A FEW YEARS AGO, a short article titled “What’s a Document?” appeared in Wired magazine. Written by David Weinberger, then a vice-president of the Open Text Corporation and now a business consultant, it begins with the provocative question, “Have you noticed that the word document doesn’t mean much these days?” Here is the article in its entirety:
Have you noticed that the word document doesn’t mean much these days? It covers everything from a text-only wordprocessing file to a spreadsheet to a Java-soaked interactive Web page.
It didn’t used to be like this. A document was a piece of paper — such as a will or passport — with an official role in our legal system.
But when the makers of wordprocessors looked for something to call their special kind of files, they imported document. As multi-media entered what used to be text-only files, the word stretched to the point of meaninglessness. Just try to make sense of the file types Windows 95 puts into the Document menu entry
The fact that we can’t even say what a document is anymore indicates the profundity of the change we are undergoing in how we interact with information and, ultimately, our world.1
Weinberger is making several claims here. Once upon a time, he says, document meant something: a document was something on paper that had an official or institutional role to play. And he is right about this, at least according to the dictionary. The Random House Dictionary defines a document as “a written or printed paper furnishing information or evidence, as a passport, deed, bill of sale, bill of lading, etc.; a legal or official paper.” The American Heritage Dictionary calls a document “a written or printed paper that bears the original, official, or legal form of something and can be used to furnish decisive evidence or information.” The word document, according to these sources, has something to do with writing, paper, and evidence.
Weinberger also claims that document is now being used to name things that lie outside this scope. Here too he is right. These days people regularly refer to various kinds of digital materials — text files, multimedia presentations, and Web pages — as documents. All of these materials fall outside the scope of the standard dictionary definitions. None is on paper; only some are “written” (that is, alphabetic or textual); and few if any function primarily as evidence.
From these two points, he then makes his most substantial claim, that new uses of document stretch the definition to the point of incoherence. But does this necessarily follow? Weinbergers point, it seems to me, is about more than just a word: it’s about the status of a cultural category. We once had a notion of written things, and it was coherent, it made sense. But now we’ve thrown other things into that category, things like text files and multimedia presentations and Web pages, and the grouping no longer makes sense. The reason we can’t say what belongs is that there’s no longer a rationale for putting all these things in the same bag. It’s like those children’s games where you’re shown a room filled with objects and asked, “What’s wrong with this picture?” What’s wrong with this picture is that multimedia presentations don’t belong with legal contracts; spreadsheets don’t belong with passports.
This might seem like a minor question, one best left to lexicographers or philosophers. Except that Weinberger believes there is a connection between the loss of this once-meaningful category and some of the big changes now taking place in the world. The disruption to our notion of documents is consequential, he seems to say, it really matters. “The fact that we can’t even say what a document is anymore,” his last sentence reads, “indicates the profundity of the change we are undergoing in how we interact with information and, ultimately, our world.”
I am certainly in accord with Weinberger on this last point. But I take issue with his main claim. We can say what a document is. Doing this, however, requires a somewhat different approach from that which dictionaries take. It requires going beyond word usage. It does require looking at the relevant technologies, but in such a way that we aren’t fixated on them, that we don’t fetishize them. Most of all, it requires immersing ourselves in the social roles these technologies play.
What are documents? They are, quite simply, talking things. They are bits of the material world — clay, stone, animal skin, plant fiber, sand — that we’ve imbued with the ability to speak. One of the earliest characterizations of documents comes in Genesis, and curiously, it is a description of human beings, not of written forms. “God formed Adam from the dust of the earth, and blew into his nostrils the breath of life, and Adam became a living soul.” In Hebrew, the name Adam is a pun, a play on the Hebrew word adamah, meaning “earth.” Adam is literally an earthling, and this meaning is directly preserved in the word human, from the Latin humus, earth. (If this account of our lowly origins makes us feel humble, it should, since the word humble is from the same Latin root. To be humble is to be close to the earth.)
The first human, according to the biblical account, was made by mixing breath with earth. But this is also how we make documents. Writing is the act of breathing our breath into the dust of the earth (not literally our breath, of course, but something very much akin to it: our speaking voice). While this act doesn’t literally bring the inert material to life, it does infuse it with an identity, a soul, you might say. (The Hebrew word neshamah means both breath and soul.) Or, to put this another way, writing is an act of ventriloquism, of throwing the voice into an inanimate object. Ventriloquism, the dictionary says, is “the art of speaking in such a manner that the voice does not appear to come from the speaker but from another source.” If this is a trick, it is nonetheless one we begin teaching our children at a very early age.
