Theses on the Metaphors of Digital-Textual History
Martin Paul Eve




ONE OF THE MOST CELEBRATED, and therefore over-remarked upon, of Jorge Luis Borges’s short stories is his 1941 “The Library of Babel.”1 In this tale, there is, perhaps unsurprisingly given the title, a library. However, the library is extraordinary. The library is the entirety of the universe. Composed of an infinite number of hexagonal pillars, each of which contains five shelves of thirty-two books, each of which is 410 pages in length, the library contains every work ever written and every work that could possibly exist. When the library is discovered, it triggers celebration and joy. Everything that could be known is now possible to know!

Nevertheless, notably and problematically, the library contains much that is useless rubbish. Most books contain arbitrary strings that are neither words nor mathematical formulas.2 Certainly, actual knowledge and wisdom are within the library. The problem is how to sort it out. Borges’s library appears, then, as a model of reality, narrated by an unreliable figure who embodies these paradoxes of totality.3 Everything we could know is, of course, presented to us in the form of the totality of existence. The “only” challenge is sifting the wheat from the chaff, sorting illusion from actuality, and separating idealism from the material.4 Without curated metadata and discoverability, knowledge and noise can be difficult to tell apart.

Readers assume, as I just have, that the Library of Babel is the universe, the one-to-one map of reality about which Borges quipped elsewhere.5 However, what if the library did exist in a form that was not just the sum of all existence? In computing systems, we use numeric representations to create digital sound files, movies, pictures, and text. Indeed, all computer files are, at heart, vast numbers. For example, let us say that we represent the letters a, b, and c by the familiar childish gamelike code of the numbers 1, 2, and 3. In this case, in binary, a would be “01,” b would be “10,” and c would be “11.” A system that puts these together might write “1110” for cb. Clearly, computer file formats are more complex than this extremely basic cipher. However, it demonstrates the point: “1110” is the binary for the number 14. Any computer file, then, can be expressed as a number.

Irrational numbers such as π, which expresses the ratio of a circle’s circumference to its diameter, contain every possible number that could ever exist within their infinitely long sequences. In formal terms, this is because mathematicians conjecture that π is normally distributed, which means that it is a disjunctive sequence. This distribution means that every digital file that could ever possibly exist—and, hence, every textual document that has been written, that has not been written, and that could be written—can be found at some point within π’s unending string of numbers. Our cb number, 14, is relatively trivial to find, occurring immediately after the first decimal place: 3.14. However, even if one wanted to find abc in our code, one would not have to look far. Under our system, “011011” comes to 27. One only has to search twenty-eight digits after the decimal place to come across the digit that one needs: 3.14159265358979323846264338327. Irrational numbers, in digital contexts, are the real-world Library of Babel.

This may all sound very hypothetical. However, Philip Langdale has created a (satirical) computing filesystem—dubbed πfs—that uses π to store (or, really, to compute) its data. Indeed, when one wishes to save or retrieve a file, the πfs filesystem calculates and searches through π until it finds the numbers that express the data in question. “πfs is a revolutionary new file system,” claims Langdale jokingly, “that, instead of wasting space storing your data on your hard drive, stores your data in π! You’ll never run out of space again—π holds every file that could possibly exist! They said 100% compression was impossible? You’re looking at it!”6

Of course, πfs is pragmatically useless and not a serious undertaking; it is geek humor.7 To calculate π to enough digits to find extremely long binary numbers, equivalent to files, would take a very long time. πfs works by substituting space and storage for time, in essence recomputing files every time a user wishes to access them. The πfs filesystem derives from a joke made around 2001 by Keith F. Lynch. As Lynch noted, if one calculates and stores π in binary, in addition to all of the benefits of the Library of Babel, one would be guilty of the following:

Copyright infringement (of all books, all short stories, all newspapers, all magazines, all web sites, all music, all movies, and all software, including the complete Windows source code)

Trademark infringement

Possession of child pornography

Espionage (unauthorized possession of top secret information)

Possession of DVD-cracking software

Possession of threats to the president

Possession of everyone’s SSN, everyone’s credit card numbers, everyone’s PIN numbers, everyone’s unlisted phone numbers, and everyone’s passwords

Defaming Islam. Not technically illegal, but you’d have to go into hiding along with Salman Rushdie.8

As Lynch went on to joke, “Also, your computer will contain all of the nastiest known computer viruses. In fact, all of the nastiest POSSIBLE computer viruses.”9 Indeed, the Library of Babel also contains every piece of malicious computer code and all dangerous knowledge and wisdom that could ever exist, assuming that one believes that we can represent all epistemic artifacts in digital form. In addition to being the font of all wisdom, it is also a dark archive that contains humanity’s (and every other possible species in the universe’s) worst.

I have chosen to open this volume with remarks on πfs and the Library of Babel because this book is about digital-textual histories and metaphors. What πfs shows us is a more basic premise of digital textuality: that text and numbers are indistinguishable from one another in the computational era and that to study electronic text implies a need to examine the numerical, digital, and computerized contexts and environments within which they are fashioned.10 In the contemporary period, the “history of text” we are writing is also the “history of computers” and the “history of numbers” because, in the present day, everything written can be represented numerically. Hence, while it is tempting, given the small Venn-diagrammatical overlap between brilliant writers and brilliant mathematicians, to think that the space of writerly quality and the arena of mathematical/computational quantity must sit at opposite poles, this is not true. No matter how many liberal humanist defenses we make of writing, πfs reminds us that all text can be refactored to numbers and some numbers contain all text that can ever possibly be written, if we only choose to calculate them. To understand contemporary digital textuality and its metaphors, we must also work to understand the technical components that sit beneath them and the computer-scientific principles that have conditioned their development.

