Reading, writhing and arithmetic

In The Uncommon Reader, the literary theorist George Steiner described how the nature of reading has changed over the past few centuries. He illustrated this by examining the depiction of reading in an 18th-century painting, "Le Philosophe Lisant" by Chardin. This painting depicts a handsomely dressed scholar studying the pages of a thick folio with ink and quill at hand. Steiner argued that the hourglass, the quill, and even the scholar's clothes are symbols that the scholar is not passively accepting the text, but is actively learning from it. Steiner contrasted this kind of engrossed, absorbed reading with that done with a paperback. Here, a reader passively "listens" to the author of the book.

Steiner argued that this is, of course, precisely the point of paperbacks: to entertain, to distract readers from hard thought, and to provide leisure. But he went on to suggest that the distance between what he called "scholarly" and '`lay" reading was becoming increasingly small in the modern world. He remarked that not only was less value being placed on deep, intellectual reading, but any attempt at such reading is interrupted by modern technology - concentration is undermined by telephones ringing, peace is disturbed by the sound of television and stereos. Steiner lamented: the art of reading is being lost.

In the early 1990s, some ten years after Steiner wrote this, others began to argue that new technology was doing the opposite to reading: instead of preventing or disrupting it, digital technology was allowing new forms of reading. Indeed, some heralded the beginnings of a new document age. What excited these individuals was the advent of hypertext - the electronic linking of one text to another. In his book Writing Space, Jay Bolter suggested that document technologies had been through a number of phases or eras. The first was the technology of the papyrus roll, the second the medieval codex, the third the printed book, and the fourth, the era of hypertext. Hypertext was likely to be the most significant of these phases, he claimed, because it has the power to alter the relationship between the reader and the text. According to Bolter, the author's intent and meaning is obliterated by the freewheeling interpretive skill of the reader as he or she constructs new (hyper)texts out of the fabric of the author's own. Thus Barthes' methodological dictum "Death to the author" was to be replaced by a technological fait accompli.

The claims of these authors were given much attention, and Steiner's lament was replaced by excitement. New technology was reinvigorating and even reinventing reading, not undermining it. Consequently, people from a variety of disciplines began to examine reading much more seriously than hitherto. Sociologists, for example, became obsessed with the written word. Indeed if for ancient Israelites, God was in the Word, for sociologists of the early 1990s, Society was in the Text. Most of this research, exemplified by Dorothy Smith's The Active Text, showed how the process of reading instantiated hierarchy and gender differences in society.

Meanwhile, in the fields of psychology and human factors engineering, the concern was for more mechanistic ways in which computers were altering reading. In the 1980s, the obsession was to explain why reading from a computer screen was slower than reading from paper. As screen technologies began to improve, these differences in reading speed began to disappear. What did not disappear, however, was people's steadfast preference for reading from paper.

Explanations emerged from new lines of research by psychologists such as Christina Haas and Kenton O'Hara. Some of this work has explored how people read in real work settings, leading to a radical rethinking of what reading really involves.

One finding is that in organisational contexts, reading occurs more frequently in conjunction with writing than it does in isolation. Thus, writing (in a variety of forms) has to be recognised as an integral part of work-related reading.

Another finding is that almost half of all reading in work settings involves the concurrent use of more than one document. Finally, this research shows that linear, continuous reading is an unrealistic characterisation of how people read.

In fact, reading in the workplace involves much more complex patterns of navigation. Readers skip from a conclusion to earlier parts of a text and back again, often glancing and cross-referring to other texts at the same time. They do all this as they annotate or concurrently write new texts. In essence, this research shows that reading has hypertextual properties avant le lettre.

This kind of research gives guidance in the design of new digital reading devices. They need to support rich, stylus-based marking and annotation. They need to be lightweight and portable to ensure they can be placed at the correct ergonomic angle for reading or writing, and so that they can be juxtaposed with paper documents or other digital displays. Flexible navigational tools are required. Two-handed input, for instance, could allow a reader to use one hand to mark a place in a document while scrolling or page turning with the other. This would also allow concurrent activities such as writing on one document while navigating through another. New modes of feedback such as audio or even tactile feedback could provide more implicit cues as to where in a document a reader is, and how long a document is.

The use of perspective and stereoscopic displays may offer a "third" dimension to better represent features such as document thickness. Finally, and perhaps most important, these new devices need to support the various ways in which people read and write across multiple documents, which suggests that multi-screen systems are the way forward.

All of this points to ways in which document viewers or portable document readers might be designed for business or educational markets. One irony of this is that the form of reading that Steiner believed was withering away - scholarly reading - may ultimately be that which is supported by these new technologies.

Like 18th-century scholars, readers will be able to mark up and annotate their texts, navigate through them in complex ways, and thereby engage themselves to the text in hand. Developing this technology will take some time. Meanwhile, electronic books are hitting the market, aimed primarily at supporting that kind of reading that Steiner so dismissed: the lay reading of paperbacks. Whether these products will succeed in this is an open question since very little research has been done into this type of reading.

The proponents of these technologies - Microsoft among them - argue that commercial success will follow from agreement to ensure that all ebooks are delivered in a standard format enabling them to be read on any device. This discounts any question of how readers interact with the texts in question.

Perhaps they will be entirely passive - simply reading, absorbing, moving between pages in strict sequence. But perhaps readers will not be so passive and the distance between scholarly reading and lay reading may not be so great.

We shall have to wait and see.

Richard Harper is director of the Digital World Research Centre, University of Surrey. Abigail Sellen designs new technologies at Hewlett Packard Laboratories, Bristol. They are co-authors of The Myth of the Paperless Office, to be published by MIT Press.