For many centuries Jewish scholars and mystics have explored this story of Adams creation, searching out its hidden meanings and making more explicit the connections between earth, writing, and life. Some of this speculation has come down to us in the form of legends or folktales about the creation of a Golem, an artificial man. In the most popular version of these, Judah Loew, a sixteenth-century scholar living in Prague, is said to have created a Golem to be his servant. Having shaped it from the dust of the earth, he brought it to life by inscribing the Hebrew word emet (truth) on its forehead. But the Golem ran amok and Loew was forced to destroy it, which he did by erasing the first letter of the word emet, leaving the Hebrew word met (dead). The Golem then crumbled, returning to the dust from which it was formed.
As literate members of a literate society, we are completely immersed in a world filled with these creatures, these talking Golems. We are so used to living with them in all their myriad forms that we’ve become inured, or blinded, to them. Under such circumstances, it’s hard to see them for the remarkable, even magical, beings they are. Somehow, we must see them afresh, with children’s, or aliens’, eyes. The literary critic George Steiner relates how the medieval scholar Erasmus, “walking home on a foul night, glimpsed a tiny fragment of print in the mire. He bent down, seized upon it and lifted it to a flickering light with a cry of thankful joy.”2 Writing in 1923, the anthropologist Lucien Lévy-Bruhl observed that the native peoples he was studying treated writing as magical, and “even when the native appears to have learnt what writing is, even if he can read and write, he never loses the feeling that a mystical force is at work.”3 It is easy, with Lévy-Bruhl, to see this as an example of “primitive thinking,” the perspective of uneducated and unsophisticated natives. It is harder to hold on to a sense of wonder.
Yet, without question, language is a wondrous achievement. Speech is something we assume to be uniquely human. For medieval Jewish philosophers, the Hebrew word ha-medaber — literally “the one who speaks” — was synonymous with “human being.”4 In recent decades, attempts have been made to determine whether chimpanzees or dolphins might possess some rudimentary degree of language skills. Understanding the process of human language acquisition is a challenge that draws researchers on. (“Language is a virus from outer space,” says William S. Burroughs.) Our inheritance from the Golem legends includes a fascination with robots, in both science and science fiction. Myths and fairy tales are filled with stories of talking animals and inanimate objects that can speak. There is active speculation about the possibility that intelligent, and presumably communicative, beings may exist elsewhere in the universe.
But why look elsewhere? For here, right under our noses, too close and intimate to be seen clearly, are creatures that share with us the ability to speak. And we have created them. Some of them — books in particular — aspire to nobility and long life. Others, such as cash register receipts and personal notes, typically have a less exalted status and a shorter useful lifetime. But all of them are bits of the material world we have taught to talk. Surely this is a remarkable feat.
An awareness of written forms as talking things has ancient roots. If it is subtly embedded in Genesis, it is much more explicitly made in Plato’s Phaedrus, which has an extended reflection on the nature of writing. Written forms may speak, Socrates observes, but they are dumb. Indeed, writing has some of the same disadvantages as painting. Painting may produce very realistic images, but they are representations, and not the real things. If you query one of them, it is unable to respond. “The same holds true of written words: you might suppose that they understand what they are saying, but if you ask them what they mean by anything they simply return the same answer over and over again.”
This is a crucial point. For Plato, it shows up the limits of writing. Written forms are pale shadows of their human counterparts. They are incapable of dialogue, the Socratic path to wisdom. This seems true enough, but it fails to acknowledge what is truly powerful about documents. For it is precisely in their ability to “return the same answer over and over again” that the utility of documents is made manifest. The brilliance of writing is the discovery of a way to make artifacts talk, coupled with the ability to hold that talk fixed — to keep it the same. The result is a talking thing, capable of repeatedly delivering up the same story at different points in time and space. This is something that documents do well and that people, by and large, don’t. It’s not that we are incapable of performing in such a manner. A messenger, after all, can deliver a singing telegram to multiple hotel rooms. But it is not of our essence to do so. Yet it is exactly of the essence of documents, a defining characteristic.
