new paradigm in Bible commentaries"[22].

I have attempted briefly to show that the art of commentary is inherently hypertextual, that the nature of commentary is hypertext - at the very least in the interrelation of text and comment.  Beyond this basic hypertextuality, however, engagement with a tradition of comment produces its own latent hypertext - seen in the copious footnotes and bibliographical information in most modern commentaries, or in the inclusion of substantial quotations from previous commentators in works such as Yitzhak Broch's Koheleth.[23] 

Given the nature and history of biblical commentary it is ironic that current offerings in the field are either:

   ●    determinedly "modern" in their attempt to present a coherent body of text (Old Testament Library or Interpretation)

   ●    or, at the other extreme, like most of the Hermenia volumes, and even more obviously the new Smyth & Helwys Bible Commentary series, straining the seams of the old media.[24] 

The Smyth & Helwys series actually offers a CD-Rom pdf version of the text to allow easier navigation of its links. This together with the slogans "as close to multimedia as you can get in print" and "a new paradigm in Bible commentaries" suggest that this publisher believes the future of commentaries will be even more hypertextual than their past.

Hypertext rhetorics and poetics

Rhetorical effects of hypertext presentation

Despite what was said above about the possibility of a print codex operating as a hypertext, in this section of the paper we shall assume the hypertext we’re speaking of is an electronic edition presented on screen. If we are to consider how the rhetorics[25] and poetics[26] of writing an electronic hypertext differ from those of a print text we need to ask basic questions about how people read.

For readers familiar with print culture text seems stable. As Bazin notes:

“The content of a book gives the illusion of being relatively independent of the mode in which it is read, because this mode stays quite stable and stereotyped, after the image of the text’s structure.”[27]

In most respects identical words from two different printings of a classic work can be, for most practical purposes, considered “the same text”. However, we should not allow the visual similarity of text in one lexia of an electronic hypertext to text (possibly with “identical” words) in a print edition to mislead us into assuming that the act of reading the two texts is also similar. In fact, though containing the same words and punctuation, the two texts may be read in strikingly different ways.

Most students assume that to “read a book” is to absorb its contents sequentially, they need to be taught to approach print books in other ways. Without such teaching many take a book from the library and then read it cover to cover. Teachers encourage them rather, first to survey the territory - looking at table of contents, index, the introduction, beginnings and endings of chapters; and only then to explore parts of the work in detail. The fact that adults in tertiary education need to be taught to read in such a non-sequential way is a vivid illustration of the power of the medium to promote sequential access. Thus a particular paragraph read from a book will most often be read “from start to finish”.

Landow claims that: “Although conventional reading habits apply within each lexia, once one leaves the shadowy bounds of any text unit, new rules and new experience apply.”[28] This does not go far enough, it is not merely when one moves to a new lexia that hypertext is read differently from text. That hypertexts contain “links” permitting, and encouraging, “jumping” predisposes readers to scan rather than to read sequentially. Empirical studies suggest that readers of a hypertext actively look for links, searching the text for routes to another lexia! Current limitations of screen technology may reinforce this tendency, but the nature of hypertext is an even more basic driver of this very different behaviour.

In theory an electronic hypertext can contain lexia of any length, and in fact some hypertexts created from existing print works do contain extremely long lexia. The electronic edition of the Anchor Bible Dictionary contains as one lexia the equivalent of over 7,000 pages of print text! Readers report that these extremely long lexia are confusing and difficult to use. Putting the text of different dictionary articles together in one lexia means that users seeking to look at the conclusion, or even a different section of an article can easily “overshoot” and get lost.

The possibility of links, and the fact that readers of hypertexts scan, mean that small lexia work better. The ideal is at most one screenful of text dealing with a tightly focused topic with key information or ideas highlighted (often as links).

Poetics of hypertext

A different technology produces different rhetorical effects, and so requires a different poetics.

The poetics of hypertext, with its rhetoric of interlinked small units of tightly focused text designed for scanning would be a disaster for the monograph. If you cannot predict which lexia a user may see, let alone the order in which they will be presented, a coherent argument leading to a particular conclusion or thesis becomes very difficult to achieve. Some websites severely limit hypertext functions to achieve such a goal. Links within a page are limited to explanations (often themselves either popups or pages with only a return link) and previous, next, home type links at the top and bottom of the document. Most such sites would be more convenient and pleasant to use as print documents, only the economics and accessibility of electronic publication have determined their medium.

