When Computers Were Human

by David Alan Grier

  In the two decades that followed the end of the Civil War, women began to find a place in the computing rooms. Just as their male counterparts were no longer educated gentlemen, these women were not Maria Mitchells or Nicole-Reine Lepautes, the talented daughters of loving fathers or the intelligent friends of sympathetic men. They were workers, desk laborers, who were earning their way in this world with their skill at numbers. In many ways, they were similar to the female office workers, who were increasingly common in the nation’s cities. Before the war, offices were closed to women, except for those well-nurtured daughters, loyal helpmates, or resilient widows. The war had opened government offices to women, as the federal agencies needed more clerical workers than they could draw from the dwindling pool of male labor. Many of these first female clerks were, in fact, war widows, women who had married in the first exciting days of the conflict and lost their husbands on some hallowed battlefield in Tennessee or Virginia. These women earned a meager salary from the government that often had to support children and mothers in addition to the worker herself.39

  Women began finding employment with private companies after the war ended in 1865. The nature of business had expanded with the war, had become more complex, and now required central office staffs to coordinate production. Business skills were taught at high schools, a new innovation in public education. These skills were similar to those needed by human computers, though clerks did not need to understand the calculus of Newton. Both clerks and computers were subordinate to a professional staff, both were paid a small wage, and both handled routine work. By 1875, one out of six hundred office workers was female, and within a decade, women would fill one out of fifty office jobs.40

  In 1875, Anna Winlock (1857–1904) approached the Harvard Observatory and asked if she might be employed in calculation. Winlock was the eldest daughter of Joseph Winlock, the director of the Harvard Observatory and the former superintendent of the Nautical Almanac. Joseph Winlock had unexpectedly succumbed to a brief illness, leaving behind a widow and five children with no obvious means of support. The officers of Harvard had been kind and gracious. They had given Mrs. Winlock a decent interval to vacate the house on the observatory grounds and had helped her to find a new home in Cambridge.41 Once they felt that the new widow was settled, they ceased all financial support of the family.42

  It fell to Anna Winlock, the eldest child at eighteen years of age, to sustain the family. Winlock had been close to her father in much the same way that Maria Mitchell had been close to William Mitchell. She had watched him work at the almanac office and learned from his example the rudiments of mathematical astronomy. When Anna Winlock was twelve years old, she had been her father’s companion on an expedition to view an eclipse of the sun. The party left Cambridge and traveled southwest to Kentucky in order to make its camp. The young girl was probably included because the trip passed through the country of her father’s birth. Along the way, the senior Winlock introduced his daughter to his various cousins and sisters and aunts. Nonetheless, the expedition was a time with the astronomers, an opportunity to prepare instruments and observe an eclipse.43

  Joseph Winlock had left the Harvard Observatory volumes of unreduced observations, a decade of numbers in a useless state. The interim director complained that he could not process the data, as “the condition of the funds is an objection to hiring anyone.”44 At this point, Anna Winlock presented herself to the observatory and offered to reduce the observations. Harvard was able to offer her twenty-five cents an hour to do the computations. Winlock found the conditions acceptable and took the position.45 In less than a year, she was joined at the observatory by three other women. The first was Selina Bond, the daughter of her father’s predecessor. Like Winlock, Bond stood in need of a steady income, as her father’s fortune had been squandered through the actions of a “rascally trustee.” The second, Rhoda Saunders, was the graduate of a local high school who had been recommended to the observatory by the Harvard president. The last was probably the relative of an assistant astronomer.46

  By 1880, the Harvard Observatory employed a complete staff of female computers. The director who hired this staff, Edward C. Pickering (1846–1919), is often considered progressive and liberal for employing women. He was called a “true Victorian gentleman in his attitude towards women and to everyone, men and women alike,” by one of his computers.47 Pickering worked closely with a female benefactor, Mrs. Anna Palmer Draper, to finance the activities of the observatory, and he encouraged an assistant to teach mathematical astronomy at Radcliffe College, the new women’s school affiliated with Harvard.48 Yet Pickering was motivated as much by economy as by altruism. “To attain the greatest efficiency,” he wrote, “a skillful observer should never be obliged to spend time on what could be done equally well by an assistant at a much lower salary.”49 The salary he offered to the women was half the prevailing rate for calculation. “[The Harvard] computers are largely women,” complained the director of the Naval Observatory in Washington, “who can be got to work for next to nothing.”50

