When Computers Were Human

by David Alan Grier

  To process the Greenwich backlog, George Airy created a computing group at the observatory that resembled the staff at Stratford’s almanac office. Airy’s computers were mere boys, some as young as fifteen, who arrived at eight in the morning. They worked at tall desks in the original observing room of the building. This room, called the Octagon Room, more closely resembled an eighteenth-century ballroom than the dome of a modern observatory. Edmund Halley had gazed through the room’s tall, slender windows to map the night sky. Those same windows admitted the sunlight to illumine the desks and papers of the computing boys. At noon, the boys took a break for supper, and they resumed their places at one. In winter months, they would bring candles to compensate for the failing light as they computed through the afternoon. Only when the clock had passed through a complete cycle and reached the hour of eight in the evening would they be allowed to go home.19

  8. Original Greenwich Observatory with the Octagon Room

  The boys were not the hairdressers of de Prony’s Bureau de Cadastre, as they came to the observatory with abilities that could be compared to those of the almanac computers of Nevil Maskelyne. They generally possessed the basic skills of mathematics, including “Arithmetic, the use of Logarithms, and Elementary Algebra.”20 Some had been educated at the Greenwich Hospital School, a school that trained boys to be seamen in the navy. Many had learned their computing skill from fathers or uncles. Among Airy’s early computers were two sons of an almanac computer. The boys had learned enough mathematics to expect a career in ocean trade or as civil servants or possibly as scholars at Cambridge. They became computers only because their father had died and left the family with no fortune or income.21

  The elder of the two boys, Edwin Dunkin (1821–1898), was seventeen when he began computing. He reported that he learned the observatory’s computing procedures from a book of printed forms that had been designed by Airy. On his first day in the Octagon Room, the chief computer placed the book in front of him and indicated the work that needed to be done. “I felt a little nervous at first,” he reported, “and a momentary fear crossed my mind that some time would be required to enable me to comprehend this intricate form, and to fill up the various spaces correctly from the Tables.” After a little instruction from the chief computer, “I began to make my first entries with a slow and tremulous hand, doubting whether what I was doing was correct or not. But after a little quiet study of the example given in the Tables, all this nervousness soon vanished.” His brother, who was two years younger, worked on simpler problems, but at the end of the day, both felt that they had mastered a small part of astronomical mathematics. “We went home tired enough to our lodgings,” Dunkin recalled, “but with light hearts and the happy thought that we had earned our first day’s stipends.”22

  Among the computers, Dunkin was one of the promising young men. His background gave him the possibility of a career beyond the computing room. Eventually, he would rise through the observatory ranks, become the superintendent of the computing office, and be admitted to membership in the Royal Astronomical Society.23 Not all of his peers would recall their time in the Octagon Room with such fondness. Some compared it to the “satanic mills” of William Blake or the soot-stained buildings of Charles Dickens, where “every day was the same as yesterday and to-morrow, and every year the counterpart of the last and the next.”24 Several observatory workers described George Airy as “despotic in the extreme” and claimed that he “was the cause of not a little serious suffering to some of his staff.”25 The charges against him were similar to those in any conflict between labor and management: the pay was too low, the hours too long, the computers could be dismissed if there was insufficient work, and they were always terminated when they reached twenty-three years of age.26 Most critics identify the twelve-hour working day as the greatest source of complaint against Airy, though they acknowledge that this schedule was due, in part, to the computers themselves. Airy eventually concluded that the length of the shift was too long and reduced it to eight hours without reducing pay.27

  Airy’s defenders have portrayed his management of the observatory as progressive and inventive. “He was a colossal-minded man,” wrote one friend, “and his ideas seemed to be executed in granite.”28 Most modern scholars have recognized that Airy brought new efficiency and new strength to the Royal Observatory, but they have also argued that he embraced the very “contagion” that he hoped to avoid by organizing his computers as if they were a “bureau of clerks.” “Airy was impressed by the power of contemporary industrialists and engineers to transform both society and the manufacturing arts,” wrote historian Allan Chapman. Chapman went on to argue that the only thing that distinguished Airy from a contemporary industrialist “was the fact that his profit was measured in terms of public utility and scientific prestige, rather than Pound Sterling.”29 In fact, Airy and William Stratford at the Nautical Almanac Office may actually have seen a connection between prestige and “Pound Sterling.” There was no market for scientific research the way there was a market for wool cloth or tin sheeting, but as Adam Smith had argued fifty years before, there was a value that could be placed on the economic efficiency of the Royal Observatory or Nautical Almanac.30

  The economic efficiency of both the observatory and the almanac was evaluated by independent organizations. Such review organizations were one more aspect of factory operation that was adopted by scientific institutions. These review organizations played a role similar to that of the board of directors of a commercial organization by representing those with a stake in the institution. The Nautical Almanac Office had faced a thorough review in 1829–31 by a committee of the Royal Astronomical Society. Beginning in 1836, the Greenwich Observatory was evaluated annually by an independent board of visitors. This board checked everything from the quality of research to the size of the staff and the condition of the instruments.31 George Airy may have exercised “despotic” control according to some, but he did not possess the unfettered freedom that had been allotted to Edmund Halley in the seventeenth century or to Alexis Clairaut in the eighteenth. In making any decision, Airy had to consider how his actions might be judged by his board of visitors.