Digital Book History

By now, the study of the history of books and their materialities is so old that the field of the history of books has its own history.11 Ever since Robert Darnton famously asked, “What is the history of books?” so the mythology goes, the field has burgeoned.12 In our digital and globalized era, though, the recent trajectory of material-textual studies has had an outward, planetary focus. As Dennis Duncan and Adam Smyth note, we have moved “away from bibliography’s historical focus on the Western codex, towards a global study of the materials and practices of reading and writing from scrolls and tablets to inscriptions on shells and bones.”13 This world-literary focus has diffused the scholarly practices marching triumphantly under the banner of “bibliography.” Indeed, at this point in their life cycles, textual scholarship and the study of material texts have become highly diverse fields of endeavor. Researchers in these fields focus on various global media over an extended period, covering the prehistory of the printing press to the resurgent orality of the digital audiobook.14 Such disparity has led Wim van Mierlo to ask whether any continuity even “exists in the methodological, conceptual and theoretical underpinnings of the discipline,” positing that we might speak instead of “discontinuity and diversification or fragmentation of methods and frameworks.”15 The histories of the histories of books proliferate.

This book, the product of a decade of thinking about critical approaches to text technology, continues this proliferation and dispersion by focusing on metaphor’s role in digital-textual production history.16 The book is wide-ranging in its subject and approaches. Some chapters of this book examine the historical evolution of digital-textual metaphorical concepts (such as “whitespace”), while other parts conduct sociological readings and interpretations of the metaphors that we use (such as “home” in computing contexts). Sometimes, the remarks are simply about technological and computing environments—all of which have a relevance for digitally consumed contemporary text. However, as a unifying feature, this is a book about the messy digital-linguistic frames that condition our current reading and writing practices in an era when virtually all texts are “born-digital” and most are disseminated via the web. As of the early 2020s, the vast majority of the world’s novelists, poets, dramatists, and even MFA students begin their writing days by settling down not with a pen in their hand but instead in front of the familiar blinking vertical line of a computer cursor and the latest version of Microsoft Word. There are exceptions. Don DeLillo and Jennifer Egan claim to resist the lure of the digital machine and conduct their initial drafting using pens, paper, and typewriters.17 George R. R. Martin sticks to the earlier, but nonetheless digital, technology of WordStar.18 In general, though, it is not possible comprehensively to understand the environment from which contemporary literature and text emerges without heed to the digital sites of their production, commodification, dissemination, and reception. In turn, paying attention to the environment entails a focus on cognitive metaphor, which tells us how to understand an aspect of a concept.19

Specifically, I hypothesize that digital-textual (and other computational) metaphors often move through three phases: they are initially descriptive, then they encounter a moment of fracture or rupture, and finally they go on to have a prescriptive life of their own that conditions future possibilities, even though they no longer seem to function as we might expect. This book looks for the moments when digital-textual metaphors break, because these instances show us how the possibilities for our future text environments have become constrained by metaphors that are untethered from their original intent. This book focuses on the ways that digital-textual metaphors do not work in order to uncover how our textual softwares become locked into paradigms that no longer make sense. This is important because, as Stuart Hall showed us, “metaphors are serious things. They affect one’s practice.”20 In a different context, but also stressing the importance of metaphor, Jacques Derrida writes that “metaphor is never innocent. It orients research and fixes results.”21 We are, therefore, right to pay heed to such digital metaphors. This degradation of digital-textual metaphor has also been noted by Jeff Jarvis, who charts it as part of the so-called Gutenberg Parenthesis theory, which posits that the age of print was merely a bracket of history from which we are now departing in the internet age. For, “in the simplest expression,” writes Jarvis, “it will become progressively less meaningful to say that we ‘turn a page’ in our lives, that our transparent selves are ‘open books,’ even that we have the ‘complete story’ when we read on screens in scrolls that never end.”22

This book also marks the first time I have written about textuality without reading a set of texts, instead focusing on the surrounding para-apparatuses that are the conditions of possibility for digital text. This book is an attempt to interpret the environments of digital textuality and text editing through their metaphors and histories. If it is true that paratexts in the physical space can be read, examined, and analyzed, then it is also true that the same can be said for the “metadata”—as we might term such para-apparatuses—of digital-textual objects.23 Indeed, other projects have also been invested in such work and have led the way. The legacy MONK project even stood for “Metadata Offer New Knowledge.” Yet when I talk about the analysis of “metadata” here, I am not referring to the traditional elements that might be bracketed under such an approach. I do not mean the titles of works, their ISBNs, and their DOIs, for example. Instead, I analyze a broadened scope of “metadata” that includes the digital paratextual contexts that condition electronic text production. These range from actual technologies and their implementations, such as Unicode or the history of white paper, to metaphors of “vision” in our operating systems. Language and technology become hybrid forms of metadata that influence how we understand text in the twenty-first century.

