Pieces That A Machine Can Comprehend

Thomas S. Mullaney, The Chinese Computer: A Global History of the Information Age. The MIT Press, 2024. 376 pp. Hardcover $34.95.

For me, a digital native who grew up typing Chinese on a personal computer since the age of seven, writing the language is almost a no-brainer: you type on a QWERTY keyboard using Hanyu Pinyin, a Romanized system taught in every elementary school to learn Chinese characters through their phonetic pronunciations. A Hanyu Pinyin-based input software is usually built-in (like Microsoft IME) or can be installed with a few clicks (QQ, Sougou, among many others).1 I never gave it much thought until I was asked by someone in the U.S. about a “Chinese keyboard.” “You must be using a special keyboard, right? You have thousands of characters!”

I almost laughed at the seemingly silly idea of a “Chinese keyboard” and the possibility of its existence. At the same time, as a Chinese native, I had never seriously considered this intriguing question: how did inventors and innovators come up with a solution that can “translate” a completely different writing system with hardware designed for Latin languages? If this question is not curious enough, Thomas S. Mullaney’s book, The Chinese Computer: A Global History of the Information Age tells a far more fascinating and winding story of this technological venture, featuring cultural and political forces from China, the U.S. and Europe on a global stage. The book, carrying a theme closely related to Mullaney’s influential monograph The Chinese Typewriter: A History (2018), explores how the contemporary Chinese computerized input method developed. It reveals how different computer inventors and innovators tackled the problem of Chinese input on a “Western” computer by designing mechanisms to let the computer “understand” and display the language. Mullaney, a scholar of Chinese history, investigates how such human-computer interaction evolved with different methods of key depressing, character formation, and word prediction. This process involves breaking down Chinese characters—or even the Chinese language—into pieces that a machine can comprehend and build up, which sometimes challenges conventional perceptions of a language with thousands of years of cultural significance and evolution. I am genuinely fascinated by Mullaney’s dive into this rich history, and amazed by how this book shows the details of Chinese input. From a historian’s point of view, however, I was a little lost in the book’s details and struggled to find a strong argument.

Using records from multiple archives, from public museums of science and technology as well as corporate archives located in China, the U.S., and Europe, the book’s research provides concrete details of how various Chinese input methods took shape and were developed. Mullaney also consulted product manuals, publications in both English and Chinese, and conducted oral histories to complement the archival research, especially on technical details of how certain input methods work.

The Chinese Computer is organized into six chapters, with an introduction and conclusion, and operates roughly in chronological order but each era carries a distinct theme and different themes overlap in terms of historical time period. After the introduction, chapter one tells the emergence of a Chinese input method heralded by IBM in the early days of the computer age, using a coding system like telegraph, with a story of a female typist, Lois Lew, who managed to memorize codes for thousands of characters. Chapter two explores the Sinotype method, and the innovative features of “predicting” the Chinese text.2 The third chapter expands the scope into the hardware space, detailing a few inventions that tried to break the QWERTY default and design a “Chinese keyboard” (spoiler: they all failed). “Input Wars,” the fourth chapter, focuses on several players trying different ways to dissect the composition of Chinese characters and optimize the input in a competitive market scene. The penultimate chapter explores the ancillary devices aiding the display and printing of Chinese characters, while the final chapter circles back to the idea of predictive text in using the popular Hanyu Pinyin as the base of input method.

Mullaney argues that the concept of hypography (the type of technical writing used to input Chinese characters into a computer) is the key to solving the puzzle of Chinese input, and enables the computer to achieve a meaningful presence outside the Americas and Europe.3 In the book’s introduction, he introduces this concept by showcasing a winner, named Huang Zhenyu, in a Chinese typing competition who, at an astonishing speed, typed a completely bizarre sequence of letters on a QWERTY keyboard to generate a long paragraph of text. The input of the letters might make sense to a human to some extent, or not at all, but it is essentially a message for the computer to retrieve the correct letter stored in memory. In other words, the input method itself does not construct Chinese characters but serves as an encrypted “code” for the computer to understand. The coding and decoding process is infinite; some methods resemble how Chinese is learned and understood by humans such as the Hanyu Pinyin input method; or the competing ways of dissecting characters and locating them on the keyboard, while others require memorizing thousands of four-digit codes, similar to the earlier Chinese telegraph system.