  As Astronomer Royal, Airy had some experience in reviewing the work of others. In 1842, the British government asked him to evaluate the Difference Engine of Charles Babbage. After inspecting the drawings of the machine and looking at the prototype parts, Airy brusquely dismissed the engine. “I can therefore state without the least hesitation,” he wrote afterwards, “that I believe the machine to be useless, and that the sooner it is abandoned, the better it will be for all parties.”32 It was, perhaps, an unnecessarily harsh conclusion, and it caused Airy trouble in later years. The historian Doron Swade has observed that Babbage took his revenge by portraying Airy as an “unimaginative bureaucrat—a mediocre but influential insider.”33 Yet Airy’s evaluation was more true than false. The Difference Engine could not process data and hence could not assist his computers in their efforts to reduce the Greenwich backlog. Though it might be able to prepare an ephemeris, it would be forced to do those computations in small pieces. Between each computation, the machine would have to be stopped and reset. The process of resetting the machine required substantial mathematical analyses that Babbage had not done, and hence Airy could easily believe that the Difference Engine would not be worth the effort required to operate it.

  Babbage had received a gold medal from the Royal Astronomical Society for designing the Difference Engine, but the prestige of that medal did not translate into the pounds sterling needed to complete his machine. William Stratford and George Airy labored not for gold medals but for the simple approval of their actions by their oversight committees and boards of visitors. In the course of their careers, they created the premier computing laboratories for their day. The computing staff of the British Nautical Almanac and the computers of the Royal Greenwich Observatory would provide models for computing offices for the next eighty years. Such offices
would have a central computing room, an active manager, preprinted computing forms, standard methods of calculation, and a common means of checking results. They would also have regular hours of operation, a clock on the wall beating out the hours, and an oversight board that would ensure that the work was well done.

  CHAPTER FOUR

  The American Prime Meridian

  The eyes of others have no other data for computing our orbit than our past acts, and we are loath to disappoint them. …

  Ralph Waldo Emerson, “Self-Reliance” (1841)

  THE METHODS of computation migrated to North America with the navigators who guided the ships across the North Atlantic and the surveyors who delineated the European claims upon the continent. Throughout the eighteenth and nineteenth centuries, the Greenwich Observatory and the British Nautical Almanac served as sources of computational techniques for those traveling west. The observatory had employed Charles Mason before he departed with Jeremiah Dixon to survey the border between Pennsylvania and Maryland.1 One of the almanac computers, Joshua Moore, emigrated to the United States and corresponded with President Thomas Jefferson on subjects mathematical.2 The British Nautical Almanac itself was reprinted in Salem, Massachusetts, and was freely available for purchase at the ports of Boston, Nantucket, New York, and Philadelphia.

  The models of organized computational labor emigrated more slowly to the United States than did the mathematical methods of astronomy, navigation, and surveying. Through the first decades of the nineteenth century, American science was the work of individuals rather than organizations, the effort of Benjamin Franklin or Thomas Jefferson rather than of an almanac office or the computing factory of an observatory. “Amongst few of the civilized nations of our time have the higher sciences made less progress than in the United States,”3 observed the French writer Alexis de Tocqueville (1805–1859). When de Tocqueville visited the United States in 1831, he found a country that had supported only a few large scientific projects, most notably the survey expedition to the Pacific Northwest of Meriwether Lewis and William Clark. De Tocqueville drew the general lesson that “those who cultivate the sciences amongst a democratic people are always afraid of losing their way in visionary speculation,” but in reaching that conclusion, he failed to understand the connection between the nature of American democracy and large scientific projects. Even in the early nineteenth century, the big scientific endeavors received government patronage. In England, the crown’s government had provided financing for the Nautical Almanac, the Royal Observatory, Babbage’s Difference Engine, and even the construction of the railroads. In the United States, the citizens deeply distrusted the power of governments and national institutions. Just before de Tocqueville visited the United States, the American Congress had rejected a proposal to create a national university and national observatory. The proposal had been drafted by President John Quincy Adams, who had a deep interest in science and learning. Adams was well read in the classic literature of political philosophy, but such knowledge did not make him a skilled leader. “It would have been difficult … to propose a more unpopular measure,” observed historian Hunter Dupree.4

  Before 1840, the United States government operated only one permanent scientific agency, the Coast Survey Office. As the name implied, the Coast Survey was responsible for cataloging harbors and navigational hazards of the Atlantic shore. It had been founded in 1807 but had accomplished little, as it was bedeviled by a weak leader and vacillating congressional support. Congressional interest in science changed only as the country’s population began to move from farm to workshop. In 1800, less than 5 percent of the adult population was involved in manufacture. By 1840, that fraction had risen to 25 percent. The new workers brought to power the Whig Party, a group that advocated improvements to the national infrastructure, including the construction of roads, the digging of canals, the expansion of ports, and the creation of scientific institutions. The navy was the first government office to be touched by this political shift. In 1842, the naval secretary reorganized the entire command structure of the navy and created two scientific offices. The first was a small ordnance proving ground, a testing place for cannons and mines. The second was an astronomical observatory, originally given the name of National Observatory, though the title of “National” was soon changed to “Naval.”5

