In describing the process of creating Book from the Sky, one as compelling as the finished work, Xu Bing helps us gain a clearer sense of what these dimensions entail. “My requirement of these characters,” he begins by explaining, “was that they resemble Chinese characters to the greatest extent possible, while still not being Chinese.” To achieve this objective, Xu set out on a painstaking analysis of real Chinese characters in order to extract those qualities he would imbue in his fake font. To begin, the particular number of fake characters he chose to produce—four thousand, as compared to one hundred or perhaps one million—came not at random but because it mimicked the statistical realities of character frequency and common usage within actual Chinese. “I decided to make just over four thousand fake characters,” Xu explained, “because average reading material is also made up of around four thousand different characters.” “When you’ve learned four thousand or more characters, that is to say, you can read and thus you’re an intellectual.” Structurally, Xu explained, the creation of his fake characters “required me to observe the internal structural principles of characters.” To that end, Xu carefully examined the Kangxi Dictionary to determine the average stroke count of real Chinese characters, as well as their distribution across a curve from lower- to higher-stroke-count characters—all of which would then inform his production process. Stylistically and aesthetically, meanwhile, Xu Bing did not develop a “fake” font for Book from the Sky, but instead patterned his character forms after one of the most conventional Chinese fonts of all: the Song-style character or songti, widely used in printed Chinese matter well into the present day. “As for the font,” he explained, “I thought to use Song-style characters. Song-style is also called ‘court-style.’ Often being used in important documents and in serious matters, Song-style is the font with the least individual flavor to it—the most standard font.”22 Xu Bing even pushed his exercise in verisimilitude into the realm of taxonomy. He created his own system for organizing his movable type blocks, so that he could retrieve them during the printing process—just like a real compositor.
When viewed through the yin-yi-xing framework, then, Book from the Sky might strike us as an exercise in rupture and discontinuity. When viewed within the technolinguistic realms of taxonomy, instrumentality, statistics, and materiality, however, we see that Book from the Sky was just the opposite: an exercise in continuity, or more accurately, an exploration of just how far one could push technolinguistic continuity and at the same time produce a Chinese script that violated everything that a Chinese script was supposed to be within the confines of the classic yin-yi-xing triad.
Rather than pursuing the history of Chinese script through the conventional fiction of the yin-yi-xing triad, a fiction in which script “comes loose from the metal” (to draw again from Carter), our examination will dwell primarily in a technolinguistic realm that makes the sound-meaning-shape triad possible.23 We will climb into the manholes, crawlspaces, and airshafts of Chinese, exploring all of the complex and fascinating meaninglessness that makes meaning tick.
In concrete terms, what happens when we shift our attention to the technolinguistic? What happens when, following Carter’s observation, we do not let the face come loose from the metal? First, we find ourselves less prepared. Traditionally, scholars of China are exceedingly well-trained at excavating meaning from the sound-meaning-shape triad, and attuned to detecting shifts therein. Indeed, as soon as the words “Chinese language reform” are broached, the thoughts of the historian turn almost instinctively to familiar subject matter: the deluge of newly coined Chinese words derived from other languages; the vernacularization movement of the 1910s and its call for greater congruence between written and spoken Chinese; the widespread development of specialized and professionalized Chinese-language discourses in fields as diverse as paleontology, aesthetics, law, constitutional reform, ethnology, feminism, and fascism; and efforts to forge a “national language” (guoyu) out of the welter of mutually unintelligible dialects. Other common invocations include calls for the Romanization of Chinese and the simplification of Chinese characters for the purposes of mass literacy.
