What does a history of Chinese computers say about the globalization of computing technologies? Thomas Mullaney’s new book picks up from where he left off in The Chinese Typewriter: A History (MIT Press, 2017) and completes the long story of Chinese “typing” by introducing previously unexamined figures and machines from the 1940s onward. At the center of this book is what Mullaney calls the Latin-centric bias of Western computer technologies, exemplified by the QWERTY keyboard, which assumes that the simplicity of typing English and Latin-scripted languages is by nature the easiest, fastest, and most efficient method. “What could be simpler,” asks Mullaney rhetorically, “than the ‘immediacy’ of ‘Q equals Q,’ ‘W equals W,’ and so on?” (p. 9) This common practice of “what-you-type-is-what-you-get” has occluded the Western vision, according to Mullaney, and prevented the West from seeing that the imagined superiority of QWERTY has already been overthrown by Chinese computational methods of inputting characters. Situating this technical paradigm shift at the heart of the book, Mullaney argues that the world has already entered a new age of writing: Hypography.
Hypography is an entirely new beast in the world history of writing, for it operates below (hypo) the level of writing (graphy). The conventional world of writing, proposes Mullaney, functioned at the level of orthography and “orthographs,” such as cuneiform, hieroglyphs, and the alphabet, i.e., physical signs that were inscribed on various surfaces and that signified concepts. In contrast, the world of hypography operates through the digital mechanisms of search and retrieval that secretly work beneath those surfaces. Pressing “Q” on a keyboard for Chinese input, for instance, gives a whole range of possibilities (e.g., 請清祺七 etc.) that rely on the deeper statistical mechanisms of search and retrieval that help the users type significantly faster. In hypography, input and output are not directly correlated, and what-you-type-is-what-you-get is never the working paradigm. Hypography, in short, is a novel way of writing that computationally functions through searching and retrieving orthographs. As Mullaney vividly demonstrates, Chinese input methods (IMEs) offer unique examples for it.
Chronologically arranged, the chapters are each structured around different input methods and their inventors, and the narrative is often embellished by the author’s own anecdotes that turn the dense technicalities of input methods into lively stories. The first chapter is centered around Lois Lew, the woman typist who made famous Chung-Chin Kao’s IBM-funded electric typewriter in 1947. As opposed to present-day input methods, Kao’s typewriter had only one way to input characters: four-digit codes, as in a Chinese telegraph codebook. Lois Lew’s job was thus extremely difficult, since she had to memorize the four-digit numbers for 5400 characters, herself becoming the IME. While the typewriter did not result in the technological breakthrough that Kao had hoped for, Lew’s story offers a charming start to the history of input.
Input methods have come a long way since then. The second chapter turns to Sinotype, a computational Chinese phototypesetting system invented in the 1950s by MIT’s Samuel Hawks Caldwell. Caldwell was one of many Westerners fascinated by Chinese while lacking any knowledge of it. Luckily, his student Francis Fan Lee and the Harvard historian Lien-Sheng Yang showed him the basics of Chinese writing and the three men came up with the first hypographic interface that used alphabetic letters to input Chinese characters. This interface assigned certain strokes to the letters on a keyboard: B was a heng 一, G was a pie 丿, A was a mo 末, and so on. Building on this method, Caldwell and his team’s efforts produced two critical devices for Chinese input that were the precursors of present-day technologies. First was the pop-up menu, which showed and allowed the user to choose among different characters that were “spelled” similarly, such as BDB, which gave shi 士 as well as tu 土. The second device was autocompletion, meaning the possibility of inputting a minimal number of letters/strokes to retrieve a character composed of a higher number of strokes. Sinotype, in short, was the first hypographic system of typing Chinese, despite its eventual fall into oblivion due to lack of funds and Caldwell’s untimely death.
The next two chapters describe the rise and fall of unique keyboard designs for Chinese. The third chapter expands the history of input by examining large non-QWERTY machines for Chinese with hundreds of keys, which were invented in the 1960s and 1970s in the US, Taiwan, China, and even Cambridge, UK. These medium- and large-sized keyboards, explains Mullaney, disappeared in the 1980s, when personal computers once again began to change the landscape of Chinese input. The fourth chapter then describes the “input wars” of that decade, with hundreds of different inventors struggling to “spell” Chinese via the QWERTY keyboard, not unlike Caldwell. Previously little-explored figures populate these pages, such as Zhi Bingyi, who while imprisoned during the Cultural Revolution spent his days thinking of ways to input characters via alphabetic letters. His difficult years paid off later, when the West German Olympia Werke company used Zhi’s input to build a Chinese word processor.