This attention to metaphor has been the focus of other studies, most notably Marianne van den Boomen’s 2014 Transcoding the Digital: How Metaphors Matter in New Media. Crucially, for van den Boomen, building on Lev Manovich’s idea of “transcoding” as a process of “‘conceptual transfer’ from the computer world to culture at large,” metaphor saturates all of our interactions with contemporary computation.24 As she points out, “we barely realize” the extent to which most of our digital terms “are metaphorical.”25 This idea of “transfer” or “transport” has long been central to our understanding of metaphor. Indeed, the etymology of the term is what allows Marshall McLuhan to state that all media are metaphors. As he put it: “The word ‘metaphor’ is from the Greek meta plus pherein, to carry across or transport.”26 Whether it is “mailing,” “chatting,” or “searching,” whenever we describe the things we do on computers, we usually use metaphors that “carry over” from other domains. (Although, as I will go on to discuss, this may, in fact, be the case with all language.) Reading such metaphors, as I do in this book, requires a delicate balancing act. On the one hand, in a point to which I will return, we must be able to recognize what van den Boomen calls the “compressed metaphoricity that stands in for a complex dynamic machinery” while seeing, on the other, that metaphor is grounded in real “things in itself.”27 The balance is between how we do still understand metaphors, even when they do not wholly “work” and have gone past the “break” point, while pointing out the limitations of just how literally we can take them.

In focusing on language and metaphor, I do not mean to understate these digital spaces’ crucial materiality. Computer systems are distinguished, argues Matthew G. Kirschenbaum, “not by virtue of their supposed immateriality but by virtue of their being material machines conceived and built to sustain an illusion of immateriality.”28 In Friedrich Kittler’s extreme formulation, all computing systems—all digital spaces—boil down to just “signifiers of voltage differences” within hardware circuits.29 Hence, despite our investment in the idea of the virtual as “unreal,” we are also finely attuned to the interplay of the real and the virtual. For example, the most significant contemporary spokescorporation of internet retail, Amazon, is dedicated to the process of making something tangible from seemingly nothing. That is, perhaps the most prominent emblem of the virtual space would mean far less if, when one clicked “Buy now,” nothing physical turned up on one’s doorstep (although in recent years, the sales of virtual servers through Amazon’s AWS business have been a major boon for the company). As we will see over the next few years as artificial intelligence (AI) language models gain access to real-world systems and become fully-fledged “agents,” the interplay between virtuality and materiality is where computational power resides.

However, the metaphors that we use to describe digital environments (as this sentence demonstrates, there is no way of not doing it) impute a materiality to the simulated interaction with the machine. Even to refer to the digital “realm” or “environment” as a virtual “space” of sorts is already falsely to hypostatize its existence as an immaterial other-place (perhaps the hideously overused Foucauldian concept of “the heterotopia”).30 Virtual reality becomes, in such language, at once a reality and only virtual, a word that—aptly, given Kirschenbaum’s repetition of “virtue”—comes to us from the late fourteenth-century Medieval Latin virtualis and virtus, meaning “excellence, potency, efficacy” but also, literally, “manliness” and “manhood.” We arrive at the contemporary meaning of “virtual” in the sense of having a true essence (nonreality) that is separate from a surface effect (solidity/reality) around the middle of the fifteenth century, probably from the meaning of “capable of producing a certain effect.”31 However, to understand the impact of the disjunct between the virtual’s surface effect and its true nature on contemporary textual production, we must examine the masking effects of the metaphorical language we use to speak of “the digital” and its implications.

In this book I rewrite a set of our digital-textual histories around the following seven theses—theses on the metaphors of digital-textual history, if you will—which correlate to each of its chapters:

1. The virtual page almost never existed.

2. The history of digital whitespace is the seriality of musical silence.

3. Digital text is geopolitically structured.

4. Digital text is multidimensional.

5. Windows are allegories of political liberalism.

6. Libraries are assemblages of recombinable anxiety fragments.

7. Everything not saved will be lost.

Each of these theses focuses on a metaphorical grounding: the idea of the “page” or “whitespace,” for instance, using these concepts as earthing points to reappraise the concrete historical unfolding—and breakage—that resulted. Some of the chapters are very much historically rooted. In the chapter on the history of the virtual page, I interview the creators of the PDF format to rewrite this format’s story. In other chapters, the debate is much more conceptual or focused on language. For example, I examine the history of the term “safety” in text processing and argue for the multiple axes across which this metaphor operates. In these senses, while some portions of this book concern new histories of digital text, other parts aim more simply to fracture and pull apart—or read and interpret—the metaphors that underpin our digital text processing.

But is “metaphor” even the correct term? A great deal of this book shows how and why our computational metaphors fall short (and analyzes the implications of this). It unpicks how we can explain particular metaphors to new users. (For instance, I ask at one point why a “menu” is called a menu, given that no food is involved, and why a “window” is called a window, given that you cannot see “through” it, although it is “framed.”) However, this problem has plagued computer designers since the 1960s; it is not a new challenge. Indeed, as Thierry Bardini put it, as just one example, “The virtual desktop was not a mere metaphor, since the user did not identify the false residual of the metaphor.”32 In other words: users took only the parts that held, ignoring the points where the metaphor fell short. We can best see this challenge when software designers do not know which metaphors they should use. Take, for example, the well-known metaphor that interface design is a “conversation” between the designer and the user. John Walker, of Autodesk Inc., poured scorn on this approach: “I believe that conversation is the wrong model for dealing with a computer—a model that misleads inexperienced users and invites even experienced software designers to build hard-to-use systems. When you’re interacting with a computer, you are not interacting with another person”—an interface-to-face—“you are exploring another world.”33 However, “exploring another world” is another metaphor that does not wholly hold. Computational metaphors involve selecting and judging parts that work and parts that do not. There are good and bad computational metaphors, but no metaphor is a direct one-to-one correlation with reality because, at that point, the metaphor would not be a substitution for the thing but the thing itself.