If scholars of China are conditioned to think critically about changes such as these, transformations in the technolinguistic realm find us far less prepared. Changes in the organization of a Chinese phone book, the application of Western-style punctuation marks to Chinese texts, the reorientation of Chinese texts from vertical to horizontal alignment, the employment of numerical coding schemes to transmit Chinese characters over filaments of wire, the statistical analysis of character frequency in order to refine the parameters by which computers retrieve Chinese characters from memory—in short, the humble yet immense information infrastructure that enables the Chinese script to “work”—all seem to constitute “nondestructive” edits within the history of the language. They are changes, to be sure, but not ones that transform Chinese characters in any essential sense, whether in terms of structural makeup, phonetic value, or semantic meaning. What does it matter that a Chinese text is adjusted to read left-to-right where once it was read top-to-bottom? What does it matter that Western-style punctuation marks have been added, or an index, or page numbers, or a bar code? What does it matter that a once paperbound Chinese text is now found in proprietary digital formats like PDF? As long as the structures of the characters, their phonemic values, and their meanings remain the same, has not Chinese stayed the same—and with it the still unbroken, 5000-year hitting streak of the world’s “oldest continuous civilization”?24
The answer is no. The technolinguistic domain does not stand apart from the better known yin-yi-xing triad; in fact, the historical transformations that take place within it—and especially those that throw it into crisis—are arguably more critical than those of sound, meaning, and shape. Consider, for example, what happens when we take a technolinguistic view of three of the subjects that China historians treat as synonymous with the subject of “Chinese language reform”: the simplification of Chinese characters, the vernacularization of Chinese, and the pursuit of mass literacy. When viewed through cognitivist and sociocultural approaches, these three initiatives constitute the very core of language reform, pursuits around which the very question of language crisis seems to pivot. What happens, however, when we are faced with historical actors who were no less passionate about language reform, but whose goals included the creation of a Chinese telegraph code, a Chinese typewriter, Chinese braille, Chinese stenography, Chinese word processing, Chinese optical character recognition, Chinese computing, Chinese dot matrix printing, and more? For these reformers, many of the conventional topics of language reform actually made the problem of Chinese linguistic modernity even harder to solve—or at best had no impact on their pursuits one way or another.
Simplified characters are a case in point. While these were no doubt pertinent to questions of literacy and language pedagogy, a character reduced from sixteen strokes to a mere five—as with “dragon” (long), in its transformation from the graph 龍 to 龙—was no easier to transmit across telegraphic wires, set in movable type, or type on a Chinese typewriter than its “traditional” counterpart. “Simplification” here simplified nothing.
Vernacular Chinese, or baihua, actually made things palpably worse. Vernacular Chinese texts are invariably lengthier than equivalent messages in literary or “Classical” Chinese, and so this movement in the early twentieth century quite literally multiplied the challenges of transmission, inscription, and retrieval. To send a message in vernacular Chinese meant having to send longer messages, which in turn exacerbated the fundamental question even further: namely, of how to transmit, type, save, or retrieve a Chinese character—any Chinese character—in the first place.
Most counterintuitively, mass literacy—that preeminent concern of language reform—exacerbated the problem of Chinese information technology most of all. No longer able to rely upon certain kinds of assumed literate subjects—the literati and Civil Service examin
ees of old, for example—developers of the modern Chinese information infrastructure not only had to build all of these new and challenging technolinguistic systems, but they had to do so with only vague, imperfect, and constantly shifting understandings of what the millions of new Chinese users of their systems would be like. How literate would these users be, and in what ways? What dialects would they speak? What would their professions and educational backgrounds be? Would they be men or women, girls or boys? In what kinds of technical environments would they be operating? These questions would impinge upon the systems being built, and yet at no point in this history would anyone have a stable answer.
