As we are now fully engaged in the digital culture, it is quite appropriate to examine the unravelling of the codex and its consequences. I shall focus on the page, which is the main building block of the codex, its DNA, its soul. Across the centuries, the changes made to the page have affected the appearance, the functionalities and the usability of the codex. After following some of these transformations across history, this article will explore the future of the page in digital format.
The definition of a page is quite straightforward: a page is one side of a leaf of paper or parchment or any other material that is able to receive a text or illustrations and be part of a set. The word “page” comes from the Latin word pagina, which designates a column of writing. Pagina itself is derived from the verb “pangere”, meaning “to mark out the boundaries”, which is essentially the function of the page, because “writing necessitates the marks of its own limits” (Zali 23). The Latin verb also means “to plant vines in a vineyard” ¾ which has given rise to the vineyard metaphor of the text, first explicitly made by Pliny the Elder, and quite frequent in the Middle Ages, inspiring Ivan Illich’s book on Hugh of St. Victor, In the Vineyard of the Text (1993).
The vine metaphor is particularly adequate because, in a papyrus scroll (Fig. 1), the pagina corresponds to the column of text. In most scrolls, the column width varies from six to eight cm. in length and holds about 25 to 45 lines, with an intercolumn margin of 2 to 2.5 cm. (Mak 2011:11). These proportions “remain remarkably stable” (Johnson 2004: 55). It should be noted that, contrary to a common misconception, the volumen unrolls horizontally, unlike the vertical standing rotulus, which served essentially for public proclamations in theatrical, liturgical or administrative settings and where the page encompasses the total length of the roll.
The page changed drastically and took on its present meaning around the first century, with the advent of the codex in the Roman Empire. The codex was first a gathering of thin wooden tablets, bound by strings and coated with coloured wax (Fig. 2).
Being easy to erase, these tablets were used for drafts or school exercises. (Zali 37; Mak 13). In the first century, wood was replaced by papyrus and then by parchment. With parchment, it was possible to use both sides of the page. The folding of the sheet in two or in four and the subsequent binding of the sheets around the first century of our era produced the codexThe page thus became an autonomous unit as well as part of a totality. In this new format, the book gained in density and was much easier to handle than the scroll. Moreover, it freed the reader’s hands. As the codex can stay open on any particular pairing pages (called an opening), the text gains new possibilities for conveying visual information. As a result, “the readable gradually moves into the realm of the visible.” (Vandendorpe 2009: 30). This shift accentuated the tension between two antagonistic yet complementary sources of absorbing information: the acoustic-verbal dimension of language and the visibility of text.
As far back as one can go, the notion of “a page” implies a space carefully demarcated for inscribing and reading. As such, it has been defined as “a temple made of writing where reading will be practised.” (Zali 23). Through the course of centuries, scribes, illuminators, typographers, printers and publishers perfected the various components of the page, in order to better seduce the eventual reader and entice her to delve into the text.
The richness of the visual dimension is the main characteristic of the illuminated manuscripts produced during the medieval era. Scriptoria established in monasteries developed a very efficient technique for the production of illuminated codices, beginning with the Vatican Vergil (ca. 420) and followed by thousands of bibles, missals, or psalters designed to reinforce the appeal of the gospel by their sheer beauty, like the Lindisfarne Gospels (Fig. 3) or the Book of Kells (ca. 800). In these books, illustration takes precedence over the text. Thus, in the chain of book production, one of the most important functions pertains to the rubricator who adds illustrations, notably incipits, ornamental initials or passages written in red ink in order to draw them to the attention of the reader. The text itself is part of the illustration and must obey its rule; this subordinate status of the text can be seen for example in the filling in of lines, whose purpose is to transform the whole page into a kind of beautiful picture.
