CHAPTER 6 . . . MARIA ALBERTO AND ELIZABETH SWANSTROM

WILLIAM GIBSON, SCIENCE FICTION, AND THE EVOLUTION OF THE DIGITAL HUMANITIES

In 2017 Chris Sevier filed a lawsuit against the state of Utah for the right to marry—his computer. The case was promptly dismissed for several compelling reasons, including the fact that Sevier’s computer was not at least fifteen years old and therefore did not satisfy Utah’s age-of-consent requirement. ¹ As ludicrous as this case might seem, however, it was neither the first—nor will it be the last—attempt to wed a computational entity. In 2009, Japanese gamer Sal9000 married Nene Anegasaki, “a character in the Nintendo DS dating simulation game ‘Love Plus,’” ² and in November 2018, Akihiko Kondo wed Hatsune Miku, “a hologram that was created by a computer as singing software.” ³ Moreover, such real-world unions have multiple literary precedents, dating back to Auguste Villiers de l’Isle-Adam’s depiction of Thomas Edison and the gyndroid Hadaly in L’Ève future (1886), further back still to E. T. A. Hoffman’s automaton Olimpia in “The Sand-man” (1816), and arguably even to Pygmalion’s marble-carved lover Galatea in Ovid’s Metamorphoses (8 CE)—examples that demonstrate a long-standing interest in these unusual accords. But the contemporary penchant for an amans computans finds its most vivid literary blueprint in the work of William Gibson, whose cyberpunk fiction delights in merging computational and human entities.

In Neuromancer (1984), which provides the master template for subsequent unions in Gibson’s oeuvre, this type of merger occurs when two artificial intelligences, code-named “Wintermute” and “Rio”/“Neuromancer,” consolidate to form the titular singularity, Neuromancer. This pattern is repeated in “Winter Market” (1986), when disabled artist Lise “dry dreams” with editor Casey, merging her consciousness with his as a first step toward merging with the digital ’net. A similar pattern surfaces in Mona Lisa Overdrive (1988), when simstim star Angie and her lover Bobby Newmark merge their consciousnesses with a vast, newly formed computational storage space called the Aleph, and again on a smaller scale in Pattern Recognition (2003), with the piecemeal creation of “the Footage” and its subsequent reconstitution by fans. Later the sustained dialog between proximate and distant futures in Peripheral (2014) follows such unions in temporal rather than spatial terms, as enabled through the medium of a quantum computer. Perhaps most visibly, in Gibson’s novel Idoru (1996) the marriage is literal, as eccentric rock star Rez weds Rei Toei, an “idoru” or computer-generated “synthetic personality” who exists only in virtual space.

Though itself an interesting phenomenon, this pattern of integrated mergers between distinct and seemingly incompatible entities in Gibson’s fiction has ramifications beyond the romantic plotlines that often surround them. In fact, we assert that this trope in Gibson’s fiction—which itself has been fundamental in shaping popular and scholarly notions about computation since the early 1980s—has particular resonance with the emerging field of the digital humanities, or DH.

Defining DH

In their introduction to Dialectic of Enlightenment , Max Horkheimer and Theodor Adorno make a scathing claim about the “totalizing” and “corrosive rationality” of Enlightenment thought, noting that within it “anything which does not conform to the standard of calculability and utility must be viewed with suspicion.” ⁴ First published in 1944, such a statement reads as a trenchant critique of an epistemology that helped pave the road to World War II, as well as a prescient excoriation of the larger culture industry that solidified in the postwar United States. Decades after its original publication, and within a technological-aesthetic system that did not exist during its time, Adorno and Horkheimer’s statement works surprisingly well to describe a particular friction within the digital humanities—specifically, between quantitative computational methods on the one hand and subjective qualitative analysis on the other.