  14. Computing room of the Harvard Observatory

  “Pickering’s Harem,” as the group would occasionally be called, served as an uncomfortable example to the government computing agencies.51 When the secretary of the navy asked why the Naval Observatory could not reduce its expenses for computation, as the Harvard Observatory had done, he was met with a defensive reply from the superintendent. He claimed that the Naval Observatory paid “its employees at exactly the same (or in some cases less) rate as in other branches of the Government Service,” deftly deflecting the issue of hiring women. Shifting to a more aggressive position, he argued, “To charge extravagance against the Observatory because its employees are paid according to a rate fixed by law for the public service at large is clearly disingenuous and tending to mislead.”52 The Naval Observatory would not hire a female computer until 1901.53 The Coast Survey and the Nautical Almanac, who had benefited from the labors of Maria Mitchell, were slightly more progressive and hired their second female computers in 1893. However, the Nautical Almanac was self-conscious about this action and identified its new employee as a man.54

  The Harvard Observatory has left an unusual document that suggests the daily routine of its computing staff and the challenges faced by the female computers. This document is a musical parody, entitled the Observatory Pinafore, based on W. S. Gilbert and Arthur Sullivan’s operetta H. M. S. Pinafore. The parody was written by a junior astronomer, Winslow Upton (1853–1914), and so reflects the point of view of the astronomers, not the computers. It shows the women struggling with their work, confronting astronomers with problems, and working in an environment that would constrain their role or even deny that they were part of a scientific endeavor.

  The Observatory Pinafore must be one of the first parodies of Gilbert and Sullivan’s opera, a show that has been adapted and modified many times. Winslow Upton probably saw the original H. M. S. Pinafore during its American premier in November 1878. This production, unauthorized by Gilbert and Sullivan, debuted in the old Boston Museum of Art six months after the show opened in London. Upton acquired a copy of the vocal score and wrote his observatory version nine months later “during four rainy days of an August vacation in Vermont.”55 He replaced the opening chorus of sailors, who tell of their shipboard life, with a chorus of computers, who sing:

  We work from morn till night

  For computing is our duty;

  We’re faithful and polite.

  And our record book’s a beauty;

  With Crelle and Gauss, Chauvenet and Peirce,

  We labor hard all day;

  We add, subtract, multiply and divide,

  And we never have time to play.56

  The song is in the same spirit as the original and suggests that Upton is going to follow Gilbert and Sullivan’s original plot. In H. M. S. Pinafore, a young sailor is in love with the daughter of his captain. Prevented from marrying by the social gulf between them, the tw
o spend much of the opera planning to elope. “Love levels all ranks,” says one character, “but not so much as that.” The show is filled with trios and dances, a silly song by a pompous bass singer, a touching lover’s lament for the soprano, and a rousing chorus expressing English superiority. Near the end, one character announces, much to everyone’s surprise, that the captain and the young sailor were switched at birth, an announcement that immediately elevates the sailor and allows him to wed his love. The other characters gleefully ignore the logical problems with this solution, notably that the young sailor is now old enough to be the father of his bride, and sing a rousing recap of the songs.

  Upton could have chosen to adapt Gilbert’s plot to the Harvard Observatory. He might have decided to have a young female computer, perhaps a recent graduate of Cambridge Girls’ High School, fall in love with an assistant astronomer who is a member of a prominent Beacon Hill family. After two acts of rewritten dances, solos, and choruses, we might discover that there was a mix-up in the two families which, after an extended lawsuit, reverses the fortunes of the two young people and allows them to marry. However, Upton did not follow this approach, as he was not interested in exploring the relations between men and women. Instead, he chose to explore the love between a young man and his science, the love that a junior astronomer had for his position at the Harvard Observatory.

  In the Observatory Pinafore, Upton recasts the female love interest as a young male astronomer, though he retains the female name “Josephine” from the original.57 The male Josephine is about to lose his position at Harvard. With no other positions available, he intends to take a job at an inferior private observatory in Rhode Island. Upton expresses nothing but scorn for the Rhode Island observatory. It is the hobby of a wealthy businessman, and hence it is tainted, less virtuous than the Harvard facility. The script portrays the businessman and the director of his observatory as vain and stupid. “I’m very proud of my degree,” sings the Rhode Island astronomer, “For it shows that I’m a man of extraordinary sense.”58 As the plot moves through its paces, the Josephine astronomer bemoans his departure from Harvard and eventually finds a way to demonstrate his astronomical skill by repairing a telescope. When he fixes the device, his value is recognized, and he is allowed to remain in Cambridge.