But at the same time that Plato underestimates the importance of sameness, he also exaggerates it. Once something has been engraved in stone, literally or figuratively, it is easy to imagine that it will last, that it will stay the same, forever. Alas, nothing, at least in the material world, is so destined. Stone, vellum, and paper all weather, though admittedly at different rates; words change their meaning; and interpretations vary depending on time, place, and many other factors. Does this mean that documents can’t actually achieve fixity or repeatability? Not quite. It just means that they can’t do it forever, for all people, for all purposes. If it is a partial victory, it is still a remarkable achievement.
Looking at documents this way sets up a strong parallel between documents and people. Both in their own ways are talking things. This is hardly an accidental parallel. Documents are exactly those things we create to speak for us, on our behalf and in our absence. It is a common enough situation, when one person stands in for and speaks for another. Lawyers and agents — press agents, literary agents — do this for a living. Lovers in some of our most famous love stories send their confidants to do the talking for them. Documents do this too, they speak for us in order to act for us, to take on jobs or roles.
The sociologist Bruno Latour uses the term delegation for this process of handing off jobs or tasks to others. He is interested in how we delegate responsibility not just to other people but to inanimate objects. He gives the example of a door. Suppose we lived in a world with walls but no doors, he begins a thought-experiment.5 Whenever we wanted to enter a room, we’d have to break a hole in a wall with a pickax. And if we then wanted to close up the wall, for privacy or to protect ourselves from the cold, we’d have to reconstruct the part of the wall we’d just demolished. A door is clearly a much more workable idea, an object to which we can delegate the task of covering and uncovering the hole.
But the door doesn’t do the whole job. There is a division of labor, a contract, between the door and its human users. The door will agree to seal and unseal the hole in the wall if we will agree to push it open and closed. The door thus assumes certain responsibilities, but in return it asks something of the people who use it. People must know how to open and close the door — they must not only have such competences (including the skill and the strength to perform these acts), but they must exercise them. In this sense, the process of delegation is reciprocal: the door delegates to us as much as we delegate to it. Of course the job of opening and closing the door can itself be delegated: to a person we call a doorman. And this task too can be delegated to an inanimate object: to a sensor and motor that automatically opens and shuts the door. The division of responsibility between humans and other humans, and between humans and objects, can be shifted.
Through examples of this kind, Latour paints a picture of a world realized through the ongoing interactions of human and nonhuman actors. We humans may make artifacts and delegate certain tasks to them, but they in turn shape our behavior, in effect delegating to us as well. It may seem odd to anthropomorphize objects in this way, to ascribe agency and intention to them. For my part, I understand this as a strategy aimed at pointing up certain truths about the social dimension of life and the sociality of artifacts. The artifacts around us, Latour seems to be saying, are more than just lumps of matter. They are bits of the material world that we have molded and shaped to participate with us — in our world, in the human lifeworld.
Documents are exactly those artifacts to which we delegate the task of speaking for us. Each kind of document, each genre, is specialized to do a certain kind of job — to carry a certain kind of information and to operate within a particular realm of human activity. Receipts are designed to witness financial transactions, handwritten notes to carry a message from one person to another, books to transmit stories and verse and knowledge deemed publicly significant. It’s useful to think of genres by analogy with human jobs. The doorman’s job is to open and close the entry door of a building. He typically wears a uniform, which signals his role, quickly establishing expectations about what we can and can’t (or should and shouldn’t) do in relation to him, what forms of service we can reasonably expect, and the nature of the dialogue we are likely to engage in with him. Of course, understanding the doorman’s role requires knowing a lot more than that he’s paid to open and close one particular door. His job is embedded in a rich set of cultural distinctions: between hotels and more-permanent dwelling places; between high-class apartment buildings and others; between tenants, guests, and employees; and so on. If we understand these distinctions, it’s largely because we, like the doorman, are immersed in the culture from which they spring.
Each document genre, too, has a uniform that signals something about the role it’s meant to play. A receipt looks like ... a receipt. A newspaper has a telltale size and shape and an easily recognizable visual structure consisting of columns of tightly packed text separated by larger headlines. A passport has its own distinctive size and shape — an official look about itself— and, internally, a characteristic visual rhythm made up of passport photo, signatures, colored stamps and seals. A pulp romance novel, a travel guidebook, and a college textbook all have highly distinctive looks as well, in terms of their shapes, cover designs, and inside typography. In each of these cases the uniform is distinctive enough to be discernible at some distance.