In order to understand how it is that both Gibson’s work, and science fiction more generally, have been fundamental to the digital humanities, it is necessary to define what constitutes DH in the first place. This is a notoriously difficult task. In Matthew K. Gold’s first Debates in the Digital Humanities collection, for example, the “Day of DH” essay offers twenty different voices, each with a distinct vision of the field. These range from the concise (e.g., John Unsworth’s “using computational tools to do the work of the humanities”) to the community-oriented (e.g., Kathyrn E. Piquette’s “community of practice” and Jason Farman’s “methodology and a community”) to the optimistic (e.g., Mark Marino’s “a temporary epithet for what will eventually be called merely humanities”) to the tautological (e.g., Rafael Alvarado’s claim that “the Digital Humanities is what digital humanists do”). Of definitions listed in this survey, only one—Ernesto Priego’s—speaks specifically to the importance of digital culture within DH: “The scholarly study and use of computers and computer culture to illuminate the human record.” ⁵ Definitions such as Priego’s are extraordinarily valuable to surveys such as ours because they acknowledge how the human elements of DH cannot be separated from their purely computational counterparts. ⁶

For our purposes, DH crosses over among several of these definitions, naming a field of research and inquiry that investigates any overlap between computational technology (the digital) and traditional humanism (the humanities). If our reconstituted definition seems to cover quite a lot of ground, that is because it does: intentionally, strategically, and even sometimes unconvincingly. Because DH does not have a straightforward, linear history—a reality that constitutes part of its appeal and its risk—it is productive to think of this field as a braided entity composed of three strands: media studies, experimental aesthetics, and humanities computing.

The first of these strands is most likely familiar. Media theory helps us evaluate our methods of communication—from Plato’s critique of writing in the Phaedrus to Marshall McLuhan’s critique of the alphabetic consciousness that writing helped foster in The Medium Is the Message (1967) to N. Katherine Hayles’s argument that the human body is integral to the formation of digital literature in Writing Machines (2002). The second strand, that of experimental aesthetics, is also probably familiar, although its particular attributes vary widely and necessitate a disclaimer that in this context we mean the term to signify any artistic act of making that concerns itself with both digital technology and formal innovation.

The third term—humanities computing—is perhaps less well known, though it names the field that uses computational methods to conduct statistical analyses of literary texts. While this might sound like an off-shoot of the contemporary DH frenzy, it is actually the other way around: humanities computing precedes DH by nearly forty years and has only recently expanded to invite the other two braids of inquiry—media studies and experimental aesthetics—into its fold. In fact, when the National Endowment for the Humanities (NEH) opened its Office of Digital Humanities in 2008, a survey of NEH-funded projects revealed that the NEH had been supporting DH projects since its own inception—long before DH existed as such. ⁷

Yet humanities computing did not merge with these other two strands until fairly recently. Two moments clearly enabled this union. The first has to do with the very name “digital humanities,” which resulted from a strategic publishing decision. In 2001, DH scholar John Unsworth was putting together a collection of work composed of both traditional humanities computing essays and other texts that were less technical than analytical. Although both types discussed the role of digital technology in humanities research, their various approaches prompted Unsworth to propose a more fluid name than the editors’ recommended “digitized humanities”: instead, Unsworth “suggested ‘Companion to Digital Humanities’ to shift the emphasis away from just digitization.” ⁸ The decision to avoid “computing” and “digitized” is understandable; such choices could have alienated readers by implying the need for a more substantive computing background. The alternative “digital” is also more capacious and lacks the finality of the past participle “digitized,” which suggests a computationally completed act of translation. Thus, Unsworth’s new term “digital humanities” struck a balance among these competing needs by merging their most important qualities.