Around the 13th century, the organization of the page gained in complexity. As an answer to the needs of scholars requiring clarity and coherence, scholarly editions carefully rethought the optimal ordinatio of the text, organizing the original text and its glosses on the same page, while making sure to establish a clear distinction between them. Initials at the beginning of verses or of important sections were frequently coloured. In a column of text, a pilcrow indicated the beginning of a new development. Running titles became a current practice. Moreover, thirteenth-century scribes “introduced the analytic table of contents as a guide to the ordinatio and to facilitate the readers’ access to component parts of a work” (Parkes 122). The visual appeal of the layout was not forgotten, however: during the same century, the French architect Villard de Honnecourt divised the ideal layout of the double page of the book, giving to the various margins specific proportions based on the golden section (Fig. 4). This layout is still relevant today. As the great Canadian typographer Robert Bringhurst has written: “To give the reader a sense of direction, and the page a sense of liveliness and poise, it is necessary to break this inexorable sameness and find a new balance of another kind. Some space must be narrow so that other space may be wide, and some space emptied so that other space may be filled” (Bringhurst 163). Thanks to these developments from the 13th century on, the page gained in autonomy at the expense of the linearity of the book (Demarcq 66), signalling the beginning of a new era. As noted by Parkes, “the late medieval book differs more from its early medieval predecessors than it does from the printed books of our own day” (Parkes 135). We can see this typical layout in an incunabulum of Aquinas, published in 1477 (Fig. 5).
Written by Domenico Nani Mirabelli in 1503, the Polyanthea is an anthology of citations classified in about one thousand articles on a variety of moral and theological subjects, as well as others of general interest: friendship, marriage, old age, grammar, memory, war, health, the zodiac and so on… For example, the article on marriage is divided into twelve questions ¾ “Should someone marry or not? At what age? Is beauty an important consideration?”, etc. ¾ and offers for most topics a series of citations carefully chosen from a variety of authors and revered sources. “The early Polyanthea served in part as a dictionary of hard words, offering in addition to the major articles, many very short ones, with just a definition, a Greek etymology, and one or even no quotation as an example” (Blair 178).
One of the first general reference works produced for the printed book market, it was to be found in every great library. The British Museum has ten copies from different editions and one of them was owned by King Henry the Eighth (Blair 182).
This best seller was published in many places (Lyon, Paris, Venice, Cologne, Frankurt, Basel…) and had at least 44 editions between 1503 and 1681. Interestingly, the book gradually grew in content, since some publishers made important additions: thus it went from 430,000 words in 1503 to more than 2.5 million by 1619 (Blair 125). Clearly, according to the old etymology, the author was also the “auctor”, term derived from the Latin verb augere, meaning “to grow, to augment”. This is not dissimilar from the remarkable expansion of Wikipedia, as we shall discuss later.
As such, the Polyanthea is a particularly interesting tool for following the evolution of ideas and printing across two centuries. In the first edition, published in Savona (Italy), the title page contains only the title, disposed in the form of a cul-de-lampe (Fig. 6). This disposition may look weird to us, especially since words are hyphenated without necessity: “Auditory dominance can be seen strikingly in such things as early printed title pages, which often seem to us crazily erratic in their inattention to visual word units. Sixteenth century title pages very commonly divide even major words, including the author’s name, with hyphens, presenting the first part of a word in one line and the latter part in smaller type” (Ong 118). At that time, reading did not focus on the visual aspect of the words grasped globally, but was still based on oral practices: the presentation of the text was independent of its semantic aspect.
Just after the preface, there are no less than sixteen pages of table of contents, listing all the entries of the dictionary. For us, it may seem strange to have an alphabetical table for a dictionary, but it was a sure sign of modernity and reader-friendly publishing. It was not new, however, as indexing had become popular as of the beginning of the 13th century (Blair 36).
After the Index, we find the first page of content, which is also a kind of title page, since there is an image enhancing the figure of the author, Nani Mirabelli, shown in the company of very important people, notably the Pope (Fig. 7). The author was rector of schools and archpriest of the cathedral in Savona. He also served as papal secretary.
Next, we find the first entry in the florilegium, which is the word « Abstinence », illustrated by almost three columns of quotes from such authorities as Augustine, Ambrose, Seneca, Aristotle, Cicero, and many others. At the beginning of the article, a full page is dedicated to an analytical table of the various meanings of the word “abstinence” (Fig. 8). This table is called divisio in Latin, a term that Ann Blair translates as a “branching diagram”. Made of “squiggly bracklets”, the divisio was already common in medieval sermon manuals (Blair 145). In Polyanthea, the diagrams take many of their divisions from Aquinas’s Summa. The medieval French scholastic philosopher Peter Abelard (1079–1142), as well as Hugh of St. Victor and Ramon Lull in the 13th century, underlined the importance of this device since it was supposed to help the reader to memorise. In the medieval era, reading itself was a way to expand one’s memory (Illich 35). In the colophon, Nani proudly mentions the presence of these analytical tables as an aid to memory: “[You have] some material ramified in trees so that you commit it more easily to the chest of memory”.