And rapid identification is important. Once we know a document’s genre, we more or less know what it is and how to use it, since we know what its job is. We have a basis for interpreting what it’s saying. (If I come across a piece of writing that begins, “The President must die,” it makes all the difference in the world whether it is the first line of a thriller or an excerpt from John Hinckley’s journal.) All this knowledge of form, function, use, and interpretation has been established by many years of schooling and living. To be literate is not just to be able to read and write, literally, but to be able to traffic in the many genres that surround us. Most of us may have already forgotten it, but it took considerable effort to learn how to use a checkbook, to read a job announcement, to write a resume, to fill out a job application. (And the people who, for lack of schooling or experience, are weak in these skills are at a severe disadvantage.) But this is really no different from the work we’ve had to do to understand the social roles and responsibilities, the jobs, of the many kinds of human workers with whom we constantly interact — doormen, policemen, flight attendants, and so on.
No document, no genre, is an island. Our little strip of cheap paper with blue symbols makes its mark in the world in virtue of being something with social meaning and value. To be a receipt is to be connected to cash registers, sellers, buyers, products, expense reports, the IRS, and so on. If these connections are severed — puff! — the receipt, as a receipt, is no more. It returns to the (saw)dust of the earth. It’s a curious thing about documents: you can’t see them if you don’t look at them; but you also can’t see them if you look only at them, ignoring the surroundings in which they operate.
To delegate to others is of course to exercise some measure of power and control over them. You get someone, or something, to do your bidding. But it also inevitably involves a loss of control. Someone else, an autonomous actor, now takes over, who will surely do things differently than you would — and perhaps not up to your standards. Indeed, the Golem legend can be read as a parable on the problems of delegation. You just can’t get good help anymore. It’s probably no accident that of the many versions of the legend, the one that retains the greatest currency has the Golem going berserk. This is the version that is continually retold — in the sorcerer’s apprentice in Disney’s classic film Fantasia, as well as in countless retellings of Frankenstein.
Few documents actually go berserk. But in endless ways they manage to escape the chains of their creators. Once these creatures go out into the world, they live a life that is at least partly independent of their creators’ concerns. (Think of my valedictory address.) They can be endlessly interpreted and reinterpreted, reused, subverted, and coopted for other purposes. Over the centuries the Bible has been used to argue both for and against many positions; in the nineteenth century it was invoked both in defense of and in opposition to slavery Delegation, like so much of life, is at once powerful, complex, and partial
But the question arises: Shouldn’t we say that all artifacts speak? Don’t all the things around us have a story to tell? Each was created at a certain time and place. It comes out of a certain community, exemplifies a certain style and design aesthetic. It has been subject to wear and tear, has perhaps been broken and repaired; it bears evidence of the passage of time. A Shaker cabinet has much to tell us about Shaker life and living. Its simple, spare design and careful construction arise from, and give voice to, a steady, quiet, and reverential attitude toward life. A house — any house — speaks not only of the era in which it was built, but of the lives that have been lived in it. Cars have much to tell us about their owners, about how they see themselves and the kind of image they wish to present to the world. Does this mean that all artifacts — cabinets, houses, and cars — are documents?
Actually, no. For although documents speak in these same ways, it’s not what they are primarily about. Unlike cabinets and houses and cars, documents are representational artifacts. As with crystal balls, we peer into and through them. Documents are made to carry and offer up very particular kinds of stories and in very particular ways. Perhaps the simplest instance of such a representational stance is the knife or piece of turf, mentioned in chapter 1, which would have been given as evidence of the transfer of property in medieval times. The knife was meant to stand for, to bear witness to, and to call to mind the declaration through which property was transferred from one owner to another. This is an extremely limited case of the kind of talk documents exhibit. The knife as an undifferentiated whole is meant to stand for the fact of property transfer, and it is only in the context of particular witnesses and their memories that one can have any idea what the knife is saying — or that it has anything to say at all. Documents talk much more powerfully and broadly when their representational abilities are more fully articulated.
The most obvious case of such further articulation is written language — the use of alphabetic or other symbolic marks to create composites that correspond more or less directly to human utterances. But it is not only through alphabetic and textual representations that documents can be said to speak. There are many other “written” forms, especially if we allow the root meaning of the word writing — “to score or incise on a surface” — to be heard. Maps, diagrams, pictures, photographs, and all manner of other conventional and well-articulated, nonverbal representations will count as well. It might seem slightly odd to say that a Mathew Brady photograph of Lincoln or a map of the New York subway system speaks. But certainly these forms stand in for us too, they tell stories, they represent.