The fifth chapter examines “modding,” i.e., modifying the Latin-centric peripheral computing technologies to make them functional for Chinese, such as computer memory, dot-matrix printers, and screens. Since all this peripheral equipment was invented with the Latin alphabet in mind, Mullaney argues that each kind represented the deep alphabetic bias of Western computing, sometimes quite literally built into the “metallurgical depths of the alphabetic order” (p. 168). None of these types of equipment initially worked for Chinese. Mullaney thus explains the innovative solutions of the engineers who modified the technologies, thereby “hacking” the Latin bias of Western computing and finding “inroads into an exclusionary global information order” (p. 157). These engineers were not copy-cats, Mullaney argues; they were the “interrupters” of a Latin-centric modernity.
In fact, these interrupters subsequently even surpassed that modernity, according to Mullaney. The hypographic possibilities reached their zenith with Hanyu Pinyin as an input method from the 1980s onward, as he describes in the final chapter. Mullaney’s description of Pinyin as a hypographic input technology is compelling, since Pinyin-as-input harnesses the power of phonetic transcription, autocompletion, and predictive text in which the full pronunciation of words rarely needs to be entered to display characters on the screen. Typing BJ is sufficient for Beijing; LHG for United Nations (lianheguo). And herein lies the central technical observation of the book: while QWERTY for Chinese can speedily output multiple characters with single-letter input, QWERTY for English is not capable of doing the same (typing SYL does not output “see you later” in a word processor). Pinyin-as-input does come with Mandarin-centrism, but this does not seem to be an issue for Mullaney, since it enables Chinese typists to achieve high speeds, finally overthrowing Latin-centric supremacy through embracing the hypographic potential of digital input technologies.
The book ends with a critique of QWERTY and goes beyond Chinese to claim that all non-Latin scripts have indeed embraced hypography, from Arabic to Korean to Devanagari. “It would not be a stretch to say that the continued growth of computing and new media owes much of its fortune to this frenzy of hypographic ‘modding’ in the non-Western and non-Latin world,” writes Mullaney in the final pages (p. 230). According to him, non-Western hypography has hacked and made obsolete the biased assumptions embedded in Latin-centric computation. It has thus “saved” the computer by enabling it to become a global technology.
The author’s archaeology of input methods, as well as his journalistic style, make this book an engaging read, but readers anticipating a broader history of computers in China and the world (per the book’s title) might be disappointed. There are indeed plenty of places in the book that would have benefited from a historically and conceptually deeper analysis. For instance, Mullaney mentions in passing (p. 91) the Taiwanese military interest in computers in the late 1960s, with no explanation for why and how the Taiwanese army developed this interest at the height of the Cold War. Similarly, the several pages devoted to describing the development of textual processing technologies in the PRC in the 1970s (pp. 100–110) do little to help readers rethink the final years of the Cultural Revolution or question the assumed historic break between the Maoist and post-Maoist years. The emphasis on the heroic and colorful personalities of input makers seems to reflect our contemporary moment of idolizing computer pioneers, and it obscures the other forces at play, such as the state. In fact, neither the PRC nor the ROC government plays any significant role in Mullaney’s account, even though one suspects that the development of a national computer industry and state capacity were mutually reinforcing in both countries from the 1970s onward.
Such omissions were likely strategic rather than accidental. Unlike many histories of computing, this book is not about data management, cybernetic governance, military prowess, intellectual property, knowledge economy, or surveillance technologies. It has a single goal: to demonstrate the triumph of Chinese typing over Western typing. Mullaney indeed succeeds in showing that the additional layers of mediation in Chinese input eventually increased the speed of inputting characters, leaving QWERTY in the dust. This narrow focus on typing and input, however, comes with problems, for Mullaney reduces the “global information age” to QWERTY and blurs the distinction between typing and computing. He further portrays QWERTY as the Western computer’s imperial claim to universalism, even though one could make a strong case that computational universalism relies on the binary code, not the alphabetic keyboard. Regardless, given that QWERTY is synonymous with Western imperialism in this book, “hypography” turns out to be the anti-imperialist savior of the non-Western world, which raises some major issues.