Interestingly, the analytical tables no longer appear in the 1600 edition, made in Lyon. It is indicative that memory suddenly no longer had the same paramount importance, thanks to the massive availability of printed books. In 1580, Montaigne had already said that he preferred a “Tête bien faite plutôt que tête bien pleine”. In 1600, we are decidedly in the modern era.
By examining the various editions of the Polyanthea, one can see how the printed book took its modern form, abandoning progressively characteristics of the manuscript, such as the colophon. Across the 16th century are introduced the title page, the printer’s mark, and pagination in Arabic numerals, and the Roman typeface is widely adopted instead of the Gothic, at least in France and Italy. Instead of the cluttered pages of the 16th century, the books printed around 1600 prefer a clean layout quite consistent with the ideal of regularity brought about by typography. Italics begin to be used for differentiating certain words, although it will only be in the 18th century, with the Biblioteca universale sacro-profano by Vincenzo Coronelli, that the use of italics for the titles of books will become a standard practice.
The layout of the book evolves in accordance with changes in reading goals and readers’ habits. By creating a closed space, the page facilitates the engagement of the reader with a text. Its uniform layout is particularly helpful for a prolonged immersion because it symbolically shuts out the external world. A stable layout of openings or paired pages is most efficient for summoning the activation of spatial memory and for a close study of long texts. Moreover, the book as a collection of pages offers the promise of a totality of meaning.
The ephemeral triumph of liquid text
With the advent of the Web, a new revolution, much more important than the invention of printing or even the transition from the scroll to the codex, is affecting the book. And it is developing very quickly. In digital format, text gains many new characteristics that generations of readers across the centuries did not even dream of, such as portability, ubiquity of access, hyperlinking, interactivity, as well as full and permanent indexation.
At the same time, the old codex has also found its digital avatar. Digitisation has made millions of books reachable by almost everybody from everywhere. With 25 million books digitized in October, 2015, Google Books is the extreme illustration of the will to avoid losing the books accumulated in great libraries, as was the case when the scroll was abandoned in favour of the codex. For some books, you can even turn the pages as if they were real –particularly in Archive.org and at the British Library.
Scholars interested in Old English know the Vercelli Bookan anthology of Old English prose and verse that dates back to the late 10th century. You can now browse all of the pages of this manuscript both in single or double page format. But the digital edition offers new possibilities that go well beyond the printed book: you can display a page of the manuscript at left and its diplomatic transcription at right, and you also have a magnifying glass to use on the manuscript, as well as a search engine. In sum, the codex, in digital format, is gaining a new life with new features that make research easier.
But text created in digital format has known a difficult start. We can trace the origin of digital text to Claude Shannon, in 1945, when he abstracted the idea of the message from its physical details and reduced text to a mere sequence of characters. In the model of communication he later published with Warren Weaver, he describes “noise” as “everything that corrupts the signal, predictably or unpredictably” and assigns to it a prominent place in his theory of information (Gleick, chapter 7).
Shannon’s mathematical theory of information (Fig. 9) allowed the processing of text by computers and gave text the fluidity necessary to adapt to any device. But this fluidity was gained at the expense of the visual components of the page, which were reduced to “noise” and thus seen as quite expendable. When the World Wide Web Consortium set the standard for the Web, it recommended that text in HTML be displayed running from one side of the browser window to the other, filling the screen: the size of the monitor or of the browser were, by default, the true container of text.
By reducing text to a stream of bits, the engineers dismissed the tradition of the page as a semantic space designed to ensure a maximum of readability. Commenting on this situation, an expert in the history of written language remarked: “With the computer, we risk letting ourselves be convinced that all writing can be reduced to a pattern of black dots on a screen” (Harris, 141; translation). In the first years of the Web, this “de-mediation” of the text was seen as a positive step forward, and a sure method to break away from the printed book. The page was supposed to be superseded by hyperlinks that made the old designs obsolete. In the nineties, most experts thus avoided the word “page” and preferred to use terms like spaces (Moulthrop), lexias (Landow), textons (Aarseth) or screens (Vandendorpe 2009: 137). Even today, in Microsoft Word, the Web Layout fills up the window of text from left to right without any margins.