And who precisely had the authority to make such decisions? As party to this vexed transition from Qing imperial subject to informed (and informational) citizen of the new Republic, Chinese intellectual and political elites were undergoing a tortuous transition of their own. Over the course of the late nineteenth century, and most notably in 1905 with the abolition of the Civil Service Examination system, controls that the state and establishment intellectuals enjoyed over the Chinese language steadily crumbled, eliciting acute anxieties over when and how a new regime of language would take shape, what it would look like when it did, and who would be positioned at the apex of this new hierarchy. The uncertainty of this period was further compounded by the rise of an entrepreneurial cultural class, one that rushed into the vacuum left by disintegrating state power in hopes of establishing private and profitable cultural enterprises. Just as concerns over modern information management were reaching fever pitch, then, the combined absence of state control and the rise of the “business of culture” made this an ever giddier and more unsettled time.25
Above all, by turning our attention to the technolinguistic we begin to appreciate just how strange continuity truly is—a point that returns us to Xu Bing and his Book from the Sky. Continuity is strange because, despite commonsense understandings, it is in no way synonymous with conservatism. To continue something—in this case, to continue character-based Chinese script—can be avant-garde, iconoclastic, radical, and even destructive. Phrased differently, while it is virtually a cliché to speak of the “destruction” often entailed in acts of creativity, rarely do we pause to reflect upon the destruction central to acts of continuation. What is more, continuity and discontinuity are not antithetical concepts, as we see with Xu Bing and his fake characters. The question is not to continue or not to continue. The question is what to continue, and which kinds of discontinuity will be essential for achieving that goal. If the diverse actors in our history could be said to share one worldview in common, it could be summarized through recourse to a famous passage from the twentieth-century Italian novel Il Gattopardo by Giuseppe Tomasi di Lampedusa. Uttered by Tancredi Falconeri, a young prince in a wealthy aristocratic family in Sicily, and nephew to the book’s protagonist, the brief passage contemplates how his family’s position within society might ever hope to survive the tumultuous and unificationist impulses of the Risorgimento as it swept through their Sicilian homeland on its way northward:
Se vogliamo che tutto rimanga com’è, bisogna che tutto cambi.
In order for everything to stay the same, everything must change.
Giuseppe Tomasi di Lampedusa was no turn-of-the-century Chinese language reformer, and yet this passage elegantly captures a belief and impulse that propelled many of the protagonists we will meet in our history. Like Tancredi, they too believed that in order for everything to stay the same, everything must change. “Everything”—or tutto—is the operative term here, of course, which in its repetition corresponds to two separate referents. The first tutto—the one we want to stay the same—is the very yin-yi-xing/sound-meaning-shape triad discussed above: that part of script that lies above the surface, and through which the immense Chinese-language corpus is inscribed, read, appreciated, and more. This is the part of the script that naive iconoclasts would have done away with, whose utterly impractical schemes would have seen it replaced with Esperanto, or French, or one or another alphabetic scheme. By contrast, the second tutto corresponds to something else entirely, something which through its wholesale transformation might enable the salvation of the yin-yi-xing triad. This tutto corresponds to our technolinguistic: the infrastructure of language that, in its modest immensity, enables the language to work in the first place. If this tutto could be ripped apart, broken down, and reconstituted—the way that Chinese characters were categorized, retrieved, transmitted, materialized, ontologized, and indeed conceptualized—then perhaps the Chinese script could survive and even thrive despite our age of alphabetic hegemony.
Field Notes from the Abyss
In this book, we will focus on one of the most important and illustrative domains of Chinese technolinguistic innovation in the nineteenth and twentieth centuries: the Chinese typewriter. In addition to being one of the most significant and misunderstood inventions in the history of modern information technology, so too is this machine—both as object and as metaphor—a historical lens of remarkable clarity through which to examine the social construction of technology, the technological construction of the social, and the fraught relationship between Chinese writing and global modernity.
If only it were peppered generously with moments of success and triumph, the story of the Chinese typewriter, and modern Chinese information technology more broadly, would be simpler and more joyous for the telling. If the Western typewriter marked a “revolution in production,” as one historian phrased it, and one that “greatly increased the speed and reduced the cost with which a written document could be produced,”26 our hope would be to make similar claims for the Chinese typewriter—to establish it as the unrecognized equal of its better-known Western counterpart. Another strategy might be to follow the path carved out by wildly popular “object histories,” a veritable cottage industry in which “authors attribute to their chosen commodities an exaggerated, mysterious, almost godlike power,” as Bruce Robbins has phrased it.27 If tulips, codfish, sugar, and coffee have all changed the world, it stands to reason that perhaps the Chinese typewriter did, too.