Perhaps the best indication of the contested status of “metaphor” in computer interface design was set out by Alan Kay, the noted computer scientist who worked at Xerox PARC in the 1970s and within several prestigious university computer science departments. For Kay, “One of the most compelling snares is the use of the word metaphor to describe a correspondence between what the users see on the screen and how they should think about what they are manipulating. My main complaint is that metaphor is a poor metaphor for what needs to be done. At PARC, we coined the word user illusion to describe what we were about when designing user interface.”34 Contrary to metaphor, which provides a substitutive context on which users can base their operational premises, illusions, of course, are meant to trick the viewer, although they are usually done, like fiction, with the consent of a willing and disbelief-suspending audience. For Kay, though, this illusory context is how we end up with supercharged analogous digital objects. There is no point, in his view, in creating digital paper that is “as hard as paper to erase and change.” Instead, we take the parts of the metaphor that we like and imbue the digital copy with magical (or “illusory”) powers: “But it is the magic—understandable magic—that really counts. . . . If it is to be like magical paper, then it is the magical part that is all-important and that must be most strongly attended to in the user interface design.”35 In reality, though, even using such terms as “interface” or speaking of the “boundary” between humans and machines operates, as does all language, through metaphor.36

The type of metaphor to which this book specifically refers is conceptual metaphor, stemming from the work of George Lakoff and Mark Johnson in the early 1980s. The fundamental premise behind their research is that “our ordinary conceptual system, in terms of which we both think and act, is fundamentally metaphorical in nature.”37 For Lakoff and Johnson, “the essence of metaphor is understanding and experiencing one kind of thing in terms of another,” and, at its heart, almost all of what we do, say, and understand uses a relational approach to definition.38 Importantly for this early work on conceptual metaphor, “metaphorical entailments can characterize a coherent system of metaphorical concepts and a corresponding coherent system of metaphorical expressions for those concepts.”39 Put otherwise, metaphors must remain coherent and consistent in their usage. Lakoff and Johnson postulated that, in most cases, even where there are overlapping metaphorical referents, such language systems do retain consistency and coherence.

Metaphors are also, though, only ever partial. “If,” write Lakoff and Johnson, “it were total, one concept would actually be the other, not merely be understood in terms of it.”40 For instance, the metaphors of “the mind is a machine” and “the mind is a brittle object” give us “different metaphorical models for what the mind is and thereby allow us to focus on different aspects of mental experience.”41 In more recent parlance, different metaphors “afford” us different comprehensive capabilities for understanding an object’s workings. These metaphors become “stacked” atop one another, usually because “two purposes cannot both be served at once by a single metaphor.”42 Each separate metaphor then “allows us to get a handle on one aspect of the concept.”43 This means that there can be a temptation, when designing a new environment from scratch, to multiply the number of metaphorical referents used to explain a system. As Douglas Kellner puts it, “Dominant [digital] metaphors draw from the human body, everyday life, home and business, nature, travel, technology, and the military and space travel.”44 As a result, we are furnished with windows, menus, status bars, pointers, sites, homes, wallpapers, desktops, check boxes, text boxes, spreadsheets, pages, icons, shortcuts, notepads, files, folders, trash cans, hourglasses, pings, dragging, dropping, clicking, right-clicking, scrolling, deleting, writing, redacting, signing, zooming, calling, playing, working, mailing, cutting, copying, pasting, snipping, screenshotting, saving, archiving, backing up, powering on, shutting down, spooling, buffering, loading, and downloading.45 And what sort of coherence or consistency do we imagine might sit at the intersection of the overlap of these terms?

Indeed, Jingfang Wu and Rong Chen have partially set out how cognitive metaphors operate across a set of nonconsistent metaphorical contexts in the computational domain.46 For instance, “a computer is a person” is a common framework in which the “CPU [Central Processing Unit] is the brain.” At the same time, a “computer is a factory” in which the operating system must “schedule computational activities to ensure good performance.” Concurrently, “a computer is an office” in which there are “notepads” that are “dropped” into “folders.” But our computers are also “containers” of “folders” that can be “emptied” from the “recycle bin.” Wu and Chen further detail the terminologies we use to describe the internet: We have “highways” carrying cyber “tourists” and “digital natives” even while the internet is also personified as “born in America.” Even as the internet is a “cyberspace,” it is a “sea” on which we “surf,” and it is a “community,” although this community space is also a “library” and a “market.”47