As research on the mechanics of reading has made abundantly clear, the reader’s eye does not follow a continuous linear path but proceeds by saccades and fixations of about a quarter of a second, during which an average of 7-9 characters are acquired (see Wikipedia, “Eye movement in reading”). The longer the line, the more frequently a saccade may jump to another line by mistake. Faced with very long lines of text, readers react by speeding up their eye movements and will sometimes be content with skimming the text, focusing only on the beginnings of the lines and on a few words here and there. This is why the column of text, from the oldest scrolls to the modern tabloids, rarely exceeds 80 characters per line. The shorter the line, the more appealing it will be for the common reader.
As monitors became larger, a full screen web page could easily count hundreds of characters, which is not ideal for reading. When I pointed to this problem, I sometimes got the answer that I could easily resize the page by resizing the browser. But, when you do that, the tabs at the top of the browser are reduced to almost nothing: it is decidedly not user-friendly to force readers to constantly resize their browser window according to the sites they are visiting.
It took several years to publishers to realize that they could not decently display books and articles in the absence of limits, and that the white margins surrounding the text are not useless space to be dispensed with; on the contrary, they help to recreate the page as a container and delimit the text, allowing the eye to rest from the tension caused by reading.
It is interesting to study how the layout of the page has evolved in newspapers over the last twenty years. This can be done thanks to the Wayback machine at Archive.org, where one can walk into our past. In this screen copy of the Globe and Mail, dated September 16, 2001, for example, the text fills a central column with no fixed length (Fig. 10).
Two years later, the design has changed. The page is enclosed in a box with a fixed length (Fig. 11).
By necessity, editors of newspapers and magazines became aware of the physiology of reading and decided to return to the layout of the printed page, with a column of limited width, and white space in the margins. Instead of being displayed at the full width of the browser, text is now generally organized into a fixed-width column of 60 to 90 characters per line, and the box has disappeared.
This is the layout we find today in most newspapers. It has also been adopted by the main platforms for bloggers. Similarly, Google, Yahoo and MSN display search results in columns of about 1024 pixels of total width and 800 pixels for text. Most learned journals have also adopted a fixed-width column counting about 75 characters, not very different from the printed page of the codex, or the column of the papyrus.
In the case of the scholarly journal Terrains/Théories the editors have chosen to display the footnotes in the margin at right, just like the Polyanthea did five centuries ago (Fig. 12). This layout is certainly more sophisticated and more user-friendly than footnotes positioned at the bottom of the page. Moreover, the numbering of paragraphs will help readers’ interactions about this article. Looking at these numerous examples, one can now say, without any doubt, that the page has found its new avatars in the digital culture.
Wikipedia : a special case
Then again, maybe I am too optimistic. Today, in a sense, the true spirit that inspired the Polyanthea is probably best represented by Wikipedia, whose strengths go much further than print encyclopaedias thanks to hyperlinks and extensive categorisation. But, Wikipedia’s layout did not evolve as other platforms did. One of the reasons may be that it was born in 2001, when engineers were still in exclusive command of the web look and feel. And for a programmer, the building block of text is not the page but the line of code: the wider the window, the easier it is to revise a program and look for a bug. If we compare a Wikipedia page from November 2002 (Fig. 13) and one from May 2006 (Fig. 14), we can see that there has been some improvement in the readability, thanks to the introduction of pictures and subdivisions, as well as a table of contents. But, since then, the design hasn’t changed much, except for the introduction of an Infobox.
A newcomer arriving on Wikipedia in 2016 may be struck by the fact that the text still occupies the full length of the window, except for the left margin. This radical departure from the page format adopted by most media today tends to confirm the status of Wikipedia as a work-in-progress.
One could say, however, that with a long line, the eye is able to scan rapidly through an article, and this is typically what one does with an encyclopaedia: we rarely read an article from beginning to end, but rather scan it in order to find a specific information. However, this is also a kind of self-fulfilling prophecy since this layout discourages the user from reading the article, and tends to foster superficial browsing rather than close reading. This is hardly ideal for an enterprise fostering the rise of knowledge.
