Digital Historiography: Paper

Nicholson Baker, Double Fold: Libraries and the Assault on Paper (New York: Random House, 2001).

Jonathan M. Bloom, Paper Before Print: The History and Impact of Paper in the Islamic World (New Haven: Yale University Press, 2001).

Abigail J. Sellen and Richard H. R. Harper, The Myth of the Paperless Office (Cambridge: The MIT Press, 2002).

I am well aware of the irony: a columnist for an on-line journal is writing an article about paper. Let me state right from the start that I am not anti-paper, nor do I believe that anyone who works on this journal is against paper. The editors and contributors of JAHC have, from the start, been interested in exploring the networked computer as a medium for scholarly communication. In so selecting this medium we have not necessarily said that we are now crusading against paper. (If I were an artist and decided to paint in oil, does this make me anti-acrylic?) It is true that some technophiles, in their rush to embrace computing, scoff at paper "dead trees." In the preparation of this article, I happened upon a discussion of an on-line article titled "What happened to the paperless office?" Some of the contributions to the thread spoke of paper as if it were a bad habit that some people just couldn't break, like a nicotine addiction, because they were either clinging to the past or simply incapable of embracing the electronic future. To which I ask, "what's wrong with paper?"

Abigail J. Sellen and Richard H.R. Harper answer "nothing at all." The authors study the use of paper in the modern workplace, observing how knowledge workers, economists, lawyers, office managers and flight controllers actually use paper in their work. This book is a part of a growing literature in ethnography, sociology and anthropology on the study of modern technology. Notable works include On Line and On Paper by Kathryn Henderson, who studied the effects of computer-aided design technologies on engineers, and documented their continued use of and preference for paper. When ethnographers and sociologists go "out into the field" today, they are just as likely to be observing activities in libraries, office places, Kinko's copy centers and schools, studying how people use technology in their daily lives. Rather than assuming that technologies are neutral or that their properties are inscribed by their designers, a growing consensus of this literature suggests that users play an important role in shaping and defining technology. This emerging field provides an excellent corrective to the speculations and utopian dreams that have accompanied the development of electronic technologies.

At least since the invention of the telegraph, inventors and technologists have been predicting the imminent downfall of paper. The authors briefly attempt in the beginning of the book to pinpoint when and where exactly this "myth of the paperless office" first sprang. This proves to be a difficult task, however, as the paperless office is much like an urban legend: many people know of the story but cannot determine where, when or who started it. The authors identify two recent periods where the myth was particularly powerful: the late 1970s, with the development of the first PCs, and the 1990s, with the explosion in growth of the Internet. Prophets of the Internet claimed that in the not too distant future many activities that were once carried out in physical space would be transported into placeless cyberspace. In the electronic world, bits of information have become separated from the atoms that had once transported them. Since bits can now flow through the evanescent electronic ether, we no longer need to worry about transporting the atoms. Why go to a store when you can simply order your purchases on line? Why travel to a university when all the information stored in the world's libraries and archives can be accessed from your home? Why walk in a park when you can socialize with friends in a chatroom? Eliminating paper was only the tip of the iceberg; paper was a part of a physical world that cyberspace was quickly supplanting. And, like an urban legend, the myth of the paperless office is false: atoms still continue to matter. While many consumers shop on line, brick and mortar stores still dominate retail sales (consider what has happened to all those dot.coms). While some educational activities are carried out over the Internet, students continue to flock to real classrooms. And, as Sellen and Harper document, we still use paper, apparently more and more of it, despite the proliferation of electronic tools. They conclude that paper and screen coexist in the modern workplace. This might appear to be a somewhat obvious and commonsensical observation, however, their book is far from trivial. They carefully document the complex interplay between paper and screen, and are especially interested in understanding why paper persists.

Simply put, we continue to use paper because it is useful. Sellen and Harper in particular study the "affordances" of paper. What does paper afford, the authors ask; that is, what types of activities does paper allow a user to perform? "An affordance," they write, "refers to the fact that the physical properties of an object make possible different functions for the person perceiving or using that object. In other words, the properties of objects determine the possibilities for action." (17) The analysis here, then, is not to examine the physical properties of paper but rather the affordances of the medium, or, more precisely, the uses to which people put those affordances.

