After World War II, the Hammond principle became part of a versatile and successful composition system called the Varitype that remained the most economical way to set mathematical copy before the computer. Beginning in the 1960s, IBM introduced the Selectric typewriter with its tilting and rotating golf-ball style element; a number of manufacturers still make word processors with rotating type wheels of the kind Hammond pioneered. Why, then, did the Sholes typebar triumph? What it lacked in design elegance it more than made up for with one of the best systems for product improvement that the industrial world had yet seen, and the credit goes to Sholes’s associate James Densmore, one of America’s most underrated entrepreneurs. Densmore not only risked all of his $600 in savings in a quarter-share of the invention, he gradually bought out the other partners. Equally important, he realized the power of steady incremental improvements. Densmore relentlessly pressured Sholes to produce one new model after another. Believing that court reporters would be an important early market, he sent models to James O. Clephane, one of Washington’s leading practitioners, for rigorous testing that led to new rounds of changes—as many as thirty models over four years. Densmore also found an ideal manufacturing partner in the Remington Arms Company of Ilion, New York, where expert mechanics had benefited from the advanced machine-shop practice developed at America’s national armories and then introduced into private manufacturing.17
The Remington Model 1 typewriter, introduced in 1874, employed most of the principles of a twentieth-century mechanical typewriter, from the spring-driven carriage with its rubber platen to the keyboard arrangement and basket of typebars. But although it had been through years of relentless development, the Model 1 was limited even by the day’s standards. It was slow and expensive, at $125 about the real price of today’s high-end personal computers. Fewer than a thousand a year were sold during the 1870s, even after Remington offered shifting for upper and lower case. The machines were far more popular with court reporters than with businessmen, whose customers often suspected typewritten letters of being printed handbills intended for the semiliterate. The technique of penmanship was still esteemed, and the legal validity of typewritten documents and signatures was debated. Sears, Roebuck & Company sent handwritten correspondence to its rural customers, and the U.S. government did not start to authorize typewriter use until the end of the century Handwriting was laborious, and many people yearned to do without it, but when the opportunity appeared they hesitated. Palmer’s penmanship schools at first spread faster than the machine that was supposed to put them out of business.18
What finally overcame obstacles to the typewriter was not any price reduction or further technological breakthrough. It was the changing face of organizations and society. Companies and government bureaus were growing larger, and their internal as well as external communication needs grew even more rapidly. Fine penmanship might still be a welcome courtesy to an individual customer, yet it could not address this volume of business. As the number of public and private clerical employees rose, all the schools’ penmanship exercises could not assure the uniformity that new bureaucratic life demanded. Eliminating confusing individual variations in writing, the technology of the typewriter permitted standardized fonts that were as closely tied to the emerging industrial and commercial society as the late medieval textus quadratus, Gutenberg’s model, had been to the scriptures and devotional books of its own day. Black letter had achieved shortcuts in writing and reading with dozens of ligatures and abbreviations; typing achieved the same through speedier use of a more limited character set. The typewriter fonts familiar well into the twentieth century, pica and elite, were as anonymous in design and as ubiquitous as black letter had been. Although a variety of typefaces was available from the earliest days of typing, there seemed to be an overpowering unconscious will to suppress variation in the interest of interchangeability It did not matter that each manufacturer used slightly different matrices or that each typewriter produced unique details of letterforms and alignment. Often ornate printers’ fonts, especially for advertising, were proliferating in the nineteenth and early twentieth centuries, but the typewriter retained its chaste uniformity.19
By 1886, Remington and other makers were selling fifty thousand typewriters a year, and organizations were embracing textual uniformity The Writer, a leading journal for authors, repeatedly reminded subscribers of magazine publishers’ preference for typewritten copy The advantages of standardized text for compositors working with hundreds of manuscripts each year were obvious; book typesetters could afford to be more indulgent, and only after World War II did nearly all of them require typewritten copy Even so, as early as 1889 the city editor of the Boston Globe could declare that a typewritten manuscript improved chances of acceptance by 10 percent.20
The uniformity of type had a social advantage, too, at least for employers. Carbon paper banished cumbersome methods of duplicating handwritten documents and made it possible to generate a copy of every outgoing letter automatically. Even more important, the keyboard responded to an emerging management trend: turning the previously flexible definition of clerical work into a hierarchy with more limited opportunities for advancement. As part of the trend, stenography and typing were feminized once the New York Young Women’s Christian Association began to offer courses. Women, who had become so fluent at the piano, were now thought to have a special affinity for keyboards. Best of all for the initially skeptical employers, women could be hired for wages 25 percent lower than those men earned for comparable work. In Frank Lloyd Wright’s state-of-the-art Larkin Building (1904) in Buffalo, where we encountered the “suicide chair” in Chapter Five, messengers took (male) correspondents’ “graphophone” cylinder customer-service responses to the (female) Typewriter Operators’ Department, a novelty with a balcony for visitors. Economic rationalization and gender stereotyping went together.21