However, as I will detail in the section below on cultural phenomenology, an essential part of comprehending metaphor is that it cannot be divorced from the experiential encounter with contextualized language. Hence, “no metaphor can ever be comprehended or even adequately represented independently of its experiential basis.”48 This experiential encounter is broken down into ideas of prototypical primitives that exemplify the difference between definition and metaphorical explanation. While such prototypes remain contextually grounded and their properties are hardly inherent—the prototypical “chair” will depend on the context, say whether it is a formal dinner or a more casual affair—the metaphor must be understood in terms of the chairness of the chair in specific contexts.49

One of the core objections that a skeptical reader might mount to much of this book is anticipated by Lakoff and Johnson. Namely, that “it is easy to find apparent incoherences in everyday metaphorical expressions.”50 In other words, the accusation (especially by those with a disdain for analytical/language philosophy) will be that such needling is more nit-picking than new knowledge. However, as Lakoff and Johnson showed, most metaphors they examined “turned out not to be incoherent at all.”51 Yet the question remains: Are the metaphorical aspects that I am extracting and here tormenting incoherent, inconsistent, or merely partially focused? Or is it, in fact, the case that we do not “understand concepts of one kind in terms or concepts of another kind at all” but instead “only that we can perceive similarities between various concepts and that such similarities will account for the use of the same words for the concepts”?52 For example, as I will discuss, the “window” of a computer system is neither transparent nor fixed, but it is openable and framed. These windows also appear, apparently, on a “desktop” which, paradoxically, has been “wallpapered.” As Theodor Holm Nelson memorably put it as far back as 1990, “I have never personally seen a desktop where pointing at a lower piece of paper makes it jump to the top, or where placing a sheet of paper on top of a file folder causes the folder to gobble it up.”53

Indeed, computational metaphors usually “form no single image,” but they do somehow, within their own logics, rather than the logics to which they refer, “fit together” with a type of coherence until they hit the “break” point.54 The consequence of this “metaphoric ideology,” as Nelson terms it, is that “first, these mnemonic gimmicks are not very useful for presenting the ideas in the first place; second, their resemblance to any real objects in the world is so tenuous that it gets in the way more than it helps; and third . . . the metaphor becomes a dead weight” in which “once the metaphor is instituted, every related function has to become a part of it.”55 The question then becomes one of transference: What is carried across in the sharing of metaphorical terms between real-world and prototypical objects and the computational environment that we aim to make hospitable? Moreover, how might it become possible to stem the harmful proliferation of functional association to which Nelson gestures? The investigation of broken metaphor gives us scope to investigate the histories of digital-textual interfaces and whence they originate in material circumstances.

This partiality of metaphor returns us to Kay’s points about illusion and to note that “only part of [a metaphor] is used to structure our normal concepts . . . they go beyond the realm of the literal.”56 Yet, when metaphors are extended in ways that go beyond our regular, day-to-day comprehension of the referents to which they gesture, they become “idiosyncratic, unsystematic, and isolated.”57 At least part of my contention in this book is that this adjectival trinity serves as a good description of much of our computational metaphor. We have, in many ways, built entirely separate systems of language and domains of practice that do little but reenforce idiosyncratic, unsystematic, and isolated rituals of digital reperformance. Importantly, these rituals of reperformance in the computational domain reflect the original metaphorical contexts and their points of divergence. As a result, as we shall go on to see, older technologies begin retroactively to be described in terms of the digital. We say that old printed papers are “like the scrolling computer screen,” when really the likeness travels the other way.58 Indeed, as Lakoff and Johnson claim, “new metaphors have the power to define reality”; and that includes past realities and history.59 Alternatively, as Susan Leigh Star has put it, these battles about metaphor matter, because “power is about whose metaphor brings worlds together, and holds them here.”60 New computational metaphors end up being how we conceptualize extant and past technologies. At the same time, these metaphors often end up being a long way from any correlated reality.

One way we can comprehend this independence from reality is through the lens of structural metaphor. Ontological metaphors are those that impute a state of being to noninstantiated concepts. For example, in Lakoff and Johnson’s example, “Time is a substance” and “Labor is a substance” are ontological metaphors that denote both time and labor as sharing a material state. Because the shared object state is the same, the two conceptual domains are made commensurate. By contrast, structural metaphor asks us to “induce similarities” between areas. Lakoff and Johnson’s example is that “ideas are food” so you can digest, swallow, and devour both of these terms. Notably, while “the concept of swallowing food is independent of the metaphor,” by contrast, “the concept of swallowing ideas arises only under the metaphor.”61

Similar probings might well be applied to our experience of computational interface elements. For example, the framed rectangles that contain (another metaphor) our user interfaces are windows that can be “opened” and “closed” purely because we think of the ingrained “window” metaphor, even though, in reality, such virtual windows might more accurately be said to appear and disappear. The “frame” of the window may constitute the content-surface that allows for cross-domain substitutability. However, the structural-comparative metaphorical elements allow us to transfer function between these areas. They “arise out of orientational and ontological metaphors,” as Lakoff and Johnson put it.62 Yet again, though, the substitution of function is idiosyncratic, unsystematic, and isolated.