The first issue is the politics of technology. Mullaney repeatedly notes that computerization, from character encoding to monitors to printers, has been biased towards the Latin alphabet, resulting in “the story of one digital ‘Chinese exclusion act’ after the next” (p. 18). While provocative, this portrayal of the West versus China conflates political bias with technological limitation. Every technology embeds the social and political codes of its circumstances, as historians have long argued. When a camera only recognizes white skin, for instance, it reflects the deeper structural biases of a racially segregated society. But not every technological limitation or exclusivity implies a political bias. The Chinese brush is not suitable for writing alphabetic letters; it is hardly biased against the alphabet. Similarly, ASCII, dot-matrix printers, computer screens, and so on, were certainly exclusive mediums, since they were initially invented for the Latin letters of the English alphabet, but it would be a stretch to claim that this Latin-centrism manifested a bias against Devanagari, Hebrew, Arabic, or Chinese. It instead posed technological limitations that disadvantaged many non-Latin scripts, some more than others.
This brings me to the second issue. Does overcoming the limitations of a given technology constitute “hacking” the hegemonic order? As Mullaney lucidly demonstrates, hardware and software were both big hurdles to overcome for Chinese engineers. IMEs, printers, screens, keyboards, and all other peripherals were in dire need of adjustment to enter the global information age. It is indeed fascinating to see the innovative dynamism of Chinese engineers. But they were not alone. Western companies were also aware of the limitations of Latin-centric computers and were trying hard to overcome them. Xerox, Apple, IBM, Microsoft, and others were investing resources into breaking the Latin-centrism of computer technologies, especially from the 1980s onward, with the primary goal of increasing corporate profits by expanding into the world market, including East Asia. Designing hardware and software for non-Latin scripts was in fact critical for the emergent global digital order, and that design effort was partially spearheaded by what we now call Big Tech. “Modding” in that regard aided the new order rather than challenging it, for going beyond QWERTY, one could argue, was the new age of corporate empire. Describing the global information age as a tug-of-war between the West (QWERTY) and the non-West (hypography) thus becomes misleading. To put it differently, by reducing the complexity of the global information age to the technical limitations of QWERTY, Mullaney conveniently bypasses the significance of larger forces such as global corporations, digital governance, increased datafication, information capitalism, and so on, leaving the reader with a simplistic narrative of non-Western triumph over the West. Modifying the extant hardware is certainly an important story, but the enthusiastic vocabulary of “hacking” is inaccurate.
Thirdly, Mullaney makes a big jump in moving from the particular case of Chinese hypography to the larger case of non-Western hypography. The description of Chinese typing as a hypographic practice is an original contribution to the literature and is beautifully examined by the author, but Chinese writing is special in this regard, because no other writing system (except Japanese and Korean to a certain extent) uses the possibilities afforded by hypography as much as Chinese does. In other words, given the particularities of Chinese, such as the number of homophonic characters and the frequent use of combined characters, the Chinese case is not comparable to other non-Western scripts. Arabic is a good example. “While still imperfect,” writes Mullaney, “Arabic is for the first time appearing on the screen and the printed page in its correct, connected forms [thanks to hypography]” (p. 230). This is simply untrue, for the definition of “correct, connected forms” has long been a contested political topic, and the appearance of Arabic on the screen, which has been happening at least since the 1980s, has received criticism from users and experts alike, as it continues to reproduce certain mistakes that go back to nineteenth-century Arabic typography.Footnote 1 Fixing, modding, or hacking it is also not easy, as this “hypographic” system is tied to larger structural and political forces, such as the development of font technologies, global information protocols, and governmental decisions. While I would like to agree with Mullaney’s suggestion that hypography is a new stage in the world history of writing, I find it an exaggeration to claim that non-Western hypography “saved” the Western computer in any meaningful sense of the term. It might instead be more fruitful to explore the global politics of hypography itself by examining the myriad forces that structure the development of non-Latin scripts.
These criticisms are not meant to deny the significant value of this work. The author’s decision to put the Chinese engineers at the center of computational innovation is commendable, as it dispels long-held assumptions about the absence of innovation in China. In addition, the intriguing conceptual framework of hypography will be useful for historians working on non-Western scripts. Overall, the author’s long years of research, unique archival finds, pleasant prose, and thought-provoking arguments bring Chinese history into conversation with a growing literature on the global history of computers. This book will long serve as a point of reference for future histories of computing in East Asia.