The keyboard turned out to be not only a tool but a management weapon. After an operators’ strike in 1907, the Western Union Telegraph Company, which had spurned earlier automatic transmission systems, began to replace its unionized, mostly male Morse operators with women using automatic transmission equipment with typewriter-style keyboards. After sixty years of prototypes and experiments, anonymous keyboarding replaced the distinctive techniques and knowledge of a formerly proud craft. The new generation of female telegraphers would never have the chance to produce its own Edison. Bookkeeping, too, was shifted to machines that combined alphanumeric keyboards with mechanical calculators. The keyboard was dividing the clerical workforce into semiskilled and professional castes. Only the mechanical compositors were able to continue corporate traditions with the new technology. Ottmar Mergenthaler’s Linotype, which prevailed in newspaper work, had a distinctive keyboard pattern needed to accommodate stacks of matrices. The keys were arranged by declining frequency of use in English. Because no typewriter used this arrangement, it took decades before typists could be enlisted in periodical typesetting. Male hand compositors, unlike their telegrapher brethren, welcomed automatic technology. Women had been gaining ground as manual typesetters. While the Linotype Company insisted they could also be excellent operators of its machines, the men were able to exaggerate the strenuousness and danger of working with heavy equipment and reservoirs of molten lead. Even after punch tape operation removed this pretext, hot metal composition remained masculine keyboarding.22
A TECHNOLOGY IN SEARCH OF A TECHNIQUE
What did the men and women of the 1880s and 1890s actually do with their expensive new machines? Ottmar Mergenthaler’s company trained Linotype operators, and Thomas Edison insisted on sending a representative to demonstrate his new phonograph, but typists and their employers were on their own. The keyboard came into the world with no recommended technique. Many typewriter inventors like Christopher Sholes and enthusiasts like Mark Twain were former newspapermen who thought of setting text as the rapid motion of thumb, forefinger, and middle finger. At first neither they nor their customers c
onsidered the fourth and fifth fingers strong enough to be used regularly—even though pianists had been applying them to fortissimo passages for years. Early typing manuals were cursory, and many were based on the alternative keyboards of other manufacturers. The Caligraph and other machines had dedicated keys for capitals, like the Linotype, above or to the side of the lowercase keyboard. Operators worked out their own systems, nearly all searching for each letter. Typists working from stenographic or plaintext copy would look at a sentence or phrase, then tap it out. Four-finger typing could be fast enough.
Not inventors but users were the first to see that typing was a new skill that, like musical keyboard fingering before it, could benefit from systematic analysis and practice. The pioneers formed a competitive and cooperative community of skills, like the late-nineteenth-century bicyclists and swimmers we have discussed, or the surfers and skateboarders of the twentieth century. Many early typists probably memorized the keyboard and used all ten fingers and thumbs, but one stood out. Frank R. McGurrin, a young law clerk in Grand Rapids, Michigan, in 1878 competed with his employer on the office’s Model 1 Remington typewriter. The idea of touch typing was originally a joke by his employer, who claimed that a local court stenographer’s typist was able to work rapidly from dictation while looking out the window. As McGurrin later recalled, “I made up my mind that whatever a girl could do I could do,” and he started to use all his fingers instead of two or three on each hand. By the end of the year he was able to type ninety words per minute—still considered an excellent speed now—from new copy without looking at the keyboard. Only two years later did he meet the stenographer’s typist and learn that she had never even tried the new method. His employer’s teasing provocation had spurred him to discover the obvious: a musical keyboard style could be adapted for text.23
By the 1880s, typewriter manufacturers were sponsoring speed contests, and newer machines needed less force, making higher speeds possible. McGurrin became known as the fastest Remington operator. In an 1888 contest in Cincinnati, he confronted Louis Traub, an agent for the Caligraph, possibly the best selling of the two-keyboard machines. McGurrin won decisively, achieving ninety-five words per minute from dictation and ninety-eight from copying against Traub’s eighty-three and seventy-one. Traub’s lower score on copy helped doom the Caligraph. Because its keys were too widely dispersed to permit the touch method, he had to take his eye repeatedly from the keyboard. Traub soon replaced all the Caligraph machines in his typing school with Remingtons and began instruction in touch typing, signaling the end of the double keyboard. McGurrin continued on the exhibition circuit, promoting Remington as he achieved up to 125 words per minute.24
For all his prowess, McGurrin had no students and published no books. Mrs. M. V Longley of Cincinnati, whose husband operated a “Shorthand and Typewriter Institute,” developed her own system of touch typing in the early 1880s after seeing a Remington and reading in its brief instructions the manufacturer’s advice to use only one finger of each hand, at most two fingers of the right hand. She taught classes with the all-finger method she developed, and published the first textbook for its use in 1882. Citing piano and organ fingering, she observed: “If the hands are held over the keyboard the fingers will reach the extreme right and left, and each be in a position to do duty and the thumb will be in readiness to strike the space bar.” This method, now self-evident and ubiquitous, was a bold departure. The goal was not so much raw speed as the ability to transcribe accurately while looking at the copy rather than the keys.25
Mrs. Longley had retired by the mid-1880s, but a Maine shorthand and typing instructor named Bates Torrey published an even more influential manual, Practical Typewriting, in 1889. Citing McGurrin, and also blind typists who had mastered the all-finger method, Torrey presented a powerful case for keeping the hands in fixed positions and typing a given word consistently with the same finger motions. It was also Torrey who recognized the importance of touch, and who first applied the word “touch” to all-finger typing. The success of his book, designed for self-instruction and correspondence as well as school use, inspired a wave of other texts, and the touch method spread across the Western states. One author, A. C. Van Sant of Omaha, developed the system of fingering that has been taught ever since and is still the basis of computer typing CD-ROMs. Within about twenty-five years of the modern typewriter’s introduction, the community of instructors and typists had stabilized a technique.26
A TECHNOLOGICAL FOSSIL AND ITS CRITICS
The Remington typewriters used by McGurrin and the early textbook writers arranged their letters in a pattern that by the turn of the century was called the Universal Keyboard. Christopher Sholes had begun with a purely alphabetic arrangement but had revised it because the hammers, which were arranged in a circle in the earliest Remington, tended to jam. Sholes and Densmore worked out a new arrangement of keys for their No. 2 Remington machine, patented in 1878. Just as the musical keyboard was stabilized before 1500, its textual counterpart remains almost identical to the diagram on the patent.