By contrast, as we saw above, other genres of academic writing such as the commentary are by their nature hypertextual. They constantly refer to other works or to other sections of the same work. Different readers each seek different things. Such works do not have one thesis to prove, but rather should allow readers to construct their own meaning from information provided. These genres will achieve their goals better in a true hypertext format.

Commentary seeks to explicate a prior text. Each reader comes to the commentary with different prior knowledge and experience of the text being explained. Thus each needs a different collection of information. Hypertext links allow this. At various points pictures, diagrams or explanations of detail will be useful – links to them can be provided – their “presence” may not be desirable at each location because some readers arriving at that lexia will have already seen them.

Hypertext Bible commentary

Since 1996 I have been investigating how a hypertext commentary might function. This project has focused on practical rather than theoretical issues, since I chose to approach the task by producing a commentary on the book of Amos.

Amos - Postmodern Bible

Certain features of the current Amos commentary ( were determined by the rhetoric of the genre commentary, others are made possible by the additional rhetoric of the medium electronic hypertext.

There are two main windows, one for the text commented, and the other for explanation and comment – thus the relationship of comment to text is visually represented.[29] As well as links from text to comment, other sorts of material are provided. Words and phrases within the comment are explained, people and places are described – this material is like that found in a Bible Dictionary. Technical terms are defined and explained – rather like material in a Dictionary of Biblical Studies. Because the author cannot predict what parts of this material readers will want to see at particular points in their reading, access to this dictionary/encyclopaedia material is provided through a permanent menu system.

This encyclopaedia material already suggests ways in which the contents of the hypertext commentary move beyond that of its print equivalents. Links from the comment to explanation of ideas and terms makes the discussion accessible to a wider range of users. Thus, although the intended level of the comment is similar to that of most print series (students and professionals in the discipline) among the users who have written with comments about half describe themselves as “lay people” with no tertiary training in the discipline. They are able to understand the discussion because of these explanations. On the other hand a significant fraction of users are professional biblical scholars, they are able to access further discussion of the reasons or evidence behind what is said through other links.

The rhetorics of hypertext (as discussed above) makes it difficult to present an argument for a particular case, but easier to present information from which readers draw their own conclusions. This feature of the medium has also been exploited in the composition of the commentary. Traditional print commentaries are aimed at segments of the wider commentary audience, some aim at particular styles of scholarship (historical, form critical, literary etc.) other commentary series are more application focused and may be aimed at particular ecclesial communities (Catholic, Evangelical etc.). Evidence from user feedback suggests that the Amos - Postmodern Bible commentary is used by Jews, Catholics, Evangelicals, mainstream Protestants and non-religious readers.

One common reaction by users is present in a recent email:

The Postmodern Bible is a breakthrough in 21st century bible studies method.  It appear (sic) to be able to incorporate the scripture (written and spoken) with bible commentary, pictures, and modern day linkages that help to bring the scriptures to life.[30] 

For writers and readers, the interesting conflict is not between print and electronic forms of publication – that is an issue of convenience, preference and economics. It is between text, largely uni-sequential and usually single; and the deliberately pluri-sequential multiple linked texts that we call hypertext. This conflict is likely to be decided in different ways for different genres (at least as far as scholarly publishing is concerned). Some genres of academic writing require the orderly progression that print technology makes pleasant to use, but others contain information and ideas that “leap off the page” and are better presented in hypertext formats that electronic technology made convenient.

[1]  Compare Haas’ analysis of the arguments used to support different attitudes to computer technology in the humanities. Christina Haas, Writing Technology: Studies on the materiality of literacy. (Mahwah, NJ: Lawrence Erlbaum Associates, 1996) ch.7.

[2]  Annie Proulx. "Books on top." New York Times (May 26 1994): A23.

[3]  Raymond Kurzweil. “The Future of Libraries.” in CyberReader, ed. V. J. Vitanza. (Boston: Allyn and Bacon, 1999): 291.

[4]  In several of the papers in Geoffrey Nunberg. The Future of the Book. (Berkeley: University of California Press, 1996) perhaps most comprehensively Paul Duguid “Material Matters: the past and futureology of the book”, 61-101.

[5]  Roger Chartier. Forms and meanings : texts, performances, and audiences from codex to computer. (Philadelphia: University of Pennsylvania Press, 1995): 5.

[6]  Paul Duguid “Material Matters: the past and futurology of the book” in Geoffrey Nunberg. The Future of the Book. (Berkeley: University of California Press, 1996): 77ff..