A second turning point can be found in the “big tent” theme of the annual DH Conference at Stanford in 2011. The metaphor of a giant circus tent opened DH to a more expansive sense of research that included media theory, aesthetics, and cultural studies without requiring the demanding technical expertise that had defined humanities computing for decades. With so many new recruits welcomed into its fold through this merger of interests, DH seemed poised to be, as a New York Times article had predicted a year earlier, “the next big thing.” ⁹

Even though the union of humanities computing and DH has resulted in a wonderfully wide latitude in terms of research options, this latitude can also be confusing to people inside and outside DH’s “big tent.” For those outside the tent, the term “DH” is so amorphous that it often mystifies. For those inside the tent, there is the same difficulty of finding common ground among so many different approaches and objects of study, but there is also the larger problem that C. P. Snow articulated in his 1959 Two Cultures lectures. This is a problem that has plagued Western epistemology since the scientific revolution: the divide between scientific and humanist knowledge production. As both a practicing scientist and a literary creator, Snow experienced this divide firsthand, writing that he was constantly “moving among” two groups: “comparable in intelligence, identical in race, not grossly different in social origin, earning about the same incomes, [and] who had almost ceased to communicate at all.” ¹⁰ Though many of his subsidiary arguments are no longer applicable in the same way today as they were in his time, his central claims about two groups resonate especially well in the context of DH.

What we mean by this is that although the “merger” has ostensibly happened—we’re all under that same big tent—our training and approaches to DH are often at odds, even when we make serious efforts to understand each other. Cultural critics are often dismayed by what they perceive as decontextualized “data dumps” coming from the humanities computing side of things, and those with more technical expertise in computation are often mystified by the qualitative analyses about computation that offer little evidence of concrete knowledge regarding the nuts and bolts of computational procedure. ¹¹

This conflict between the “two groups” of DH is only compounded by how humanities computing and DH can be weirdly out of step with the times, but in completely different ways. Those who focus on digital culture are often at the cutting edge in terms of cultural and political theory—but at the same time, they are often behind in terms of technological skills and knowledge. By contrast, humanities computing scholars are computationally literate but often behind the times in terms of cultural theory and sometimes even regarding the social and political implications of their work.

This tension recalls a darkly comedic moment in Neuromancer , in which the AI Wintermute confides to Molly Millions how frustrated he has become with the Tessier-Ashpool family that owns him. When she recounts the scene later, it is not without sympathy: “They were always fucking him over with how old-fashioned they were, he said, all their nineteenth-century stuff.” ¹² Although here neither the cultural-critical nor the humanities computing sides are working from a nineteenth-century paradigm, their lagging methodologies make it seem this way. Thus, turning to science fiction and to Gibson’s cyberpunk in particular can provide useful ways of reframing this divide.

Computers in Outer Space: Precyberpunk Machines

Before diving into Gibson’s work, it is also useful to review the science fiction lineage that precedes it, as this overview reveals an interesting shift in our thoughts about computation and digitality. Whether personal, desktop, “bank,” or otherwise, computers provide the fundamental means of creating, accessing, and participating in digital spaces—for those who work in humanities computing and those who study the computers that populate science fiction narratives. Because of the computer’s central role in our understanding of digitality, then, its changing roles in science fiction merit closer examination, but the apparent simplicity of relationships between computers and digitality can also be misleading—in reality, the science fiction computer is not so neatly delineated. It ranges dramatically in form, from the impersonally mechanical, as with the text-generating “Engine” at the Academy of Projectors in Lagado in Jonathan Swift’s Gulliver’s Travels (1726); to a totalizing architectural hive, as with the titular entity in E. M. Forster’s “The Machine Stops” (1909); to the blinking consoles of intergalactic spaceships, such as the vacuum-tubed Maraax in Stanislaw Lem’s The Astronauts (1951); to the punch card–controlled machines of Illium in Kurt Vonnegut’s Player Piano (1952) and Garson Poole’s punch card–controlled heart in Philip K. Dick’s “Electric Ant” (1969); to the militaristic and meddlesome, as with D. F. Jones’s Colossus (1966); to the massive and godlike, as with “Deep Thought,” the city-sized supercomputer meant to calculate an answer to “life, the universe, and everything” in Douglas Adams’s Hitchhiker’s Guide to the Galaxy (1979)—not to mention the Earth itself in the same series, a supercomputer intended to complete Deep Thought’s great work before its untimely destruction.

