The Program Era

Track Changes: A Literary History of Word Processing BY Matthew G. Kirschenbaum. Belknap Press. . $29.

I can’t remember the last time I used an electric typewriter. It most likely would have been in the course of typing out an address on an envelope—but then again, I can’t readily call to mind the last time I did that with anything other than that old-fashioned technology, the ballpoint pen, which itself is not really all that old school. The mass commercial distribution of the ballpoint pen in the United States dates only to about 1945, which means its triumphal appearance in the writing market occurred just under twenty years before that of the Magnetic Tape Selectric Typewriter, IBM’s radically rethought typewriting device. Released in 1964, the MT/ST was the first machine of its kind, equipped with a magnetic-tape memory component that allowed you to edit text before it was actually printed on the page. Corporations were considered the primary beneficiaries of the new technology, a wrinkle on the electric typewriter that arrived with considerable media enthusiasm. The makers of the MT/ST saw the contemporary office groaning under the weight of metastasizing paperwork and envisioned making money off companies hoping to streamline the costs of secretarial labor and increase productivity. Writers were something of an afterthought: Whatever effect IBM’s product would have on authors—high or low, commercial or experimental—was collateral.

But if the introduction of a new type of word-processing machine started a slow-burning revolution in how writers went about their business, it was a revolution nonetheless, drastically altering how authors did their work. The primary focus of Matthew G. Kirschenbaum’s new history of word processing, Track Changes, is a twenty-year span, from the moment that IBM brought out the MT/ST until 1984, when the Apple Macintosh first offered a glimpse of an unchained future with its televised appeal to a nation of would-be Winston Smiths. (As with Bobby Thomson’s home run, millions still claim they remember exactly where they were when they saw Ridley Scott’s celebrated commercial for Apple during the third quarter of an otherwise forgettable Los Angeles Raiders Super Bowl blowout.) The word-processing program that the new Mac included, MacWrite, was fairly primitive—it couldn’t handle documents longer than eight pages, a boon only to the briefest of short-story writers—and it would take years for the mousy point-and-click innovations to knock off the market Goliaths of WordStar and WordPerfect. But the timing of Apple’s campaign couldn’t have been better. “In 1978 or 1979,” Kirschenbaum notes, “writers using a word processor or a personal computer were part of the vanguard. By 1983 or 1984 they were merely part of the zeitgeist.”

As Kirschenbaum’s history reminds us, the story of personal computers supplanting older systems dedicated to word processing—and writers’ larger commitment to abandoning pens and ink and typewriter ribbons and correction fluid—was hardly the fait accompli that we sometimes think it was. His book attempts a full literary history of this shift. To do so, he ranges across a number of phenomena: the technical and managerial prehistories of the word-processing revolution; the imaginative, sometimes allegorical literary responses to how work was managed (from Stanislaw Lem’s 1971 “U-Write-It,” which fantasized a fully automated literary production line, to John Updike’s 1983 poem “INVALID.KEYSTROKE,” a sort of ode to the little dot that appeared on the screen between words in early word processors like his own Wangwriter II); and most prominently, how word processing both tapped into and reflected writers’ anxieties about their whole enterprise. The last didn’t appear with the first wizardly word processor or dazzling software program, and it hasn’t gone away. What Kirschenbaum doesn’t do is reflect on how the “program era” affected authors’ sentence structure, book length, and the like. Track Changes is less concerned with big data than with bit-by-bit change.

A detail from a 1982 Radio Shack advertisement showing Isaac Asimov and a TRS-80 computer. Radio Shack

The technological history of word processing is indeed a series of changes, some rapid, others lumbering, in hardware, software, operating systems, and data structures. The early word-processing environment wasn’t Windows or Mac or even DOS (introduced by Microsoft in 1981) but CP/M, which powered such popular pioneering machines as the Osborne 1 (also released in 1981) and the Kaypro II (1982) and provided the programming “platform” of WordStar, released in 1979. Hard drives were virtually nonexistent, and disks were truly floppy. Spell-check, thesauri, and built-in dictionaries were a thing of the future. Compared to a laptop, the screens were dinky (but compared to an iPhone, not overly so?). The ease of using devices has changed as well. Learning how to operate earlier systems took diligence, with coded combinations of keys that allowed users to manipulate chunks of text. Plus, the various systems weren’t mutually compatible. If you bought a Kaypro or a Tandy or a Commodore, you were stuck with the limitations (or enjoyed the advantages) of the particular product in ways that seem impossible to imagine for anyone using a laptop today. When Charles Bukowski wrote a poem titled “16-bit Intel 8088 chip,” he observed: “both Kaypro and Osborne computers use / the CP/M operating system / but can’t read each other’s / handwriting / for they format (write / on) discs in different / ways.”

