Graeme Davison, Department of History, Monash University.
Years ago, when I used to teach a course on the history of the industrial revolution, we sometimes discussed what it must have been like to live through that era of cataclysmic technological change when steam engines, cotton mills and railways radically transformed patterns of work and play, relationships between city and country and even people's conceptions of space and time.  We tried to imagine ourselves into the mental world of the rick-burners, the Luddites and the handloom weavers, whose traditional ways of life had suddenly been rendered obsolete by the new machines.
We surveyed their fate through the Gothic windows of a university still recognisably patterned on it s medieval prototype, teaching students who sat in libraries and took notes with pen and paper. Only in the 1980s and 1990s has the industrial revolution finally arrived in academia. Now, as we contemplate the approach of the virtual university, we know something of the mingled excitement and anxiety that gripped the handworkers and artisans of the newly mechanised trades of the first industrial revolution.
We are in the midst of revolution in the ways in which knowledge is not only transmitted, bu t generated, packaged, absorbed and interrogated. Our libraries are committing an increasing proportion of their funds to electronic data-bases. Most of us are now tied by an invisible umbilical cord to our personal computers, which have become not just word processors but gateways to a world-wide web of information and communication. Monash's Faculty of Arts has just made a large bid for funds to enable many of our courses to be made available by 'flexible delivery'. Several of us are now involved in res earch projects, such as the ever-increasing numbers of Oxford Companions and encyclopedias, which have been made possible only by the ease with which electronic texts can be compiled, edited and translated into print or multi-media.
The academic world often seems to be divided between techno-freaks and Luddites, those who unreservedly welcome the new technology and those who try to hold it at bay. Most of us, however, have quietly absorbed the computer into the pattern of our academic lives, only oc casionally pausing to consider how it is also changing the character of what we know and how we know it. Of course historians have been using computers for research as well as writing for several decades. 'Cliometrics' -- the statistical analysis of larg e bodies of population and economic data from manuscript censuses, tax records and the like -- was a vogue of the 1960s and 70s. History and Computing, a collection of papers published in 1987, shows the cliometricians hard at work reconstructin g early English villages, tracing the Huguenot migrations, modelling the fluctuations of British mining company shares, editing medieval texts, and analysing images in medieval paintings. But there is hardly a paragraph in the whole volume about the ways in which the computer is actually transforming the nature of historical knowledge. Only towards the end of the volume do Bob Morris and Charles Anderson, describing a teaching experiment with computers at the University of Edinburgh, pause briefly and ask :
Do we need to go further than this and ask what will the machine do to the teaching of history? Will it distort and bias the teaching of history towards certain styles of inquiry, towards certain types of answer and subject material? I have sympathy with my colleagues who asked at a presentation of this [teaching] project, if we would still buy books for the library. We live in a generation which has taken aboard the ideology of 'information technology'. The very phrase is a significant piece of ideology, leading us to forget that many important and legitimate forms of information have little to do with this micro-electronic technology. Now machines ask for information in particular shapes and designs. They prefer informati on in repeatable, standardised forms. They are most effective with questions that involve endless repetition of quite simple enquiries. Will this be in conflict with the particular nature of history with its focus on specific events in time and space? 
In 1987 some of those questions might have seemed alarmist, but a decade later the virtual library may be nearer than we think: only last week I learned of a large government department with a well- established library of journals and monographs which has junked the lot on orders from a new manager who claims that you can now get all the information you need on the Web. The tendency for discussion to slide from 'information' -- which means the kind that you can store and manipulate electronically -- to 'learning' and from 'learning' to 'knowledge' is endemic in discussions both outside the universities and within. Already I notice that the message is abroad in the schools that you don't need to go to the library if you can get access to the Web.
It is in this general context that I think Sven Birkerts's The Gutenberg Elegies: The Fate of Reading in an Electronic Age may have something to say to us. Birkerts is a self-confessed Luddite, a bibliophile lost in the electronic labyrinth. A self-conscious literary man -- he is a regular book reviewer for The New Times -- his book breathes the rarefied atmosphere of the small American liberal arts college, the Barnes and Noble bookshop and the literary lunch. It would be easy -- perhaps too easy -- to write him off as one of yesterday's men.