  The navy’s actions were soon followed by a major reorganization of the Coast Survey Office. In 1843, Congress gave the survey office an expanded budget, a broader scope of operations, and a new, dynamic superintendent.6 The next year brought the founding of the Smithsonian Institution (1844) as an office for the “increase and diffusion of useful knowledge among men.”7 These government agencies were matched by two important private scientific institutions, the Harvard Observatory (expanded in 1843) and the American Association for the Advancement of Science (1848). The last major scientific institution of the 1840s, the American Nautical Almanac Office, was created in the last year of the decade.8

  “The peculiarity of American institutions,” wrote the historian Frederick Jackson Turner, “is the fact that they have been compelled to adapt themselves to the changes of an expanding people—to the changes involved in crossing a continent, in winning a wilderness, and in developing … out of the primitive economic and political conditions of the frontier into the complexity of city life.”9 In many ways, the founders of the early American scientific institutions were working in on intellectual frontier, borrowing and adapting ideas from their European counterparts. The Coast Survey purchased equipment from France and Germany. The American Association for the Advancement of Science took its name and purpose from a British organization whose founders included Charles Babbage. The American Nautical Almanac based its operations upon the ideas of Nevil Maskelyne. Not all such ideas were successful in their transit across the Atlantic. The U.S. Navy mistakenly constructed its new observatory on a river bluff site that resembled the placement of the Royal Observatory in England. Unlike the park at Greenwich, the navy’s perch over the Potomac River had little to recommend it. From the first nights of operation, astronomers complained that river mists fogged telescope lenses and that marshes bred swarms of infectious mosquitos. They might have borne such trials more bravely if the site had had the advantages of Greenwich, such as easy proximity to a navy yard or the view of ocean-bound ships. The only vessels that passed the American observatory were canal barges bound for the Ohio Valley, boats that did not need an almanac in order to find their way.

  Though all of the American scientific institutions of the 1840s were touched by the “primitive economic and political conditions” of the North American continent, they were more profoundly shaped by competition with European institutions. This is especially true for the largest computing organization of the age, the American Nautical Almanac. The almanac was generally viewed as a practical means for improving the navigational skills of the navy and the mercantile fleet. “But for the Nautical Almanac of England or some other nation,” claimed one supporter, “our absent ships could not find their way home nor those in our ports lift their anchors and grope to sea with any certainty of finding their way back again.”10 Yet many individuals saw the almanac as a way of demonstrating America’s intellectual accomplishments. It “would be a work worthy of the nation,” wrote one scientist, “and might engage our ablest astronomers and computers.”11 Such thoughts were echoed by the political leaders. An almanac “was important to the character of the country,” argued one member of Congress, “important to the national pride, national honor and independence.”12

  The almanac was shaped by two individuals, Lieutenant Charles Henry Davis (1807–1877) of the U.S. Navy and Harvard College professor Benjamin Peirce (1809–1880). The two were related by marriage and lived in neighboring houses in Cambridge, Massachusetts.13 Their front doors were just a few steps from Harvard Yard; their back windows looked across the fields to the distant ships at the docks of Cambridgeport. Davis’s son recalled that children moved freely betw
een the two homes and that their families “dwelt almost as one.”14 Their friendship had begun some fifteen years before, when Davis had taken a house in Cambridge following an extended voyage. He had once been a student at Harvard College, but he had left without a degree in order to take an officer’s commission. During his travels, he had retained an interest in learning. One of his commanders described him as “intelligent in his profession, energetic in his character, and devoted to the improvement of his mind.” While on an early voyage, he studied navigation, learned “French, Spanish and a good smattering of Italian,” and read the complete works of William Shakespeare.15 In 1840, he had returned to Cambridge for a year of intensive mathematical study with Peirce. According to his son, Davis did not always “follow the transcendent flights of Peirce’s genius” but persevered in his study and ultimately acquired “a working familiarity with mathematical tools.” For his efforts, he received a Harvard degree, conveniently backdated to suggest that he had graduated with his original classmates.16

  9. Lieutenant Charles Henry Davis of the American Nautical Almanac

  By all appearances, Benjamin Peirce was an unlikely friend for Davis. Davis was a practical and disciplined officer from a privileged Boston family. No sign better captured his bearing than his prominent moustache, groomed with military precision. Peirce was one of the “bearded ancients” of Harvard College and sported an unruly head of hair. His lectures were rambling, difficult affairs that were often incomprehensible to students. The one quality that may have caught Davis’s sympathy was Peirce’s ability to maintain his intellectual bearings. In the 1840s, Cambridge was awash in the Transcendental movement, the philosophical current that saw in every aspect of nature “a symbol of some spiritual fact.”17 Peirce befriended the leader of the Transcendentalists, Ralph Waldo Emerson (1803–1882), and occasionally referred to God as the “Divine Geometer,” but he never lost sight of the theories of Newton or the mathematics of calculus.18