All too often, those who push for the paperless office fail to take into account these affordances. Interestingly, it is precisely paper's physicality and tactility, the very opposite of the cybernaut's anti-atom bit world–that are the source of many of its affordances. Paper is easy to grasp, carry, manipulate, fold, and mark upon. The marks on paper are fixed, and any markings made after it is printed upon are readily apparent. In the modern workplace, paper often affords collaborative work, especially when documents must be shared among a number of workers. The sharing of paper documents facilitates important social connections between workers. "Because paper is the physical embodiment of information," observe Sellen and Harper, "actions performed in relation to paper are, to a large extent, made visible to one's colleagues." (66) In their studies of economists at the IMF, Sellen and Harper observed which reports received the most attention, which were simply glanced at or leafed through, which were set aside. These gestures are not incidental to the work, but are vital to the flow of information. Such gestural information is harder to decipher when observing users looking at screens. (What is that person doing: reading your report or looking at email?) Further, at the IMF authors of reports would often hand-deliver copies to prospective reviewers as a way to heighten the gravity and importance of the report. A mass emailing of an attached file does not carry the same social importance at the IMF.

Another of paper's affordances is its spatiality. Paper documents are easy to spread out so that several of them can be viewed and compared at once. Paper laid out in space is easier and more efficient to navigate. Air traffic controllers, especially, value this affordance of paper. At the London Air Traffic Control Centre, controllers rely upon paper flight progress strips, 1x8 inch strips of paper that contain information on a specific flight. Controllers mark real time data directly on these strips, and arrange them to the more easily see the relationships between different flights. Like the activities of the IMF, these activities usually occur collaboratively, and paper makes the process transparent. Thus, people who continue to use paper rather than electronic forms of information are not necessarily Luddites or those simply unwilling to change their habits. In many of the cases Sellen and Harper study, people continue to use paper because it facilitates work, it affords work in ways that electronic tools have yet to improve upon. These affordances become apparent when companies decide to "go paperless." Sellen and Harper document two companies who decide to switch to electronic forms of document manipulation, eliminating paper from their workplace. In one case, a company eliminated paper because of its association as an old fashioned technology. Paper in an office makes a company appear less modern and cutting edge. This is often a poor reason to eliminate paper, argue the authors. This company found that by eliminating paper many of the practices of the workplace were disrupted, causing frustration and a drop in productivity among the workers. This company failed to heed the affordances of paper. Another company, however, wished to alter how their employees worked, and thus eliminating paper was part of this larger goal. Replacing paper with screen is not simply a matter of exchanging an old technology for a new one; workers will need to learn new habits of work that the screen affords. As Neil Postman once observed, technological systems are like ecologies. Adding a caterpillar to an environment does not yield the same environment plus caterpillars; you have created a totally new environment. Replacing paper with screen does not yield the same workplace; you have created an entirely new workplace.

The question is not which medium is better or worse, paper or screen. Indeed, paper has notable drawbacks. Unlike a screen, paper cannot display dynamic information. It is harder to update information quickly and easily. Compared to computers, paper affords bulky information storage. Linking to related materials is much slower with paper. Paper does not display multimedia information with much efficiency. Paper is a flexible, mobile medium that can only display static, fixed data, where the screen is a relatively immobile medium that can display dynamic and multimedia information. The question is, what type of activities do you wish to perform, and which medium best affords those activities?

One of the drawbacks of paper, according to Sellen and Harper, is that, while it affords the activities that workers are performing now, paper is less suitable for the long-term storage of that work. Many of the paper documents that Sellen and Harper study are what they term "hot files," those documents that need to be close at hand for work that is occurring in the immediate present. "Warm files" are those that, while not needed at the moment, must still be within easy reach for on-going projects. They probably sit in "to-do" boxes or in desk drawers. "Cold files," on the other hand, are those that record previous work activities, are stored in filing cabinets, and are usually kept by businesses "just in case." "Paper simply lacks the affordances necessary to allow organizations to make effective use" of cold files, argue Sellen and Harper. Paper cold files take up too much space, they become cumbersome to manage and organize, making it difficult for someone later on to mine the documents for valuable information. "There are many ways," they conclude, "in which paper is an outdated, unsuitable technology for preserving and leveraging knowledge of the past." (170) Imagine if an historian were to utter these words. Imagine if an archivist or librarian were to utter them.