With this historical sampling in mind, it is worth a reminder that the word “computer” is a term that initially did not refer to machines at all—a fact our students are always rather surprised to learn. Instead, computers were people who “[made] calculations or computations” to support practical or scientific work, and the earliest recorded usages of the word stem from a 1613 religious reader and a 1646 encyclopedia. ¹³ Human computers were involved in fields ranging from eighteenth-century astronomy to World War II–era nuclear fission, NASA projects such as the 1969 moon landing, and the construction of the groundbreaking ENIAC computer. These human computers—often women—interacted with vast amounts of raw data and performed complex, foundational calculations. From its inception, the concept of a “computer” in science fiction and science fact has simultaneously drawn from and connoted a complex relationship with and among bodies, knowledge, and various means of transmitting information.

Thus, there is an incredible diversity of work related to computation in science fiction that precedes Gibson’s cyberpunk. Although it is risky to overgeneralize, three texts from this expansive time period strike especially interesting monuments along the route toward Gibson’s treatment of digitality—and from there, toward what we now call the Digital Humanities. Robert Heinlein’s The Moon Is a Harsh Mistress (1966) follows a set of lunar colonists who rebel against the strictures of government by Earth, aided by the HOLMES IV supercomputer “Mike,” who controls most of the main lunar colony’s resources and develops a sense of humor that makes a revolution sound like fun. Arthur C. Clarke’s 2001: A Space Odyssey (1968) of course (re)introduces the infamous HAL 9000, ¹⁴ who slowly kills off most of his human crewmates but also leans heavily on a series of monolithic computers including the “New Rock” and TMA-1. Finally, Anne McCaffrey’s The Ship Who Sang (1969) follows the trials and travels of Helva, a “brainship” who seeks a human “brawn” to complement her on diplomatic missions. Though these three texts are incredibly different in terms of narrative, they exhibit interesting parallels in their treatment of computer or computer-like figures, and these similarities anticipate Gibson’s later portrayals of digitality in interesting ways.

All three texts, for instance, are somewhat cagey about what a computer is , while at the same time being very certain about what a computer is not . Human protagonists from Heinlein, Clarke, and McCaffrey define and describe their computer counterparts in contrast—but not necessarily in opposition—to themselves. In this model, the humans of 2001 are depicted as “builders” rather than creators or makers, ¹⁵ and Helva is “made” from a disabled fetus and cutting-edge hardware but immediately distinguishes between herself and “you people” without hardware shells. ¹⁶ Meanwhile, computers are posited as “one of Your [God’s] creatures” in Heinlein. ¹⁷ Yet these two apparent “sides,” human and computer, are never positioned as equivalents: that is, one does not move from human to computer, or from computer to human, even though a movement of some kind seems implied. Such uncertainties about what a computer is demonstrate an interesting preoccupation with how the computer “acts” and what it is possible for that computer to do.

Digitality, code, data, and binary signals do, of course, have material roots and components, but they are not tangible in the same way that precyberpunk computers such as Heinlein’s Mike, Clarke’s HAL 9000, and McCaffrey’s Helva are. The fact that these earlier science fiction texts struggle to define computers by their actions and capacities rather than their mechanical components prefigures how Gibson later fuses human and computational entities. Precyberpunk computers complicate boundaries between human and nonhuman in abstract, noncorporeal ways by questioning whether the two can share similar mental, emotional, or even spiritual capacities. But where his predecessors question divisions between human bodies and computer terminals—or between human cognition and computational capacity—Gibson strides across such divisions altogether.

Wares from “The Winter Market”: A Case Study

Since numerous critics (Hayles, Nakamura, Moody, Nixon) have demonstrated how Gibson’s cyberpunk Sprawl trilogy (Neuromancer , Count Zero , and Mona Lisa Overdrive ) provokes concerns about digitality in regard to race, sexuality, embodiment, cognition, and subjectivity, it will be useful to turn to one of Gibson’s lesser-known works as a means of revisiting some of the concerns these critics foreground. Published in Stardate magazine in 1986, “The Winter Market” presents a sprawling, open-air bazaar of garbage, heartbreak, and technological compost and provides a poignant meditation on embodiment, mortality, celebrity, and disability, among other concerns. A close reading of this story brings some of the issues raised by Hayles and others into sharp relief.