[7]  The standard treatment of the origin of the codex is Colin H. Roberts and T.C. Skeat, The Birth of the Codex (London: Oxford University Press, 1983)

[8] This development is summarised by the standard text on the transmission of classical literature L.D. Reynolds and N.G. Wilson. Scribes and Scholars: a guide to the transmission of Greek and Latin Literature. (Oxford: OUP, 1991 [3rd edition]): 34-35.

[9] M. Valerius Martialis (ed. William L. Carey). Apophoreta (Epigrammaton liber XIV): clxxxiv, clxxxvi, clxxxviii, cxc, cxcii [downloaded 12 March 2003].

[10] Harry Y. Gamble. “The Codex.” in The Anchor Bible Dictionary. ed. David Noel Freedman. (New York: Doubleday, 1992): Vol. 1, 1068; M. Valerius Martialis (ed. William L. Carey) Apophoreta (Epigrammaton liber I) [downloaded 12 March 2003].

[11]  Reynolds and Wilson, 34-35; Gamble, 1068.

[12] Roberts and Skeat, 37.

[13] T.C. Skeat. “The length of the standard papyrus roll and the cost advantage of the codex.” Zeitschrift für Papyrologie und Epigraphik. 45, (1982): 169-175.

[14]  James J. O’Donnell “St. Augustine to NREN: the tree of knowledge and how it grows” [downloaded 16 April 2003].

[15]  On these see H.-J. Martin and J. Vezin, edd., Mise en page et mise en texte du livre manuscrit (Paris: Iditions du Cercle de la Librairie-Promodis, 1990); O’Donnell discusses particularly the example of Cassiodorus’ means of facilitating non-sequential access to his commentary on Psalms, and the nature of gospel parallels as hypertext links.

[16] Many of these elements form a sort of rudimentary hypertext, see G.P. Landow and P. Delany. "Hypertext, hypermedia and literary studies: the state of the art." in Hypermedia and Literary Studies. eds. P. Delany and G. P. Landow. (Cambridge: MIT Press, 1991): 4.

[17]  O’Donnell.

[18] Theodore Holm Nelson. Literary Machines 93.1. (Sausalito, CA: Mindful Press, 1992): 0/2.

[19] The term is adapted from Roland Barthes, and has been commonly used to speak of the textual units within a hypertext- following George Landow. Hypertext in Hypertext. (Baltimore: Johns Hopkins UP, 1993): 52. If lexia is used for the information unit to or from which a link is made “node” would then refer to the exact location of the link within that lexia. Thus a lexia might contain several (both incoming and outgoing) nodes.

[20]  Such a hypertext is not non-sequential, rather it is ‘plurisequential’ for each reader chooses their own sequence. See Luca Toschi. “Hypertext and Authorship” in Geoffrey Nunberg. The Future of the Book. (Berkeley: University of California Press, 1996):203.

[21] From Eliezer Segal. Interactive Rabbinic Bible. downloaded 2 August 2001.

[22] Advertising material for the Smith & Helwys Commentary series downloaded from: on 23 March 2005.

[23] Yitzhak I. Broch. Koheleth: The Book of Ecclesiastes in Hebrew and English with a Midrashic commentary. (Jerusalem/New York: Feldheim, 1982).

[24]  See e.g. Walter Brueggemann 1 & 2 Kings (Macon: Smyth & Helwys, 2001): 17-19.

[25]  Rhetorics is understood in this context as the study of how the functioning of a medium makes certain forms of argument or expression more or less possible, easy or difficult.

[26]  Understood as the study of the principles behind the actual expressions of a particular genre.

[27] Patrick Bazin. “Towards Metareading” in Geoffrey Nunberg. The Future of the Book. (Berkeley: University of California Press, 1996): 165.

[28] George P. Landow. Hypertext 2.0: The Convergence of Contemporary Critical Theory and Technology. (Baltimore: The Johns Hopkins University Press, 1997): 4.

[29]  In contrast to IBM’s 1990 showcase hypertext commentary on Tennyson’s “Ulysses” which replaced the text of the poem by other (often video) material, discussed by Jay David Bolter. “Ekphrasis, virtual reality, and the future of writing” in in Geoffrey Nunberg. The Future of the Book. (Berkeley: University of California Press, 1996): 262

[30]  Response form from by David Watkins, Wednesday, April 16, 2003 at 14:36:15