Whether metaphor or magically augmented illusion, user interface design fought in the battle between two competing cultures in the 1960s and 1970s. On the one hand, some developers felt that digital technologies should be “user friendly” and easy to learn. This group included Larry Tesler, who would go on to be responsible for our “copy and paste” metaphorical paradigm, and Jeff Rulifson, who worked on ARPANET, the precursor to the internet. On the other hand, though, was the paradigm pioneered by Douglas Engelbart, one of the inventors of the computer mouse. For Engelbart, user interface design was a way to modify users rather than being something that should fit users’ expectations. Engelbart saw, in computer interface design, a way to improve humanity by training it afresh. Tesler and Rulifson, by contrast, planned, from humanity, how to improve their user interfaces.63

This conflict came to a head under the leadership of Butler Lampson at PARC in the 1970s, who insisted on a “new ethic” where every product had to be “engineered for a hundred users” to give a broadly applicable design paradigm.64 Such a paradigm was about “tailoring the user interface to what designers could find out about or imagine about how people actually do their work” instead of using the interface to “force people to learn to do it in a new and better way.”65 This approach sounds the most obvious of methods; we should make our technologies easy to grasp and use. However, it was by no means a sure thing at the outset of computational design history. Instead, the battle was between one ideology that thought computers might improve the way we do things and another that thought we do things pretty well already and should make computers conform to existing practices.

“Practice” is an apt term to explore at this point. Talk of metaphor often moves us into conceptual and cognitive arenas. Metaphor is how we translate between an observed practice (what people do) and a cognitive frame (how they think about and understand it). However, the aforementioned debate about the role of metaphor and the place of interface design in shaping or being shaped by the user shows us the value of N. Katherine Hayles’s notion of an “incorporating practice.” Hayles sets out a system in which “inscription” is the opposite of “incorporation.” In this model, abstract signs, when written down, are given an independent existence from the writer. Written forms exist independently, but with traces, of the incorporated form that produced them. On the other hand, an incorporated gesture “such as a good-bye wave cannot be separated from its embodied medium, for it exists as such only when it is instantiated in a particular hand making a particular kind of gesture.”66

Computer user interfaces straddle this inscription-versus-incorporation boundary and must consider the interplay between cognitive and bodily actions. As Hayles defines it:

I mean by an incorporating practice an action that is encoded into bodily memory by repeated performances until it becomes habitual. Learning to type is an incorporating practice, as both Connerton and Merleau-Ponty observe. When we say that someone knows how to type, we do not mean that the person can cognitively map the location of the keys or can understand the mechanism producing the marks. Rather, we mean that this person has repeatedly performed certain actions until the keys seem to be extensions of his or her fingers. Someone can know how to type but not know how to read the words produced, such as when a typist reproduces script in a language that the typist does not speak; conversely, just as someone can be able to read a typescript without knowing how to type.67

Hence, the argument above is a debate about the site of inscription and the inscribability of the body. Those who wanted user interfaces to be easy to use saw the body as an inscribing agent, etching its incorporated practices onto machine interfaces. By contrast, the second school of thought saw the human body as inscribable, as quarried rock awaiting the sculpting influence of the interface designer. It is convenient for the subject of this book—writing with computers—that our comprehension of interface metaphor has been couched, previously, in terms of inscription. Because the question then becomes: Do we write with computer interfaces, or do computer interfaces write on us?

Given the aforementioned remarks on the maturation of this disciplinary area, it is unsurprising that I am hardly the first to explore such terrain. Indeed, this book is perhaps most indebted to the work of Matthew G. Kirschenbaum, whose Mechanisms: New Media and the Forensic Imagination (2008) and Bitstreams: The Future of Digital Literary Heritage (2021) inspired the mode of formal and forensic material thought that informs my analyses. In particular, this work takes a cue from Kirschenbaum’s consistent attention to specifics: specific technologies, specific authors, specific actants.68 I also bring to the table a work that is interested in computationally specific technicalities refracted through a cultural-studies (or cultural-phenomenological) lens.

As were Kirschenbaum’s, this book is also an attempt to move away from the phenomenon that Nick Montfort has dubbed “screen essentialism,” the tendency that we have to privilege only the final stage in the translation of digital texts into photonic forms.69 Certainly, in this work, I examine the role of visual display units. Such an angle enables my query of why whitespace is white in chapter 3. We cannot overlook the ways in which many digital technologies continue to privilege the visual mode, an aspect to which I turn in chapter 6. However, I am also concerned by the extent to which screen essentialism obscures the underpinning historical and other-material realities, which often work in contradiction to the metaphors on our screen. Whitespace, for instance, wasn’t white in the earliest forms of papermaking or in early display technologies. Further, I show, on the one hand, how the development of specific screen ideologies and metaphors—such as the virtual page—have been historically conditioned by unexpected material correlates. On the other hand, I also show how material correlates are often less predeterminate than we might expect.

What will this book examine? Many branches of literary criticism focus on narrative, style, and interpretative effect.70 Material-textual studies further abstract this, inquiring about the enframing conditioning textual-materialities that alter our understanding of a work. For example, how and why does it matter that a scribe was left-handed and left a different mark down the left side of a manuscript? What does the particular degradation of paper tell us about the historical worth accorded to a particular text and its preservation economics? Digital-material textual studies move one step further in this dialectic and ask similar questions of computerized objects. The questions I ask include: How does the term “whitespace” relate to the fact that early computer monitors were black? How and when was the virtual “page” born? What relation does that page have to the history of print pages? Are they the same histories? Should we preserve computer viruses when storing digital text, and how are they like real viruses? How did the concept of a “code library” emerge from real-world library (that is, text-lending) systems? These metaphors condition how we think of the pasts and futures of digital text technologies.