“The Winter Market” begins with the memory of a rainy day in Vancouver, where narrator Casey has just learned that his former lover, Lise, has “merged with the net,” a message that propels him to contemplate suicide. ¹⁸ As Casey contemplates jumping into an icy river and ending it all, his actions suggest criticisms about cyberpunk fiction’s disdain for the human body—in this moment, after all, the reference to Lise’s merging with the net suggests the death of a human body, what N. Katherine Hayles equates to her posthuman nightmare. ¹⁹ This revelation of Lise’s “death”-by-net prompts Casey to consider adding his body to the tally, as he stands on “a concrete ledge two meters above midnight.” Eventually, Casey steps away from the ledge, still grappling with the complicated emotional realization that “she was dead, and I’d let her go . . . she was immortal, and I’d helped her get that way . . . I knew she’d phone me, in the morning.” ²⁰ Like Casey’s contradictory realization about Lise—that she is at once dead, alive, and technologically connected—the implications of the human body as an afterthought or fashion accessory thanks to digitization are complicated in “The Winter Market.”

Through Casey’s flashbacks, we meet Lise, a disabled, nerve-damaged woman who cannot move her body of her own volition; instead, a robotic exoskeleton does this for her, powered via a digitally enabled neural jack in the back of her neck. The image of Lise’s human body, at once obscured and animated by the exoskeleton, is eerie, suggestive of technology that threatens to eclipse the human form. By the same token, the story suggests that this technological device allows Lise a greater range of motion than she otherwise would have—her entire capacity for mobility, in fact—and hence that this device provides her with greater autonomy. Yet as Kathryn Allan notes, the disabled body is often figured as aberrant in science fiction and “therefore in need of control by others.” ²¹ This aspect of Lise’s experience is unmistakable, as when Casey recounts Lise’s being “saved” by their mutual friend, an artist named Rubin. When she is found by Rubin, the batteries of Lise’s exoskeleton have died and she lies in an unresponsive heap next to other piles of technological detritus, “back in an alley.” ²² This moment aligns Lise’s unresponsive body with the surrounding trash and coincides with Hayles’s vision of how cyberpunk fiction treats the body: as a discarded accessory, almost indistinguishable from the technological trash surrounding it. Lise is recovered from this pile of refuse against her will—“saved” by Rubin—and her only desire to “live” is as a computer-generated simulation, which seems to confirm Hayles’s and Allan’s readings.

Throughout the story, Lise’s status as machine-dependent human only increases in complexity. Readers see this in the way Lise, once recovered, meets Casey and dares him to take her home, where the two of them “jack, straight across”—a term for sharing each other’s neural input without mediating the resulting sensations. ²³ Jacking straight across would seem to complicate the boundaries of a liberal humanist subject whose boundaries are divinely granted, inviolable, and stable—and because it offers a way to share human experience, the practice would also seem to challenge Hayles’s critique of cybernetics as devaluing embodiment. In “The Winter Market,” however, this moment of jacking straight across is not one of overcoming separation or of sharing sensations to achieve a humanistic synthesis or empathic unity. Instead, it plays out as an artistic battle of the sexes, pitting Lise’s raw emotions against Casey’s inability to cope with them. In the end, Lise wins; after she hits him with pure, single-minded ambition, with “a void that stank of poverty and lovelessness and obscurity,” Casey “couldn’t look at her. I heard her disconnect the optic lead. I heard the exoskeleton creak as it hoisted her up from the futon. Heard it tick demurely as it hauled her into the kitchen for a glass of water. Then I started to cry.” ²⁴ Here emotional “sharing” does occur, but it is violent and one-sided, devastating one of the human subjects involved.