Nicholson Baker's criticism is that many librarians and archivists have been uttering exactly these words for some time now. His book is–if I may be permitted the understatement–controversial. His argument is that in their rush to embrace microfilm, which promised to reduce space and ensure the longevity of records, librarians and archivists were too quick–even negligent–in destroying the original paper copies from which the films were derived. The controversy raised by his book has as much to do with its accusatory and conspiratorial tone as it does his skepticism of the durability, utility and cost-effectiveness of film or his disgust at the practice of gutting and destroying original runs of American newspapers. Baker identifies the culprits and "names names," often times resorting to ad hominem attacks. He implies that the destruction of old newspapers was a by-product of the Cold War, carried out not only by the Library of Congress but by large corporations and government agencies, indeed the entire military -industrial complex. By destroying the paper, we have forever lost an important connection to the past, for once the microfilm copies disintegrate–and he claims that the paper will last at least as long as the film–we will have no original from which to copy.

If one is able to look past his conspiracy theories and name-calling, a reader of this book would observe that Baker is not anti-technology or even particularly pro-paper. It is not the practice of microfilming per se that Baker finds offensive as it is the destruction of the paper originals. "There is nothing intrinsically wrong with microfilming," he writes. "Taking tiny black-and-white pictures of things isn't objectionable so long as the picture-taking isn't destructive."(25) Aside from serving as original sources, Baker sees value in the paper that microfilmed copies do not preserve. For instance, the paper itself conveys information that cannot be replicated on film. The physical composition of the old newspapers themselves—whether they were made from rags or from wood pulp–contains information that would be of interest to economic historians or historians of technology. I am reminded of an account in John Seely Brown and Paul Duguid's The Social Life of Information that resonates here. Duguid recounts an experience in an archive, where the smell and dust made him pine for a digital system that would eliminate the allergy-inducing stench. Another historian sat next to him, examining a similar box of documents. Rather than reading the words written on them, the historian sniffed each letter, occasionally jotting down a few notes. Duguid later learned that the man was a medical historian studying an outbreak of cholera. Letters were often sprinkled with vinegar to keep the disease from spreading. Business letters like the ones they were examining would rarely have made overt mention of the disease–bad for business–but the tell-tale smell of the vinegar betrayed the words on the page. Microfilming or digitizing those letters would preserve the words–the bits of information–but the information contained in the atoms would be lost in the translation. Microfilming breaks up the layout of the original page. "The size of newspapers is indispensable to our experience of their content," argues Baker.

The newspaper reader proceeds nonlinearly, not as he would holding a typical book, but circling around the opened double-page spread, perhaps clockwise, or counterclockwise, moving his whole head as well his eyes, guided by island landmarks like photos and ads. Even the papers that have no pictures at all have a visual exorbitance, a horizon-usurping presence that microfilm's image (which one observer in the seventies compared to "kissing through a pane of glass") subverts and trivializes. (24-25)

The spatial arrangement of the information on a specific page is itself a source of information, but this contextual information is lost in many microfilmed versions. Reading a microfilmed copy is like reading a document in translation: it might be acceptable, but much is lost in the translation. Any historian will tell you, it is always better to read the original.

Microfilmed copies of newspapers often ignore or badly reproduce illustrations, pictures and other images. As someone with a scholarly interest in images, I found this a particularly unsettling portion of Baker's book. Microfilmed versions tend to focus on the words more than illustrations, cartoons, and advertisements, all excellent historical sources. When microfilmers do pay attention to the images, these are often badly reproduced, especially color images which lose much information when copied to black and white. Baker's concern is that once the microfilmed version becomes the only source at our disposal–the original having been destroyed–we will have lost a significant portion of the historical record that might have been easily saved.