Nevertheless, in order to help readers, Wikipedia offers the possibility of creating a kind of book with one or various articles, by clicking on the link “Create a book” in the left margin. The choices of layout offered are very limited, however: you can choose between the paper size “letter” or its European equivalent. The letter format is fine for printing, but who still prints articles today? Wikisource offers a wider choice of layouts and allows you to download a book in various formats: PDF, EPUB, MOBI…
On tablets and mobile applications, Wikipedia articles are designed in a limited-width format. If you use the mobile url on a laptop, you will access articles in a limited-width format also, as one can see with the article on William Blake (Fig. 15). Wikipedia is also developing a skin that could be applied to existing pages and would reformat all the articles in this new layout.
The smartphone today is the platform of choice for more and more people. As we can see on a page from Middlemarch by George Eliot (1819-1880), the column of text displayed on an iPhone Plus counts about forty characters (Fig. 16), but the user could choose to make the character bigger or smaller. The author of this long novel certainly could not have anticipated that she would be read one day on mobile devices in the most unexpected situations, since many readers today turn to their phone to fill any period of waiting. Thanks to mobile reading, one can say that the page has found a new digital avatar.
After the page, the book
Even if the page, as a fixed-width space designed for reading, is now commonly accepted in its virtual reincarnation, there are still some protocols to implement before one can say that the book has been rescued from a widely anticipated death. As noted by Peter Brantley, in order for books to be fully replicated in the digital world, we must address the problems posed by annotation management, ensure perfect compatibility between the dedicated environment of the e-book reader and the web thanks to HTML5, and, most of all, ensure that the new protocol can “reproduce a core definitional aspect of the traditional book: its self-containment”. Both the W3C and the International Digital Publishing Forum (IDPF) are working closely toward that objective and the next e-book standard, due to be released at the end of 2016, “will marry itself fully to the Open Web Platform, and books, magazines, and pamphlets — online or off — will be the peers of any document on the web” (Brantley).
With this next step, a media that has been in use for two thousand years and produced so many illuminating landmarks for the human mind will continue to be alive in our culture, adapted to convey new inquiries into human nature and to shape the minds of future generations.
- Aarsth, Espen. 1997. Cybertext. Perspectives on ergodic literature, Baltimore and London, Johns Hopkins University Press.
- Baudin, Fernand. 1994. L’Effet Gutenberg. Paris : Éditions du Cercle de la librairie.
- Blair, Ann N. 2010. Too Much to Know. Managing Scholarly Information before the Modern Age. New Haven & London: Yale University Press.
- Bradley, Peter. 2016. “Books, in a browser”, https://medium.com/@naypinya/books-in-a-browser-375df76207ce#.qkd9tbisy
- Bringhurst, Robert. 2008. The Elements of Typographic Style. Vancouver : Hartley & Marks Publishers.
- Demarcq, Jacques. 1999. “L’espace de la page. Entre vide et plein, ” in L’aventure des écritures : la page. Edited by Anne Zali. Paris : Bibliothèque Nationale de France.
- Gleick, James. 2011. The Information : A History, a Theory, a Flood. New York : Random House.
- Harris, Roy. 1994. La sémiologie de l’écriture. Paris : CNRS.
- Illich, Ivan. 1993. In the Vineyard of the Text. A Commentary to Hugh’s Didascalicon. Chicago. The University of Chicago Press.
- Johnson, William A. 2004. Bookrolls and scribes in Oxyrinchus, Toronto : University of Toronto Press.
- Landow, George P. 1992. Hypertext: the convergence of contemporary critical theory and technology. Baltimore and London: Johns Hopkins University Press.
- Laufer, Roger. 1982. “L’espace visuel du livre ancien,” in Histoire de l’édition française. Edited by Henri-Jean Martin, Roger Chartier and Jean-Pierre Vivet, vol. 1. Paris: Promodis.
- Mak, Bonnie. 2011. How the Page Matters. Toronto: University of Toronto Press.
- Ong, Walter. 1982. Orality and Literacy: The Technologizing of the Word. New York: Routledge.
- Parkes, Malcolm Beckwith. 1976. “The Influence of the Concepts of Ordinatio and Compilatio on the Development of the Book,” in Medieval Learning and Literature. Edited by J.J.G. Alexander and M.T. Gibson. Oxford : Clarendon Press.
- Vandendorpe, Christian. 2009. From Papyrus to Hypertext. Toward the Universal Digital Library. Urbana and Chicago : University of Illinois Press.
- Zali, Anne. 1999. L’aventure des écritures : la page. Paris : Bibliothèque nationale de France.