The thesis of his book is well conveyed by its title and can be best summarised, its tone as well as its substance, by two short passages from the introduction:
Over the past few decades, in the blink of the eye of history, our culture has begun to go through what promises to be a total metamorphosis. The influx of electronic communications and information proce ssing technologies, abetted by the steady improvement of the micro-processor, has rapidly brought on a condition of critical mass. Suddenly it feels like everything is poised for change; the slower world that many of us grew up with dwindles in the rear v iew mirror. The stable hierarchies of the printed page -- one of the defining norms of the world -- are being superseded by the rush of impulses through freshly minted circuits. The displacement of the page by the screen is not yet total (as evidenced by the book you are holding) -- it may never be total -- but the large scale tendency in that direction has to be obvious to anyone who looks.
. . .I speak as an unregenerate reader, one who still believes that language and not technology is the true evolutionary miracle. I have not yet given up the idea that the experience of literature offers a kind of wisdom that cannot be discovered elsewhere; that there is profundity in the verbal encounter itself, never mind what further profundities the author has to offer; and that for a host of reasons the bound book is the ideal vehicle for the written word. 
There are obvious echoes in Birkerts's text of the American culture wars -- of the debates within the universities over core curricula and the canon, E. D. Hirsch's worries over the cultural literacy of the young, the American Right's fears of a loss of coherence in cultural life. Here Ned Ludd meets Edmund Burke. Where Birkerts excells is in his ability to evoke the feelings of regret and nostalgia that readers feel when books -- the objects which were once almost synonymous with knowledge -- seem to be made redundant. But his critique goes further than this. He knows -- while rather too indiscriminately deploring -- that the electronic revolution is changing the very nature of the intellectual enterprise in ways that we have barely begun to understand.
As a historian I was especially struck by Birkerts's reflections on the linkage between print and the exercise of the historical imagination. Narrative, he argues, is a form of thinking deeply bound up with the technology of print and the experience of reading. Hypertext and the Net, he predicts, will erode that sense of time as a continuous deposit of inscribed memory which underlies our sense of history:
As the circuit supplants the printed page, and as more and more of our communications involve us in network processes -- which of their nature plant us in a perpetual present -- our perception of history will inevitably alter. Changes in information storage and access are bound to impinge on our historical memory. The depth of field that is our sense of the past is not only a linguistic construct, but is in essential way represented by the book and the physical accumulation of books in library spaces. In the contemplation of a single volume, or mass of volumes, we form a picture of time past as a growing deposit of sediment; we capture a sense of its depth and dimensionality. Moreover, we meet the past as much in the presentation in books of a specific vintage as we do in any isolated fact or statistic. The data-base, useful as it is, expunges this context, this sense of chronology, and admits us to a weightless order in which all information is equally accessible. Birkerts's image of historical time as a kind of archaeological deposit of books on a library shelf may strike a particular chord with historians. For me it conjured up the very specific experience of working in the Cambridge University Library, where as some will know, the books are catalogued, not according to Dewey of Library of Congress, but in accordance with the Library's own idiosyncratic but instructive system. Volumes are listed and shelved in broad subject groupings by date of acquisition. By scanning the shelves you can trace the development of scholarship on the topic in question in a clear historical sequence. Even if we are not in Cambridge, or in a library at all, we know that more is conveyed about the time in which a book is written by its physical embodiment -- size, typeface, binding, etc -- than could be conveyed by the same characters presented in electronic form. There is something different too in writing with the books on the shelf beside me, or piled around with yellow stickers marking pages, and writing with the texts tucked away in electronic files, but invisible until I summon them to the screen. But although some valuable time signatures might be lost in the process of translating from printed text to electronic text, and although reading from the screen is different in some indefinable way from reading the book, I wonder if is it really as different as Birkerts makes it out to be? Surely it is only a bibliophile -- even a bibliomaniac -- who could seriously argue that without a book or a library we would lose all sense of historical time?