As we might expect, Baker is critical of digitalization as well, especially the costs associated with upgrading. His argument is a familiar one to historians: since software makers and computer manufacturers constantly change technologies, any documents electronically copied today might be unreadable in only a few short years. (I am thinking about all of the files I still have saved on 5 1/4 floppy disks in WordStar, files I simply cannot read now.) Digitalization cannot serve the functions of a true archive, writes Baker, since "a true archive must be able to tolerate years of relative inattention; scanned copies of little used books, however, demand constant refreshment, software-revision-upgrading, and new machinery, the long-term costs of which are unknowable but high." (242) Paper originals do not require systems upgrades, and can sit unattended for much longer periods, a conclusion based on Baker's assessment that paper lasts longer than librarians and archivists are willing to concede. Businesses might find paper "cold files" a logistical nuisance, but historians earn their livings exploring such files. "One of the important functions, and pleasures, of writing history," according to Baker, "is that of cultural tillage, or soil renewal: you trowel around in unfashionable holding places for things that have lain untouched for decades to see what particularities they may yield to a new eye." Baker fears–if I may borrow the language of Sellen and Harper—that our culture will only want to retain "hot files," thus ignoring the value of "cold files" to our culture.

We mustn't model the digital library on the day-to-day operation of a single human brain, which quite properly uses-or-loses [information], keeps uppermost in its mind what it needs most often, and does not refresh, and eventually forgets, what it very infrequently considers–after all, the principle reason groups of rememberers invented writing and printing was to record accurately what they sensed was otherwise likely to be forgotten. (244-45)

As with his critique of microfilm, Baker does not dismiss digitalization technology completely out of hand. He cites approvingly the case of John Warnock of Adobe Software, himself a collector of books. Warnock has digitally photographed his own book collection, producing high-resolution images of the originals. He has founded the Octavo Corporation, which seeks to preserve old books through digitalization, without destroying or damaging the original versions. Because eventually, argues Baker, low-resolution black and white microfilmed versions of old newspapers will eventually prove of little value to scholars, who will wish to see the originals. Or perhaps a company such as Octavo may want to scan a volume of newspapers, at high resolution, nondestructively, as if it were a fragile sixteenth-century folio. The president of the company will make inquiries at a library of historical society in his city. He will be led into a room that holds four hundred gray cabinets of microfilm. "But we need the originals," the company president will say. "Where are the originals?"(253)

Paper has been affording all sorts of activities long before the advent of the modern workplace. In the West, historians dutifully nod to paper's critical role in the development of printing and the explosion of knowledge that revolution engendered. Western historiography tends to see paper solely as a footnote in the larger history of printing, since the two technologies developed at roughly the same time in the West. As such we tend to entwine printing (and its subsidiary technologies) and paper together. If the computer is quickly displacing printing technologies, then paper must surely be close behind. Only when we cleave paper and printing, however, may we begin to understand paper's longevity as a material for documenting thought and knowledge.

Paper's role in Islamic civilization was quite different than what transpired in the West, as Jonathan Bloom documents. Printing developed much later, in the eighteenth century, nearly a millennium after paper was introduced into the Islamic world. This lag time between the introduction of paper and the spread of printing was not wasted; paper afforded its own revolution in thought and knowledge, shifting Muslim civilization from an oral culture to a scribal culture. Paper also produced dramatic effects on Islamic visual culture, from book illustration, pottery and crafts, to mathematics, geography and commerce. "The importance of printing in the great transformation of European civilization is undeniable," argues Bloom, "but paper, the material that Islamic civilization carried from China to Europe, was an even greater invention in the history of humankind." (226)

Western historians draw attention to the role the printing press played in shifting Europe from a scribal culture–manuscript-based–to a typographic culture. They have tended to pay less attention to the shift from oral to scribal culture, which is what paper engendered in Islamic civilization. The Abbasid caliphate's bureaucracy first switched from parchment and papyrus to the use of paper for official documents in the eighth century. Official documents required a careful and elegant script, difficult to achieve with the materials of the time. Smooth, high-quality paper afforded this surface, which accepted the cheaper and easier to produce carbon-based ink more readily than parchment. With cheaper inks and high quality papers, Muslim scribes forged the broken cursive script that dominated government documents, copies of the Koran and, later, manuscripts of all kinds. Paper, Bloom argues, afforded calligraphy, one of the highest of Islamic arts.