A final reflection for this section: Many of the observations about digital text in this book could be said to apply to computational systems more broadly. Indeed, text is so central to “what we do” with digital systems that you can find it in almost all corners of the virtual world. However, by using the frame of digital text, we gain a narrower entry point for such investigations while refraining from overly broad claims. I believe, though, that the wider arguments herein about computational metaphor are transapplicable beyond the limits of digital text.


1. See William Goldbloom Bloch, The Unimaginable Mathematics of Borges’ Library of Babel (Oxford: Oxford University Press, 2008); and Paul Gooding and Melissa Terras, “Inheriting Library Cards to Babel and Alexandria: Contemporary Metaphors for the Digital Library,” International Journal on Digital Libraries, 18.3 (2017), 207–22,

2. For a demonstration of this, see the digital recreation of the library: Jonathan Basile, “Library of Babel,” 2023,, accessed 6 August 2023. Note that, if flicking to a “random” book, most of the text therein consists of unreadable and arbitrary strings.

3. Jonathan Basile, Tar for Mortar: The Library of Babel and the Dream of Totality (Santa Barbara, CA: Punctum Books, 2018), 21–22.

4. Jorge Luis Borges, “The Library of Babel,” in Collected Fictions, trans. by Andrew Hurley (New York: Penguin, 1998), 113–14.

5. Jorge Luis Borges, “On Exactitude in Science,” in Collected Fictions, 325.

6. Philip Langdale, “πfs: Never Worry about Data Again!,” 2021,, accessed 16 November 2021.

7. The history of the term “geek” is fascinating, taking a detour through circus sideshows. Mike Sugarbaker, “What Is a Geek?,” Gazebo (The Journal of Geek Culture), 1998,, accessed 30 October 2020; J. A. McArthur, “Digital Subculture: A Geek Meaning of Style,” Journal of Communication Inquiry, 33.1 (2009), 58–70,

8. Keith F. Lynch, “Converting Pi to Binary: DON’T DO IT,” Comp.Risks, 2001,, accessed 16 November 2021.

9. Keith F. Lynch, “Converting Pi to Binary: DON’T DO IT,” Comp.Risks, 2001,, accessed 16 November 2021.

10. For more on this, see Steven Connor, Living by Numbers: In Defence of Quantity (London: Reaktion Books, 2016), 12; and Michael Gavin, Literary Mathematics: Quantitative Theory for Textual Studies, Stanford Text Technologies (Stanford, CA: Stanford University Press, 2023).

11. Alexandra Gillespie, “The History of the Book,” in New Medieval Literatures, vol. 9, ed. by David Lawton, Wendy Scase, and Rita Copeland (Turnhout, Belgium: Brepols, 2007), 245–86,

12. Robert Darnton, “What Is the History of Books?,” Daedalus, 111.3 (1982), 65–83.

13. Dennis Duncan and Adam Smyth, “Introduction,” in Book Parts, ed. by Dennis Duncan and Adam Smyth (Oxford: Oxford University Press, 2019), 6.

14. Matthew Rubery, The Untold Story of the Talking Book (Cambridge, MA: Harvard University Press, 2016).

15. Wim van Mierlo, “Introduction,” in Textual Scholarship and the Material Book (Amsterdam: Rodopi, 2007), 1–12,

16. Another worthwhile study on this theme is Adrian Currie, “Of Records and Ruins: Metaphors about the Deep Past,” Journal of the Philosophy of History, 17.1 (2023), 154–75,

17. Martin Paul Eve, “Jennifer Egan,” in Dictionary of Literary Biography: Twenty-First-Century American Novelists, vol. 382, ed. by George P. Anderson (Columbia, SC: Bruccoli Clark Layman, 2018), 78.

18. Matthew G. Kirschenbaum, Track Changes: A Literary History of Word Processing (Cambridge, MA: The Belknap Press of Harvard University Press, 2016).

19. George Lakoff and Mark Johnson, Metaphors We Live By (Chicago: University Of Chicago Press, 1980), 97.

20. Stuart Hall, “Cultural Studies and Its Theoretical Legacies,” in Cultural Studies, ed. by Lawrence Grossberg, Cary Nelson, and Paula A. Treichler (New York: Routledge, 1992), 282.

21. Jacques Derrida, Writing and Difference, trans. by Alan Bass, Routledge Classics (London: Routledge, 2005), 19.

22. Jeff Jarvis, The Gutenberg Parenthesis: The Age of Print and Its Lessons for the Age of the Internet (New York: Bloomsbury Academic, 2023), 14.

23. For more, see Jeffrey Pomerantz, Metadata, The MIT Press Essential Knowledge Series (Cambridge, MA: MIT Press, 2015); and Martin Paul Eve, “On the Political Aesthetics of Metadata,” Alluvium, 5.1 (2016),

24. Lev Manovich, The Language of New Media, Leonardo (Cambridge, MA: MIT Press, 2002), 47.

25. Marianne van den Boomen, Transcoding the Digital: How Metaphors Matter in New Media (Amsterdam: Institute of Network Cultures, 2014), 12.