Birkerts's argument veers close to a form of technological determinism. Notice, in the passage I have just quoted, how the computer is said to `inevitably' change our sense of time, is 'bound to impinge' on historical consciousness. He may also exaggerate the novelty of electronic forms of communication such as the hypertext. As historians we have been using a primitive form of hypertext -- the footnote -- for centuries. If we could track the eye movements and mental processes of a historian reading a history book -- from text to notes, from introduction to index, to text again -- we would observe that the process of reading even a narrative history is nothing like as linear as Birkerts's account makes it seem. Here is the view of David Dobrin, a radical sceptic about the supposed revolutionary effects of hypertext:
I happen to think that hypertext is not a new text form. It is not an evolutionary advance. It forces no reconsiderations. It has not potential for fundamental change in the way we write or read. Hypertext is simply one text structure among many, made unique by the text conventions it has, conventions that guide the readers' attention and allow him or her to navigate through the text. . . Thus reading hypertext is very much like reading encyclopedias or comic books; you have to teach how the conventions work, and once you do, you've taught people to be literate in hypertext. `Built hypertext is no new category of experience; it is a minor subcategory of read text, something we all know a great deal about.' Other students of hypertext, however, argue that its potential is much more revolutionary. For example Ilana Snyder argues that hypertext radically undermines all internal hierarchies in texts, such as that between writer and reader, text and footnotes. 
Evangelists of hypertext often present themselves as radical libertarians - by breaking down the hierarchies between writer and reader, they argue, hypertext will free the reader to construct knowledge for him/herself. But while hypertexts, such as electronic encyclopedias or data-bases, may permit the user to search the text in diverse ways, the paths they follow are nonetheless built into the design of the hypertext, and freedom they allow may be less real than it seems. Indeed, being hidden and inexplicit, the design of the hypertext may arguably be more coercive than the more explicitly didactic text of the book.
Birkerts is not only a technological determinist: he is also a technological pessimist. All change is for the worst in the worst of all possible worlds. Yet perhaps the computer also opens up new ways of teaching about time and history that go much further than just the use of computerised data-bases. Consider, for example, the standard critiques of narrative history, of the kind made by Hayden White, Dominick La Capra, Leo Mink et al.  Narrative history, they argue, is always written with the advantage of hindsight. The historian's explanations of events are not like scientific hypotheses, subject to disconfirmation by subsequent events but are constructed in accordance with preconceived literary forms. If a tipster explains why a horse will win on Saturday we can test his or her powers of explanation by checking the results on Monday morning. But the historian is a bit like a race commentator who never tips and only writes up the results of the meeting on Monday morning. Sometimes an imaginative or rigorous historian introduces counter-factuals - how would the race have turned out if the favourite had scratched, or the track had been wet? - but we are generally short of methodologies for modelling such scenarios. One of the advantages of the computer, and the hypertext, it seems to me, is that it offers the potential for thinking about historical relationships in new configurations. We can think of multiple beginnings and endings, and exploit the labyrinthine linkages of the hypertext to represent them.
How might these possibilities change the way we teach history? As an urban historian, I've sometimes wondered about the pedagogical potential of certain kinds of computer games - like Sim City for example- in expanding students' grasp of these causal complexities. If we did, however, we would have to begin by eliciting and critically examining the rules built into the game - that voter discontent increases as taxes rise, for example.
The hypertext may also change the ways in which history-students learn to write. Ilana Snyder offers a prospect of a different kind of student essay:
Hypertext makes possible a new kind of essay in which linear argument is replaced by multiple explanations. Argumentation in print involves guiding one's readers through a body of information towards a univocal solution of a problem. Hypertext, by contrast, provides a set of possibilities through which many different arguments or lines might be traced or arranged. . . . Hypertext [she goes on] may well oblige us to redefine essay writing in a way that reveals not just the difficulty but perhaps even the irrelevance of producing a continuous and systematic argument. It is here, of course, that the technology of hypertext meshes with the broader agendas of critical theory and postmodernism - the destabilisation of narrative, the introduction of multiple voices, the death of the author etc. And it is here too that I suspect that some historians - myself included - will part company with the evangelists of the hypertext. In the end, we will be inclined to say, the essay - or the journal article or book- involves the adoption of a position of implied moral responsibility; it is a rehearsal for decision-making, not just for literary play. While the hypertext may put some powerful interpretative tools at our disposal, they are in themselves inert without the educated capacity to select which among the multiple connections have meaning for us.