Once this overwhelming shared experience has happened, Casey records Lise’s neural input—from a safe distance this time, barricaded behind studio equipment and already planning to run a sanitized version by his manager as a “product” they can sell ²⁵ —and the resulting reel makes Lise wildly popular, turning her suffering into a commodity that strangers can consume and, by extension, expanding her reach in ways she seems to have always wanted. She becomes a star, able to “dive down deep, down and out, out into Jung’s sea, and bring back—well, dreams.” ²⁶ In terms of Nixon’s critique of sex and gender in cyberpunk, ²⁷ Lise conforms to the complicated formula she outlines. She is tough and astonishingly resilient, but she wields her power in ways that suggest a sexual conquest only achievable through depravity.

“The Winter Market” continues as Casey learns that a newly wealthy Lise bought her freedom, wed herself to the net, and now exists in the form of a neuro-electronic simulation. Freedom, however, is just as complicated as the depiction of the body that she leaves behind, and the notion that her uploaded, merged consciousness continues to perform after the body’s death affirms and confounds cyberpunk’s troubled reputation about embodiment, celebrity, artistry, and mortality. The cost of Lise’s apparent freedom is exorbitant, corroborating Moody’s suggestion that the presence of the somewhat normalized disabled body in cyberpunk is ultimately in service to neoliberal capitalism’s insistence on a hardy workforce. ²⁸ When Casey wonders aloud if Lise will keep creating “dry dreams” from her new transcendent state, Rubin’s response is telling: “you have to edit her next release. Which will almost certainly be soon, because she needs money bad. She’s taking up a lot of ROM on some corporate mainframe, and her share of Kings won’t come close to paying for what they had to do to put her there.” ²⁹

In summary, “The Winter Market” offers an excellent illustration of many features that characterize cyberpunk and many more that occupy the scholarship responding to it. This short story at once complicates a simple attitude toward the human body, offers a devastating critique of digitally enabled consumer culture, and performs an unsettling mode of cultural resistance. It is precisely this form of complex cultural critique that has been essential to the formation of DH.

Not Quite the Singularity, But . . .

Gibson’s work speaks to the emergence of DH as a field in many ways. In addition to the conversations his work has prompted regarding embodiment, cognition, gender, race, and so on, it has also been central to contemporary DH criticism, particularly regarding the form of what Alan Liu has called “transcendental data” and Matthew Kirschenbaum’s demonstration of the surprising durability of digital objects. ³⁰ We now offer a more exploratory and perhaps playful reading of Gibson in terms of DH—we suggest that the narrative backbone of Neuromancer provides a metaphor through which we can view DH’s potential. This reading emerges from the plotline of the distinct artificial intelligences Wintermute and Neuromancer, who, at the novel’s end, merge into a single presence or singularity.

In the novel, the Wintermute is characterized by his attention to detail, logic, forecasting, and computation. His counterpart, Neuromancer, is characterized by improvisation, creativity, and personality. Both are strong subjective entities, but for the majority of their existence they have been kept apart, each in ignorance of the other. By the end of Neuromancer , the human protagonists have successfully hacked into the information structure that will allow the AIs to merge, and the resulting merger has appealing consequences: (1) the emergence of a superintelligence whose potential for autonomous action is greater than the sum of its parts, and (2) a redistribution of wealth that serves as a sort of justice for the characters who have been maimed or otherwise compromised in their quest.

Like Wintermute, the field of humanities computing is logical, statistical, and rigorous in terms of its methodology. Like Neuromancer, the larger purview of cultural criticism and making is more creative, improvisational, and concerned with aesthetics. Although the two carried on for decades without formal acknowledgment of the other’s activities, once they were collected under the aegis of DH’s “big tent,” they were given an opportunity to become, like the merged AI at the end of Neuromancer , greater than the sum of their parts. When the two disciplines converse and collaborate, they complement each other, as can be seen in the realms of electronic literature, digital art, and more. Femtechnet, HASTAC, and the ELO are just some of the many entities that have resulted from mergers between humanities computing and DH.