26. Marshall McLuhan, Understanding Media: The Extensions of Man (Cambridge, MA: MIT Press, 1994), 89.

27. Boomen, Transcoding the Digital, 14–15.

28. Matthew G. Kirschenbaum, Bitstreams: The Future of Digital Literary Heritage (Philadelphia: University of Pennsylvania Press, 2021), 102.

29. Friedrich A. Kittler, “There Is No Software,” in Literature, Media, Information Systems: Essays ed. by John Johnston (Amsterdam: Overseas Publishers Association, 1997), 150.

30. Michel Foucault, “Of Other Spaces,” trans. by Jay Miskowiec, Diacritics, 16.1 (1986), 22–27.

31. Online Etymology Dictionary, s.v. “virtual,”, accessed 21 October 2021.

32. Thierry Bardini, Bootstrapping: Douglas Engelbart, Coevolution, and the Origins of Personal Computing, Writing Science (Stanford, CA: Stanford University Press, 2000), 165.

33. John Walker, “Through the Looking Glass,” in The Art of Human-Computer Interface Design, ed. by Brenda Laurel (Reading, MA: Addison-Wesley, 1990), 443.

34. A. C. Kay, “User Interface: A Personal View,” in Laurel, The Art of Human-Computer Interface Design, 199.

35. Kay, “User Interface,” 199.

36. Bardini, Bootstrapping, 42.

37. Lakoff and Johnson, Metaphors We Live By, 3.

38. Lakoff and Johnson, 5.

39. Lakoff and Johnson, 9.

40. Lakoff and Johnson, 13.

41. Lakoff and Johnson, 28. Allen Newell, “Metaphors for Mind, Theories of Mind: Should the Humanities Mind?,” in The Boundaries of Humanity: Humans, Animals, Machines, ed. by James J. Sheehan and Morton Sosna (Berkeley, CA: University of California Press, 1991), 160 focuses on the difference between the computer as a metaphor for mind and the computer as offering a theory of mind. He explicitly objects to the treatment of science as metaphor.

42. Lakoff and Johnson, Metaphors We Live By, 95.

43. Lakoff and Johnson, 97.

44. Douglas Kellner, Technology and Democracy: Toward a Critical Theory of Digital Technologies, Technopolitics, and Technocapitalism, Medienkulturen im digitalen Zeitalter (Wiesbaden: Springer VS, 2021), 37. Indeed, the entire second chapter of Kellner’s book provides a wide-ranging survey of the metaphors used in the digital environment.

45. Another interesting metaphorical conjunction is to be found in Leo Marx, The Machine in the Garden: Technology and the Pastoral Ideal in America (New York: Oxford University Press, 2000), 220, where the harmonization of machines with pastoral landscapes is theorized.

46. See also Sally Wyatt, “Metaphors in Critical Internet and Digital Media Studies,” New Media & Society, 23.2 (2021), 406–16,

47. See Jingfang Wu and Rong Chen, “Metaphors Ubiquitous in Computer and Internet Terminologies,” Journal of Arts and Humanities, 10 (2013), 15. This very descriptive article is fine so far as it goes, although I was significantly under-awed by the conclusion that “new computer and Internet metaphorical terms have largely enriched people’s vocabulary and brought people a lot of fun and enjoyment.”

48. Lakoff and Johnson, Metaphors We Live By, 19.

49. Lakoff and Johnson, 71, 122–23.

50. Lakoff and Johnson, 41.

51. Lakoff and Johnson, 41.

52. Lakoff and Johnson, 112.

53. Theodor Holm Nelson, “The Right Way to Think about Software Design,” in Laurel, The Art of Human-Computer Interface Design, 237.

54. Lakoff and Johnson, Metaphors We Live By, 44.

55. Nelson, “The Right Way to Think about Software Design,” 237. Emphasis in original.

56. Lakoff and Johnson, Metaphors We Live By, 54.

57. Lakoff and Johnson, 55.

58. Michelle P. Brown, “The Triumph of the Codex: The Manuscript Book before 1100,” in A Companion to the History of the Book, ed. by Simon Eliot and Jonathan Rose, Blackwell Companions to Literature and Culture, 48 (Malden, MA: Blackwell Publishers, 2007), 180.

59. Lakoff and Johnson, Metaphors We Live By, 157.

60. Susan Leigh Star, “Power, Technology and the Phenomenology of Conventions: On Being Allergic to Onions,” Sociological Review, 38 (1990), 52,

61. Lakoff and Johnson, Metaphors We Live By, 147–48.

62. Lakoff and Johnson, 152.

63. Bardini, Bootstrapping, 157.

64. A. C. Kay, personal interview by Thierry Bardini, 1992.

65. Bardini, Bootstrapping, 157.

66. N. Katherine Hayles, How We Became Posthuman: Virtual Bodies in Cybernetics, Literature, and Informatics (Chicago: University Of Chicago Press, 1999), 198.

67. Hayles, 199.

68. Matthew G. Kirschenbaum, Mechanisms: New Media and the Forensic Imagination (Cambridge, MA: MIT Press, 2008), 17.

69. Nick Montfort, “Continuous Paper: The Early Materiality and Workings of Electronic Literature” (presented at the MLA Convention, Philadelphia, PA, 2004),, accessed 15 March 2021.

70. See also Simone Murray, Introduction to Contemporary Print Culture: Books as Media (London: Routledge, 2021) for further unpicking of these layers.