To say that documents speak is perhaps to indulge in a certain poetic license, which might be avoided by simply saying that they represent or communicate and leaving it at that. But to do this would be to lose the elegant and suggestive parallel between documents and people. Like all parallels, this one is incomplete and inexact. If it is useful, however, it will not be because documents and people are exactly the same, but because their similarities and their differences are illuminating, because the partial parallel sheds light on what documents are, and perhaps also on what we humans are.
Traditionally, this parallel has been conceived of as a distinction in language — between spoken and written language, or between orality and literacy. When we humans speak directly, with our voices alone, our words are evanescent. Yet when we speak through writing, our words acquire a degree of fixity. (Verba volent, littera scripta manet, the ancient maxim states: words are fleeting, written letters remain.) There is a long history of speculation about the relationship between these two modes of communication. For much of the twentieth century, beginning with Ferdinand de Saussure, linguists have taken writing to be a straightforward derivative of speech and undeserving of sustained attention. As recently as the 1960s, the linguist Otto Jespersen declared that “language is primarily speech” and the written word only a poor substitute for the spoken word.6 Earlier in the century, another linguist, Leonard Bloomfield, was even more blunt: “Language is basically speech, and writing is of no theoretical interest.”7 If these linguists were disdainful and dismissive of written language, they were also oblivious of its nonverbal forms — or at least so uninterested as to withhold any commentary whatsoever.
This attitude has been changing, if slowly Within linguistics, Geoffrey Sampson’s Writing Systems, published in 1985, was one of the first books to take written forms as an object of study in their own right.8 About the same time, another British linguist, Roy Harris, decried the “tyranny of the alphabet over our modern ways of thinking about the relation between the spoken and the written word.”9 Operating in different disciplinary orbits, others have ventured further. In a book first published in 1968, the philosopher Nelson Goodman explored the properties of text, painting, diagrams, and musical notation as examples of “languages of art.”10 And in his Panizzi lectures, given at the British Library in 1985, the bibliographer D. F. Mc Kenzie argued for a “sociology of texts” that would broaden its horizons to include not only books, but maps, film and TV, and, in his most audacious example, natural landscape, when it fulfills a narrative function.11
Michael Buckland, a professor of library and information science at the University of California, Berkeley, has described how the European documentalists in the first half of the twentieth century argued about what could and could not be a document.12 (The documentation movement, the predecessor to today’s information science, aimed to organize scientific and technical publications.) Suzanne Briet, for example, a French librarian and documentalist, declared that “a document is evidence in support of a fact” and is “any physical or symbolic sign, preserved or recorded, intended to represent, to reconstruct, or to demonstrate a physical or conceptual phenomenon.”13 Briet’s most unusual example was that of an antelope. In the wild, she claimed, it isn’t a document, but once captured and placed in a zoo, it becomes evidence and is thus transformed into a document.
In all of this, a movement can be discerned toward greater inclusiveness: from a preoccupation with spoken forms alone to the inclusion of written forms; from the verbal or textual alone to other sign systems; from printed, bound books alone to other media and technologies. Surely one of the forces at work here is the computer, bringing together previously separated forms in a single medium. It isn’t only David Weinberger who is being led to reflect on and rework previously stable categories and boundaries.
Although all these attempts to make sense of documents are intriguing, from my perspective they work too hard: they fail to see what is most obvious and in plain sight. The simple fact is, we humans have found a way to delegate the ability to speak to inanimate objects, and have become deeply dependent on them for an endless array of services. But if I say that documents are talking things, I don’t so much mean this as an ironclad definition, as a means of drawing a tight boundary around a set of objects; I mean it as an orientation, as a way of seeing. Adopting this perspective has one immediate payoff: it gets us away from a narrow and overly exclusive focus on the details of particular technologies. Discussions of documents tend to begin with the technology — talk of clay tablets, the alphabet, the invention of paper and the printing press. While historical treatments of technology are certainly useful, they risk hiding as much as they reveal, for they tend to overemphasize the technological particulars and to mask the nature and workings of the creatures made from and with these technologies. We may miss the documents for the wood pulp.