A DH Concordance

Throughout this essay we have followed a fairly one-sided form of argumentation to make our case, that of traditional cultural-literary analysis complete with textual close reading. However, in a chapter that calls for the integration of cultural studies and data analytics, as this one does, such an approach is insufficient by itself. We close our essay by sharing our process of computational textual analysis to see how the concerns we have outlined might also be traced quantitatively. Our corpus draws from the following areas: precyberpunk science fiction, particularly the three texts analyzed here for their treatment of embodied computational entities; cyberpunk fiction, with Gibson’s work as the dominant sample; DH scholarship, which draws from a variety of DH publications; and science fiction scholarship from Science Fiction Studies , beginning with the journal’s inception in 1973 to 2017. ³¹ In all, we are looking at more than four thousand distinct documents, with the bulk coming from DH and science fiction scholarship. This was a massive undertaking, and we are still working to parse the wealth of data; what follows are but a few instructive results.

Figure 3: Instances of “digital” and “computer” in issues of Science Fiction Studies in the 1970s. Courtesy of Maria Alberto and Elizabeth Swanstrom.

Our primary interest was the shift between “computation” and “digital” that signifies the emergence of DH as a field. We started by searching for the words “digitality” and “computer” across the collected SFS issues. We began with the cat command from terminal, which combined all the separate issues and articles into one file, and then used Python’s NLTK to search for these terms. Our hunch going in was that both words would appear more frequently from the 1980s onward, but our findings were surprising. In the 1970s, there are zero instances of the word “digital” and fifty-eight of the word “computer.” Figure 3 shows how this count appears as a part of a larger search for concordances.

In the 1980s, things begin to change: we now have five instances of “digital” and eighty-six of “computer.” In the 1990s, both terms take off, with 61 occurrences of “digital” and 315 of “computer.” In the 2000s, there is a sizable jump in “digital,” whereas “computer” stays fairly stable, with 287 instances of the former and 338 of the later. In the 2010s, another leap occurs, this time with “digital” barely increasing to 295 counts and “computer” seeing more than a 50 percent increase with 582 occurrences.

Figure 4: Lexical dispersion plot of instances of “digital” and “computing” in Science Fiction Studies through 2017. Courtesy of Maria Alberto and Elizabeth Swanstrom.

By using NLTK to create a lexical dispersion plot, we can see how these terms track over time. In Figure 4 “each stripe represents an instance of a word, and each row represents the entire text” (nltk.org). In other words, the greater the density of stripes, the more frequently the word is used in that section of the text. Here we demonstrate with a single file created by merging every available issue of Science Fiction Studies through 2017.

On some level this shift is intuitive, but on another, the fact that digitization is not even conceived of as a synonym for computation until the 1980s is fascinating. The lag between usage is much longer—and wider—than we thought it would be. In addition, since the term “digital” replaces “computing” in the humanities computing/DH shift in 2002, precisely because it lacks the same mathematical, calculating connotation of “computing,” our analysis of the SFS corpus reiterates that this shift makes sense. Conceptions of “digital,” “digitality,” and “digitization” in science fiction scholarship simply do not have the same quantitative baggage that “computer,” “computing,” and “computation” do. When the “digital” as a concept comes into play, it does so in science fiction scholarship—and possibly in science fiction itself—in conjunction with Gibson’s vision of the term in his cyberpunk fiction, as well as the new genre’s preoccupations with embodiment, human–machine networks, and expansions of subjectivity. Another dispersion plot of all SFS issues, this time including Gibson’s name and the name of this important subgenre, clarifies this point (see Figure 5 ).

Figure 5: Lexical dispersion plot of instances of “Gibson” and “cyberpunk” in Science Fiction Studies through 2017. Courtesy of Maria Alberto and Elizabeth Swanstrom.