Paper was central to a revolution in mathematics. Mental calculation, the abacus and dustboards–a slab covered with a thin layer of sand that one could use to make marks upon–were the chief means of reckoning. Paper afforded the Indian system of calculation, argues Bloom. This system, which required several steps, was difficult to carry out mentally, and dustboards had to be erased after nearly every step. When calculating on paper, however, the mathematician could retain all the steps of the calculation, allowing for revision and correction and a permanent record. "The new technique of calculating with Hindu-Arabic numerals therefore spread throughout the Muslim world at exactly the same time as the new medium of paper." (134)

Paper afforded other notational systems as well. Through a similar process that moved Islam from an oral culture to a scribal culture, paper afforded the opportunity to preserve music in notational form. Cartographers explored new forms of graphic representation of the earth. Genealogical charts were also drawn out on paper. Paper appears to have been the medium of choice in these other disciplines because of its relatively cheaper cost and the fact that paper was quickly become an elite material, used by other scholars and government officials.

As paper became less expensive, its effects were felt outside of elite culture. Merchants participated in a burgeoning credit economy which depended upon the several types of commercial paper flowing through it. Probably spurred on by the example set by the Abbasids, merchants employed letters of credit, notes, accounts, wills, inventories (but, interestingly, not paper money).

More significantly, paper had dramatic effects on the visual arts and crafts. Book illustration emerged as an important form of painting in Islamic culture, made easier as the size of paper became larger, which provided greater space upon which to compose. Paper also altered many artisan crafts, from weaving, metalwork and pottery, to architecture and carpetmaking. As we might expect in an oral culture, many of the designs embossed on these crafts were retained in the memory of the craftsman. As such, the patterns on the art were relatively simple compared to later work. As paper became cheaper and more widespread, elaborate designs sketched out on paper became more and more common. These plans on paper would then be transferred over to the new medium. "Design" itself becomes possible, that is, artistic patterns divorced from any specific context, applicable to many media. This had two important consequences. Since design was removed from the memory of the builder, the act of design was separated from the act of construction. One might design an artifact without ever being involved in its actually construction. Islamic culture thus witnessed a greater uniformity of design, spurred on in part by paper plans, as "designs on paper" could move more quickly and easily over a wide area than the memory of a designer/builder, rooted as he was in place. And since design and construction were separated, designers became separate from artisans. Just as in the Italian Renaissance, designers and architects emerge from anonymity, signing their names to their works.

Paper provided Islamic culture a representational space upon which scholars could explore new forms of graphic representation. In the West, we view paper as instrumental–subordinate to printing, but instrumental nevertheless–to the spread of literacy. While we understand that Luther's propagandists printed broadsheets and penny newspapers printed large illustrations, we tend to view the print/paper revolution as one that made words available to larger audiences. While paper certainly had notable effects on literacy in Islam, Bloom's book suggests that the real revolution paper afforded was in pictoricity.

Thinking of paper in this fashion might offer an interesting way to think about the future of paper and screen. Perhaps paper will remain an excellent medium for words, while the screen might prove to be a better surface for images? As Sellen and Harper observed, paper is not as useful for displaying dynamic data, the interactive charts and information spaces that computer screens have afforded. The graphics capabilities of computers led directly to a revolution in the sciences and mathematics; the Mandelbrot Set, chaos theory and other "new sciences" were developed largely as the result of the fourth generation computer, in a similar manner that paper afforded the spread of Hindu-Arabic calculation. If paper declines as a medium of communication, perhaps it will be because we will have transferred our images—our maps, diagrams, visualizations, charts and other forms of graphical representations of thought–to the screen. When writing about "Paper After Print," perhaps future historians will observe that the computer afforded a new revolution in pictoricity as dramatic as the one that swept across Islam centuries ago.

Historians have only tentatively begun to explore the use of dynamic visual information as a form of historical scholarship. We continue to remain wedded to words and print, which may explain why articles in paper journals continue to carry more authority than do articles appearing in on-line journals like JAHC (although I believe this attitude is changing, albeit slowly). If historians do venture to use multimedia information in electronic publications, as Robert Townsend observed in the May 2002 issue of AHA Perspectives, this information is used chiefly to illustrate or supplement a written textual account, and in many cases does not add significantly to the argument. Townsend laments that historians have not fully explored the potential the electronic medium has to offer. As long as we remain so wedded to print, historians may never see the screen as a serious alternative to paper. Indeed, we may continue to believe that paper serves our needs just fine. Only when historians learn to appreciate the affordances of the computer screen–and recognize that these affordances are different from those of paper–will digital scholarship be accorded the same weight and authority as paper scholarship.

Making this statement, however, does not make me "anti-paper."