No such triumphal story awaits the reader. While it might be tempting to identify one or another of our historical figures as the “Chinese Charles Babbage,” the “Chinese Grace Hopper,” or perhaps the “Chinese Steve Jobs,” this would amount to little more than a distracting parlor trick. While the Chinese typewriter did find its way into major Chinese corporations, as we will see, as well as metropolitan and provincial governments across the country, the Chinese typewriter did not transform the modern Chinese corporation or the functioning of Chinese government. For better or worse, no history of the Chinese typewriter can stake its claim upon the idea of “impact.”
With all of this in mind, then, a fair criticism at this point might be: Did China need the typewriter at all, and if not, why do we need a history of it? Would it not be more accurate to say that China “skipped” the typewriter, moving directly to the computing age in the same way that certain parts of the world “leapfrogged” landline telecommunication, and entered directly into the world of cellular phones?28
In one respect, the answer is yes. To the extent that we demand certain things of a technology prior to admitting it into the category of “typewriter”—that it will have transformed the history of modern business communications and record-keeping, that it will have become a fundamental part of the feminization of the clerical workforce, that it will have become a cultural icon whose impact in popular culture extended far beyond its role as a business appliance—then the Chinese typewriter we are about to examine will not seem like much of a “typewriter” at all. Would it not be best simply to come clean, then, and admit that Chinese script is unsuited to the technology of typewriting, and that between the alphabetic and the nonalphabetic lies what has been called a “technological abyss”?29
In another more important way, the answer is resoundingly no. Chinese typewriting may not have approached the scale or centrality
of its counterparts elsewhere in the world, and yet in many ways China experienced and engaged with the age of typewriting—and indeed telegraphy and computing as well—much more intensively than did the alphabetic world. As early as the 1870s, the novel inscription technology was known in China, and looked upon with admiration. In his account of the 1876 Philadelphia Centennial Exposition, Chinese Customs official Li Gui wrote of the “ingenious” device:
It was set up on a small square table and was only about a [Chinese] foot high and eight inches wide and made of iron. In the middle of it was a clever device embedded with ink, and an iron plate was set up under which were arranged all the letters of the foreign alphabet, twenty-six, like chess pieces, operated by a woman on the staff [i.e., a type-writer]. Paper is placed on the iron plate then, using a technique similar to that of playing the piano in foreign countries, it prints certain letters by means of her hands pressing certain alphabetical keys, while inside the machine an impression of each letter is struck. These are connected together to form words very nimbly and quickly. Offices all buy one, since its uses are many and its cost is only in the range of a little over one hundred dollars. Unfortunately, however, it does not print Chinese characters.30
To build a Chinese typewriter—a sentiment and desire just beneath the surface of Li Gui’s closing remark—would be no small feat. To bring a nonalphabetic script into a technological domain that was built with alphabets in mind, engineers, linguists, entrepreneurs, and everyday users had no choice but to bring both script and technology into a shared critical space, posing questions that today might sound like irresolvable Zen kōans, but which in their original contexts were deeply practical ones: What is Morse code without letters? What is a typewriter without keys? What is a computer where what you type is not what you get? The Chinese typewriter was not a new type of mining drill, nor a new type of artillery, nor anything like most of the technologies being imported from abroad during the modern period—technologies that, while they undoubtedly demanded their own forms of intangible cultural, political, and economic practices and worldviews, could at the very least be “switched on” the moment they reached Chinese soil. As linguistically embedded and mediated technologies, Chinese telegraphy, typewriting, and computing explode conventional narratives of “technology transfer” and “diffusion” that have long guided our understanding of how industrial, military, and other apparatuses and practices circulated from Western loci of invention to non-Western loci of adoption.31 Technolinguistic systems such as typewriting, telegraphy, stenography, and computing came with much steeper requirements. Because these systems were conceptualized and invented in direct association with alphabetic script, even the most basic functionality of a Chinese typewriter or a Chinese telegraph code required inventors, manufacturers, and operators to subject both the Chinese script and the technologies themselves to unprecedented forms of analysis and reconceptualization—to scrutinize both Chinese and typewriting, telegraphy, computing, and more. In order for everything about Chinese characters to stay the same, that is, everything about Chinese characters and modern information technology would have to change.
The Chinese Typewriter Page 4