But if we don’t insist on defining documents with respect to particular technologies, such as paper, then we have greater latitude to embrace, and to make sense of, other technological constellations as equally documentlike. Take something like audiotape, for example. This doesn’t fit comfortably under the standard dictionary definition. Indeed, it clearly fails to meet two of the three dictionary properties: it isn’t on paper and it doesn’t involve writing. (Depending on the circumstances, an audiotape recording may or may not count as evidence.) But an audiotape recording is most definitely a talking thing. And it does achieve sameness of talk: by playing the tape over and over, we can hear the same thing again and again.
Writing, narrowly defined, is the inscribing of communicative marks in a surface. Visible symbols have traditionally been scratched or carved into a hard surface, such as stone, or puddled onto a smooth surface, such as animal skin or paper. The success of this kind of writing has been dependent on the durability, the fixity, of the resulting forms. But audiotape works differently. Certainly it works by fixing marks (representations of sound) in a medium. But these marks aren’t directly available to us: we can’t hear them or see them directly. So there needs to be another device (an audiotape player) to generate the sounds we can actually hear. These sounds, just like spoken, unrecorded speech, are evanescent. Once you’ve played a recording of Dr. Martin Luther King saying, “I have a dream,” that particular performance — the acoustic signal that just rang in your ears — is gone forever. Yet these transient sounds are documentlike too; not because they are permanent, but because they are exactly repeatable. The audiotape can be played an indefinite number of times, and each time it “will return the same answer over and over again.” A different technological means is being used to achieve the same ends.
For most of the five thousand years of writing history, all our techniques and technologies have been aimed at making visible marks stick to surfaces. It is only in the last hundred years that radically different mechanisms have been invented. Audio recording technologies, film, and videotape all achieve reproducibility of talk by making transient visual and auditory performances exactly repeatable. And now, thanks to computers, we are developing still other mechanisms and forms. These include “everything,” as David Weinberger puts it, “from a text-only wordprocessing file to a spreadsheet to a Java-soaked interactive Web page.”
Are these new forms really a breed apart, as David Weinberger argues? Not really — certainly not to the extent that he suggests. Are text files and spreadsheets and Web pages talking things? Of course they are. They speak through text and sound, through still and moving images — the same basic communicative repertoire we had before computers appeared on the scene. The fact that these signs and symbols are being realized in technically novel ways —Java, HTML, global Internet connections, and so on — is a red herring. We are simply learning to throw our voice into new materials.
But there is also good reason to be confused about all this — and not just because the technologies involved are opaque to all but the most technically sophisticated. It’s confusing because the genres now online are a mishmash of old and new, and because they’re currently unstable. Some, like academic conference papers or online newspapers, have been relatively straightforwardly imported from paper. Others, like e-mail, while still displaying their offline roots, have been around long enough to take on novel features, but not long enough to fully stabilize. Still others don’t have clear identities, and hover uneasily between parody, experimental art, and just plain weirdness.
Another question must be addressed: If digital materials are talking things, do they, can they, ensure the sameness or repeatability of their talk? The answer would seem to be no, if you accept the claims of certain hypertext enthusiasts and high-tech visionaries. For such people, digital technologies have sounded the death knell for the fixed text. Because it is so easy to edit digital materials, there will no longer be a final, stable version. Because it is straightforward to produce “print-on-demand” materials that have been customized for individual readers, editions will soon consist of single copies rather than the print runs of thousands to which we have become so accustomed, effectively doing away with the notion of an edition. And as we increasingly produce hypertext webs in which there is no determinate reading order, there will no longer be any guarantee that two people, approaching the same work, will read the same text. The transience of materials on the World Wide Web — where the average lifetime of a Web page is measured in days — would seem to offer the final truth of these assertions.
There is an inherent difference between paper and digital technologies as communicative substrates: where paper documents are fixed, digital materials are fluid. This is the essence of the claim being made. One of its clearest articulations can be found in Jay David Bolter’s book Writing Space: The Computer, Hypertext, and the History of Writing. Published in 1991, before the Web exploded onto the scene, his book nonetheless manages to capture the sense of movement and change that seems to typify the Web. “Electronic text,” Bolter says,
is the first text in which the elements of meaning, of structure, and of visual display are fundamentally unstable. Unlike the printing press or the medieval codex, the computer does not require that any aspect of writing be determined in advance for the whole life of the text. This restlessness is inherent in a technology that records information by collecting for fractions of a second evanescent electrons at tiny junctions of silicon and metal. All information, all data, in the computer world is a kind of controlled movement, and so the natural inclination of computer writing is to change, to grow, and finally to disappear.14
Attractive and popular though it may be, this understanding of the difference between paper and digital forms is, quite simply, wrong. What it fails to grasp is that paper documents, and indeed all documents — are static and changing, fixed and fluid. It also fails to see the importance of fixity in the digital world. There is a reason why text and graphics editors have a Save button, after all. And if digital materials at the moment are highly impermanent, this is increasingly seen as a problem, with libraries and other institutions working hard to guarantee the survival of these materials into the future.