In this case, the name Gibson and the term “cyberpunk” have nearly equivalent distribution. There is a slight overlap between “computer” and Gibson, but none between “computer” and “cyberpunk.” Likewise, the occurrence of the word “digital” precedes both, but only slightly. Again, this suggests an intriguing distinction in the connotations of both terms. Although this result stems from only one corpus, we believe that pursuing the implications of this distinction would be worth further exploration in both science fiction and DH scholarship.

We were also interested in the topics highlighted by our close reading of “The Winter Market”—gender, disability, embodiment, and so on. For each topic we created a list of keywords, which we searched for using Voyant, a free online tool for text analysis, and Mallet, a tool for topic modeling. Before testing our own keyword associations, we created a topic model, which reaches across multiple texts to find patterns of co-occurrence and pools them to predict how words relate to each other across the corpus as a whole. In our topic model of the complete contents of Science Fiction Studies , for instance, the following cluster surfaced: “cyberpunk postmodern feminist gender gibson body stories cyborg texts.” At first glance this cluster seems nonsensical—and it is, at least syntactically. This is to be expected, though: topics aren’t meant to be read as sentences but as words that co-occur enough times to make the relation potentially meaningful. When we created a visualization of the actual frequencies of these words with Voyant, we saw that Gibson’s name is visible but not equal in size to these other terms (Figure 6 ). When we added the names of his contemporaries to this search, however, we saw his name rise in comparative prominence (Figure 7 ).

Although we do not suggest a causal relation between Gibson and these larger concepts—a different lexical dispersion plot, not included here, shows that they both predate and postdate cyberpunk—these visualizations do suggest that his work was of special interest to science fiction scholars thinking about these issues.

Another topic that intrigued us was the depiction of human–computer mergers in classic works of science fiction as compared with cyberpunk. Our hunch was that the concept of such a “system” would be distinct in our corpora, and we wondered if there would be a visible pattern shift over time. We created a list of keywords that surrounded the concept of “system” and compared the frequency of their appearance in our sets of texts. We used the three works of classic science fiction referenced earlier—Heinlein’s Moon , Clarke’s 2001 , and McCaffrey’s Ship —as points of contrast from our other corpora, and what we found was illuminating. Again, the term “computer” was significant—much more prominent in these classic science fiction works and the DH texts, while largely absent in both Gibson’s fiction and the SFS corpus (see Figure 8 ).

Figure 6: Science Fiction Studies word cluster without Gibson’s contemporaries. Courtesy of Maria Alberto and Elizabeth Swanstrom.

Figure 7: Science Fiction Studies word cluster with Gibson’s contemporaries. Courtesy of Maria Alberto and Elizabeth Swanstrom.

Figure 8: Word cluster variations of the Science Fiction Studies corpus. Courtesy of Maria Alberto and Elizabeth Swanstrom.

Again, although this is only one approach, we believe our findings demonstrate that pursuing the implications of this distinction would be worth further exploration in science fiction and DH scholarship.

Neither of us is formally trained in computational linguistics, so we know full well that our findings are fairly shallow. At this stage there are only two of us working on this project, which involves considerable numbers of files and amounts of data. In addition, Mallet confirmed our keywords in some cases but not others—a phenomenon that may reshape our results or conclusions after further study. Finally, we used “The Winter Market” as our representative sample of Gibson’s work to generate topics of interest and the keywords characterizing them, but we acknowledge that choosing this particular text to work from was a largely subjective process.

Despite these shortcomings and snags, we hope that our findings are suggestive. We also hope that by undertaking this qualitative approach in tandem with cultural analysis, we can help familiarize new audiences with DH methodologies in a way that does not overwhelm them with data but offers a user-friendly point of departure. We know we have only scratched the surface of what such an analysis might yield, and we invite anyone with an interest to build on our work. We provide a more robust breakdown of our data, complete with a variety of annotations and instructions, on our project’s webpage (http://sf-and-dh.org ). In the spirit of the interests and questions detailed, we warmly welcome collaboration.