The concept of hypography and its evolving iterations encapsulates the technological characteristics and the process of fitting Chinese into the technical and societal landscape of computing. The speed of typing, aimed at matching Anglophone typing, was a core pursuit of many inventors and innovators. By adopting a hypographic method, a computer user can exploit redundancy in Chinese typing and scripts, “reducing time by adding steps.”(85) This was not a top-down design but rather a localized endeavor of tinkering, modding, and hacking, as Mullaney depicts in chapter five. The development of human-machine interaction is also a product of its socio-technical context, with various global powers, such as the U.S., the U.K. and China, carrying their aspirations of making Chinese computers, whether for national security, market dominance, or cultural significance.

One question Mullaney does not give a sufficient answer to is why some input methods succeeded while others were forgotten. The history of technology does not offer definitive answers but presents opportunities, contingencies, and a lineage of thoughts and experiments. There are hints about the “path-dependency” of technological innovation. In the book’s third and fourth chapters, Mullaney excavates the historical records of various Chinese keyboard designs, noting that a completely new system of computers with custom-built hardware and software was less viable despite Chinese political backing.4

Compared to his previous book, The Chinese Typewriter, where Mullaney eloquently told the story of maintaining the cultural identity of the Chinese language through mechanical innovations in the wake of globalization, this book has less certainty in its narrative. Telling human stories is one of Mullaney’s strong suits, and in a history filled with many players, competitors, and contenders—including IBM, MIT, the CIA, the Pentagon, and the Chinese government and academia—he makes great efforts to elucidate details and exchanges between them using historical records, complemented with abundant oral histories. Yet there seems to be a lack of a backbone connecting these human stories to the technical and social trends. What are the influences of the competing political and economic powers? Did they have their own agendas, and how would their agenda impact the contemporary landscape of Chinese computers? Although mentioned in several places, the story also seems somewhat detached from the global history of political economy from the Cold War to the Information Age, and does not fully explore the impact of political trends and market initiatives.

Another component largely missing from the book are the users. Previously in The Chinese Typewriter, Mullaney depicted not only the typewriters but also the typers’ embodied experiences and practices with physical objects, which in turn impacted technological development. Early in this book, Mullaney dedicates a considerable amount of space to the story of the typist Lois Lew (featured on the book’s cover), and concludes that “How much code can a person stand?…far more than the gatekeepers of modern information technology were willing or perhaps able to accept.”(60) This statement highlights that human-computer interaction involves human cognition and adaptation in certain social settings, which can, in certain circumstances, feed back into the technological system, as seen in the word prediction algorithms mentioned in chapter six. Beyond this, however, The Chinese Computer does not pay much attention to users. This of course can be a challenging task, as users of personal computers now encompass an unlimited range of occupations, communities, social classes, and people, each posing great challenges in historical writing. No single definition or representation of users exists. Further research on how school educational programs or mass media address the learning and use of IMEs (based on Hanyu Pinyin) and Wubi (based on Chinese character strokes) would be fascinating to read.

Mullaney might also draw inspiration from critical media studies and digital sociology/anthropology to examine phenomena around input methods and systems. In particular, linguist Gretchen McCulloch’s work in Because Internet: Understanding the New Rules of Language examines how people “invent” languages with online technologies. For example, an interesting aspect of the hypography of Chinese input is how people circumvent censorship by using input shortcuts and homophones, part of the predictive feature of input software. It would be illuminating to see the living and evolving nature of technology until user preferences and practices become integral to the technology itself. Like many Chinese computer users, I use Hanyu Pinyin to type Chinese every day, not because of speed or optimal compatibility between software and hardware design, but due to a language instinct technologically cultivated into an embodied “no-brainer” of human-computer interaction. Human adaptation makes the quest for innovation so captivating, and needs to be a part of the history of technology. I enjoyed reading this book and am never tired of the intricacies of technical inventions grounded in their historical and social background, but as a historian, I would like to see all the details present a more coherent story, which this book falls short of.