One very traditional literary form that has been given new life by the computer is the encyclopedia or literary companion. At the moment I find myself involved in two new reference volumes, the Oxford Companion to Australian History and The Encyclopedia of Melbourne. The last, we expect, will be produced in both printed and electronic form. The beauty of the encyclopedia, as of the hypertext, is the freedom it gives to the reader to browse, to make connections that the authors themselves may not have anticipated. Yet, how rich and subtle are the new understandings which it produces for its readers? Recently I read a generally appreciative review in the New Yorker of the newly published Encyclopedia of New York City, edited by Ken Jackson, a Fulbright visitor with us just a few years ago.
The chief pleasure of the 'Encyclopedia' [writes Jonathan Franzen] lies in a kind of Derridean lateral slide of imagination. I move from 'Terrorism' to read about 'Anarchism', then go across the page to 'Amphibians and Reptiles', on to 'Birds', and (after a side trip to 'Birdland' and a courtesy call to 'Parker, Charlie') to Cockroaches. 'which are known to be attracted to toothpaste,', which brings me to 'Colgate-Palmolive' and its founder 'Colgate, William', who fled England in 1795 'to escape public hostility toward his father, who had supported the French Revolution'. It's like a game of 'Telephone: 'Anarchism' connecting with the sans-culottes not by way of history but via 'Cockroaches'. Yet there's something empty about this pleasure. A city lives in the eye, ear and nose of the solitary beholder. You tum to literature to find the interior point of intersection between subject and city, and as a living connection to New York's history a few lines of Herman Melville or Don Lippo outweigh whole pages of encyclopedia.'Like Birkerts, Franzen has a point, though perhaps it's not the one he thinks it is. In the end, what counts, is not the machine that allows us to make the connections, but the power of the mind -- informed by theory, imagination, moral insight, call it what you will -- to know which connections are meaningful and which are not.
Birkerts and Franzen are pessimists, but it would be mistaken to dismiss all they have to say. At the very least, The Gutenberg Elegies provides us with a vivid, sometimes moving account of the response of an older literary culture to the coming of a new regime -- as vivid in its way as Cobbett's Rural Rides was as a portrait of the disappearing world of pre-industrial England. And he reminds us, if reminding is any longer necessary, that in this, as in that earlier revolution, there may be losses as well as gains. How we minimise the first, and maximise the second, is the question now urgently before us.
Graeme Davison is Professor of History at Monash University, and internationally known for his work in the field of urban history.
 This paper was originally delivered at the regular Friday book discussion session in the Department of History, Monash University, on 14 March 1997. I am grateful for the participation and comments especially of Ilana Snyder, Faculty of Education, Monash University. Alan Mayne, Department of History, the University of Melbourne, converted the original Word document to HTML.
 Peter Denley and Deian Hopkins (eds.), History and Computing, Manchester University Press, p. 313.
The Gutenberg Elegies The Fate of Reading in an Electronic Age, Fawcett Columbine NY 1994, p p. 3-4.
 Ibid., p. 129.
 David Dobrin, 'Hype and Hypetext' in Cynthia Selfe and Susan Hilligoss (eds.), Literacy and Computers, Modern Language Association of America, New York 1994, p. 308.
 Ilana Snyder, Hypertext: The electronic labyrinth, Melbourne University Press 1996, p. 58.
 For example, Leo Mink, 'Narrative Form as a Cognitive Instrument' in his On Historical Understanding, Ithaca 1977, pp. 182-203; Hayden White, 'The Question of Narrative in Contemporary Cultural Theory', History and Theory., vol. 23 no. 1 1984, pp. 1-33.
 Ibid., pp. 111-12.
 Jonathan Franzen, 'First City', New Yorker, 19 February 1996, p. 90.
Graeme Davison is Professor of History at Monash University, Clayton, Victoria 3168. Australia
He may be contacted at Graeme.Davison@arts.monash.edu.au