Nobody has been able to reconstruct Sholes’s and Densmore’s reasoning completely. It would probably be necessary to find an operating Model 1 or 2 typewriter and experiment with combinations of letters. The QWERTY keyboard, as it came to be known, was clearly a compromise. On the middle row of text there was a nearly alphabetical sequence: DFGHJKLM. The last letter was later moved to the bottom row, where the original C and X were also later reversed. On the top row was a vowel cluster (UIO) out of alphabetical order. Sholes and Densmore were both familiar with newspaper type cases, arranged not in alphabetical order but roughly according to letter frequency. The QWERTY keyboard did not follow these patterns but was conceived in a similar spirit.27
Sholes and Densmore made a fateful assumption about the operator’s technique. Compositors used thumb and forefinger and looked at the type case as they worked, and it seemed reasonable to think that typewriter operators would do the same—as, indeed, all but a few initially did. For this style of work the QWERTY keyboard was relatively efficient. Its leading twentieth-century critic, August Dvorak, found that the most frequent letters were typed with the first two fingers of the left hand and the index finger of the right. There seems to be a balance between putting all the most frequent characters near the center of the keyboard and maintaining an order that will make it easier to find keys visually, like keeping O and P as well as the middle-row sequence together.
As early as 1875, proposals circulated for more efficient keyboards. Once touch typing prevailed, it would have been logical to look for even greater speed and comfort by devising a new arrangement. Yet neither McGurrin, nor Mrs. Longley nor Torrey is known to have proposed any modification. It is unlikely that they feared devaluation of their skill by keyboard reform. McGurrin became a well-to-do banker in the West, Mrs. Longley left teaching, and Torrey and others could have sold new editions of their textbooks. The QWERTY keyboard was not the best imaginable, even in the late nineteenth century, but it was good enough that expert typists did not feel frustrated by it. Remington wisely did not try to monopolize its keyboard arrangement under patent law. And instead of developing their own proprietary systems, Remington’s major emerging competitors, Underwood and Royal, kept the Universal Keyboard. The innovators were producers of interchangeable type element machines, notably George C. Blickensderfer. The Blick’s home row, DHIATENSOR, was claimed capable of spelling 70 percent of English-language words. But even Blickensderfer offered the Universal Keyboard as an alternative option.28
The Universal Keyboard was spreading just as the Jankó piano keyboard was failing in the musical marketplace. Most typists were no longer independent writers, attorneys, or court reporters but part of a large labor pool operating machines they did not own. Just as concert halls needed pianos with standardized keyboards, businesses wanted to be able to engage typists without having either to replace or refit the machines, or retrain their recruits. Typists and typing students l
ikewise wanted to learn the most widely accepted arrangement. Later the economist Paul David called such pressures for standardization “network externalities” and cited the QWERTY keyboard in an influential paper arguing that historical contingency can, paradoxically lock inferior technology into place.29
During the early twentieth century those who challenged the QWERTY design came from outside the office machine business. Apostles of industrial rationality they would have found it inconceivable that the dead hand of the past could constrain the invisible hand of the marketplace. Frank and Lillian Gilbreth, who developed new methods for analyzing human skills, took an interest in the typewriter, producing equipment and techniques for people with disabilities. They also took notes on the sequence of motions in typing and made slow-motion study films of championship typists at work. One of Gilbreth’s assistants in his typewriter work was William Dealey who later became a professor of education in Texas. Dealey’s brother-in-law, August Dvorak, taught educational psychology at the University of Washington.30
Our Own Devices: How Technology Remakes Humanity Page 27