Bolter is quite right, however, in suggesting that digital information is “a kind of controlled movement.” But the same is true for paper documents. Look at any document — a cash register receipt, a book, a laundry list, a greeting card — and you will see the ways in which it is moving and changing and the ways in which it isn’t. Its effectiveness depends exactly on its engaging in a controlled movement: holding certain things fixed, allowing other things to change. The success of any genre, or of a particular instance of any genre, comes in its pattern or rhythm of fixity and fluidity: what information it holds fixed, when it can be changed and by whom. Different genres, to be sure, will exhibit different rhythms: a shopping list may come into existence for a brief period of time and may change often during its useful lifetime; a published volume may exist for decades and undergo relatively little change during that time.
What the death-of-fixity arguments miss is the significance of communicative stability. The ability to keep talk fixed, to guarantee its repeatability, is a basic building block of human culture. Putting talk into stable, external forms allows it to be shared, to be held in common. (The root of the word communicate, as well as of the words community and communion, is the Latin communis, meaning “common.”) Just as the stability of the earth allows us to build stable structures on top of it, so written forms provide stable reference points that help us orient ourselves in social space.
It is no accident that all our social institutions — including science, law and government, religion, education and the arts, commerce and administration — rely on the stabilizing power of documents to accomplish their ends. In the form of books and journal articles, documents are carriers of scientific knowledge. As sacred scripture, they are the central artifacts around which many religious traditions have been organized. As written statutes, charters, and contracts, they play a crucial role in constructing and regulating lawful behavior. As works of literature, paintings, and drawings, they are the tangible products of artistic practice. As textbooks and student notes, they are crucial instruments around which learning practices are organized. As receipts and accounts, memos and forms, they are critical ingredients in the way commerce, and indeed all bureaucratic conduct, is organized. In each of these cases the ability to hold talk fixed — to provide communicative stability — is crucial.
It would be strange indeed if, in making talking things out of new materials, we were somehow to omit this crucial ingredient. Better to say that we are still working out how best to achieve fixity in the digital world, not that we are trying to abolish it — or, worse yet, that fixity is inherently absent from the new medium. Most of the digital forms we are now creating and using already possess this property, to whatever limited degree, and it is crucial to their success. CNNs main Web page displays a set of stories, much as does the front page of The New York Times. Certainly their Web page changes more quickly than the Times puts out new editions. But the page exhibits a kind of punctuated equilibrium (to borrow and abuse a phrase from evolutionary theory): it stays the same for periods of time, then undergoes change, then stays the same. It may well be true that digital documents will generally maintain their sameness for shorter periods of time than their paper counterparts, but this is a different matter.
Seeing documents in this way — as talking things, as beings exhibiting patterns of stasis and change — should only increase our respect for their significance and power. Each one is a surrogate, a little sorcerer’s apprentice, to whom a piece of work has been delegated. Each one speaks out, tells its story, makes itself known. Small and insignificant though most are, when seen in aggregate they are hugely powerful, helping to make meaning and order. Of course, they don’t do this work alone — they need our help as we need theirs. And at this stage in human societal development, the conduct of life would be unthinkable without them. Is it any wonder that changes in the technology for producing and using these things are sending ripples through our institutions and practices? Even through our sense of ourselves?
For if documents are surrogates for us, they are extensions of ourselves, parts of ourselves. The best and the worst of ourselves can be found in them: our loves and hates, our destructiveness and our altruism, our honesty and our deceptiveness. Which means that looking at documents gives us a way to see ourselves. We can look at each one, close up, to see its form and content. But to make sense of these properties, we need to look at its surrounding context, the activities in which the document plays a part. And if we pan back farther still, we can see the larger slice of life, the qualities and values, that a document reflects and supports. And in seeing these forms, meanings, activities, and values, we just may catch a glimpse, like a passing reflection in a store window, of who we are and who we long to be.