Yet the first generation of word-processing systems attracted a legion of proselytizing adopters. Kirschenbaum includes a roster of acolytes—from Michael Chabon, Arthur C. Clarke, and Ralph Ellison to Anne Rice and Eve Kosofsky Sedgwick—whose motley makeup gives a sense of how the appeal of word processing crossed genre boundaries. Many were loyal to specific programs. George R. R. Martin remains a user of WordStar to this day, and William F. Buckley never abandoned it. (He once opined to Time magazine, “I’m told there are better programs, but I’m also told there are better alphabets.”) The poet Lucille Clifton relied on a Philips/Magnavox VideoWRITER 250 that she purchased in 1988 (it had retailed for $800 when it appeared in 1985), a bulky machine with an unusually wide screen, displaying one hundred characters rather than the more-typical eighty or fewer. John McPhee had customized a set of programs that run off an old IBM-compatible Kedit, magnificently designed to combine or isolate various files of notes. Yet writers were affected by these programs in surprising ways. When Jacques Derrida gave up his Olivetti electric typewriter in 1986 for what he called “le petit Mac,” he learned to respond to a program that alerted him when his paragraphs were “too long,” “submitting . . . to an arbitrary rule made by a program I hadn’t chosen.”

What made word-processing devices much more than just souped-up typewriters was not only that they gave you the ability to edit at the same time that you wrote, or that they eliminated or seriously curtailed the effort of correcting from typewritten pages. Seeing text revealed on a screen, even in the technologically costive form offered by the earliest word processors, provided an unprecedented opportunity to picture the manuscript as a whole and with an immediacy that typewriting didn’t permit. The acronym WYSIWYG—What You See Is What You Get—delivered perhaps the same frisson for writers in the early 1980s as Frank Stella’s “What You See Is What You See” had to ambitious painters a generation earlier. Even the intricate system of keyboard commands required to move passages or insert italics or signal word breaks seemed akin to the freedom of writing in longhand, the fingers never leaving the keyboard, with none of the “mechanical” intrusion of typing and retyping manuscripts: It combined the “efficiency of typing with a hands-on, no-nonsense approach to really handling a manuscript—almost as if the writer was somehow ruffling the electronic pages, marking them up and blocking off passages, sorting them into piles and flagging them with bits of scrap paper or colored ribbon or bubble gum wrappers.”

Plenty of writers balked at the joys of word processing, for a host of reasons. Overwriting, in their view, became too easy; the labor of revision became undervalued, and noisy printers and plugged-in gadgets the norm. When Gore Vidal wrote in the mid-1980s that the “word processor is erasing literature,” he expressed an uneasiness about technology’s proximity to creative writing (and the wider field of publishing) that persists. This too is the literary history of word processing, a snapshot of dread about gadget love, the seduction of the screen, and automation and the threats they pose to writers. This dread has taken various forms over the years. It lurks in the background of Sven Birkerts’s late-1990s jeremiads against the Web and Leon Wieseltier’s 2015 diatribe about what distraction was doing to the contemplative mind (and, by extension, the writer). Paradoxically, much of the hand-wringing over digital-era distraction as a mortal enemy to thinking has given rise to apps: Scrivener, or WriteRoom, or Write or Die, the last an online editor that promotes concentration by erasing your text if you pause too long between keystrokes.

The word erasure is a trendy survivor from the hard-copy era, but it points to a lot of what was worrisome to many about the rise of word processing—and remains so still. (If you doubt it, just mention an app like Write or Die to almost any writer and watch the terrified reaction inspired by the thought of losing work.) But Kirschenbaum’s history also relates a more profound erasure, that of the unseen hands, mostly women’s, the typists and office workers whose value was an intentional casualty of the revolution. There’s a good bit of ironic justice, then, in his nomination for the first novel written on a word processor: the British writer Len Deighton’s 1970 Bomber, the execution of which began sometime in 1968 on a European version of the MT/ST mastered by Deighton’s assistant, Ellenor Handley. Deighton did the laborious work of constructing a complex narrative from a typescript; Handley did the laborious work of making his scissors-and-paste method of integrating cross-cutting narratives obsolete. The first book, perhaps, to be created on a word processor was hence a collaborative production. The words “indisputably belong to Len Deighton. But the hands that recorded and processed them using the MT/ST belong to Ellenor Handley.”


Eric Banks is the director of the New York Institute for the Humanities at NYU.