Sheffield's Chinese typewriter as depicted in Scientific American, 3 June 1889, 359 (From Mullaney, "The Moveable Typewriter," Technology and Culture, 53 (2012).
In 1897 American Presbyterian missionary to China, Devello Z. Sheffield, head of the Lu He Boys' School in Tungchow (Tongzhou), developed an operational Chinese typewriter. It took Sheffield more than a decade to develop the invention. Other than a functioning prototype, it is doubtful whether the machine was ever manufactured.
"In Sheffield’s machine, which he completed by 1897, we find an intensified engagement with both common usage and somatically determined organization. First, Sheffield set out toward a “careful selection of characters for general use,” attempting to determine which characters were necessary for his machine—a question that was already central to Chinese typesetting, as we have seen. To this end, he began frequenting local foundries and typesetting offices, speaking with professionals who, through their accumulated experience of cutting, forging, and using Chinese typeface, possessed an intimate, firsthand knowledge as to which characters were most and least frequently used. Moreover, it is almost certain that Sheffield was familiar with the work of William Gamble (1830–1886), an Irish-born American publisher dispatched to China in 1858 to oversee operations of the Presbyterian Mission Press in Ningbo. With the assistance of two Chinese scholars, Gamble spent four years examining 4,166 octavo pages of Chinese printing, containing roughly 1.3 million characters in all. Gamble’s research produced what must have been a thrilling conclusion for Sinology: of the more than 40,000 characters found in the Kangxi Dictionary, only 5,150 of them were needed to produce the texts he had examined. He further expanded on his discovery by developing a novel type case that, just like Sheffield’s machine two decades later, would enable the typesetter to reach all of these high-frequency characters from a static position. Based on this research, Sheffield estimated that “the working vocabulary of Chinese scholars is easily within the limit of six thousand, and this list can be reduced to four thousand, with but rare occasion to strike outside of the list to give expression to thought.” Sheffield’s novel inscription technology differed from that of the printing press, opening up alternative avenues of exploration. He decided to take the concept of common usage one step further than his counterparts in the printing office could by not only partitioning the language into the categories of common and less common characters, but jettisoning the second category from his machine altogether. Sheffield settled on a set of 4,662 characters, excluding tens of thousands of others and affecting a sharp break in the way that infrequent characters were handled within printing" (Thomas S. Mullaney, "The Moveable Typewriter: How Chinese Typists Developed Predictive Text during the Height of Maoism," Technology and Culture, 53 (2012) 777-814; quotation from pp. 787-788).
Sheffield published an unillustrated account of his invention as "The Chinese Type-Writer, its Practicability and Value," Actes Onzieme Congrès International des Orientalists, Paris, 1897, Première section: Langues et archeologie des pays ariens (Paris: Imprimerie Nationale, 1899).