  The only computer identified by name in the Observatory Pinafore is Rhoda Saunders. She may have been chosen because of her role in the observatory, because she was a friend of Winslow Upton, or because she possessed the best voice among the female employees. The script suggests that she was strong and confident of her skills. In a scene which has no parallel in the original Gilbert and Sullivan play, she defends her work to an agitated assistant astronomer. “Oh! Oh! Oh! Oh! Oh! It’s all wrong! A fearful mistake!” exclaims the astronomer, who has just been looking at a sheet of reduced data. When the figures are identified as the work of Saunders, she quickly affirms, “I don’t believe it’s wrong.”59 The observer eventually calms down and agrees with her. As the script was intended to entertain the staff of the Harvard Observatory, this scene is probably an exaggerated version of a real event, though it gives us no real assessment of Saunders’s character. She may have been strong and outgoing or quiet and retiring. Even if it could give us a fuller picture of Saunders, it would be of little use, for the scene is one of the last times that a computer appears in the original script. By the end of the first act, the female computers all but disappear from the show.

  Though Upton changed the gender of the female romantic lead and gave most of her songs to a man, he was not such a fool as to give the best soprano song to a male voice. He assigns the lover’s lament, the song “Sorry her lot,” to Rhoda Saunders and uses it to express some of the frustrations of the computers. In the original opera, this song is one of the points where Arthur Sullivan’s music transcends the comedy of Gilbert’s words and expresses a few moments of honest human emotion. It is stately, with a delicate melody that falls from high notes to low. In H. M. S. Pinafore, the falling melody is a fall of resignation, a recognition that love may be eternal but that courtship and marriage are governed by the conventions of society. “Weary the heart that bows the head,” she sings in the last verse, “When love is alive and hope is dead.”

  Upton’s version of the aria nearly matches the language of the original. It is his best writing in the entire script. Saunders, the computer who had held her own against the charges of an assistant observer, sings of repetition, tedium, and self-doubt.

  Sorry her lot, who adds not well;

  Dull is the mind that checks but vainly;

  Sad are the sighs that own the spell

  Symboled by frowns that speak too plainly.60

  To those unversed in arithmetic, computing was indeed a difficult task, but Saunders was a good computer and served at the observatory for thirteen years. She, as well as the entire staff, had known the times when fatigue made any mental labor difficult. Under such circumstances, computers could barely add a column of figures and would make error after error in the simplest of calculations. Upton underscores the problem of fatigue by contrasting the computers, who work during the day, against the observers, who work at night.

  Happy the hour when sets the sun;

  Sweet is the night to earth’s poor daughters,

  Who sweetly may sleep when labor is done,

  Unlike their brother astronomers.61

  The sweet sleep is only a brief respite from their days of toil. Upton reinforces the tedium of the work through the chorus:

  Heavy the sorrow that bows the head

  When fingers are tender and the ink is red.62

  The computers used red ink to correct their mistakes. A page of red figures meant the day had been hard and frustrating.

  Once the song is finished, Saunders has only a few more lines in the script about mistakes and red ink before her role is finished and she leaves the stage. Just before she departs, there is a brief exchange that suggests the anxieties created by the presence of women at the observatory. The Rhode Island astronomer asks Pickering about the size of the computing staff. Pickering replies that it is “quite large—most enough for a good dance in spare hours.” After hearing this, the other astronomer exclaims, “Do you allow, sir, your assistants to dance? Physicians tell me it is promotive of inaccuracy in computation.” To modern sensibilities, this response is a quaint remnant of Victorian superstition, but it actually expresses a deep anxiety of the times. No one was entirely confident that men and women who were unrelated by family ties could work together in offices without being influenced by physical attraction. Taken at face value, the script suggests that E. C. Pickering may not have thought about this question, for the observatory captain responds, “Indeed! I was not aware of that.”63 However, it seems likely that Upton is being ironic, for Pickering was comfortable with the presence of women in his observatory, while the leaders of the Naval Observatory and the Nautical Almanac were not.

  In 1880, shortly after Winslow Upton had finished the Observatory Pinafore, the director of the real Harvard Observatory, E. C. Pickering, announced that “a handsome carpet has also been purchased for the Computing Room in the east wing.”64 It was a pleasant addition to the room, for it deadened excess noise and provided a layer of insulation against the frozen ground of the Massachusetts winter. Pickering undoubtedly purchased it with all goodwill, yet the carpet was also a symbol of divided labor, of tasks for men and women. The literature of the time talks of separate spheres for the two genders, a metaphor that borrows from the classical model of astronomy. The sphere of Mars encircles the sphere of Venus, but both spin on the same axis. Men could move freely through the observatory, even into the telescope room, “a dark and dingy place,” Upton tells us, “all clattered up and smelling strong of oil.”65 The women had their computing room, with its desks and its carpet.

  The Harvard Observatory marks the end of experiments with dividing the labor of computation. Subsequent computing offices draw their models, consc
iously or unconsciously, from the laboratories that were founded before 1880. The Harvard Observatory also marks the end of astronomy as the dominant force in scientific calculation. Indeed, the Observatory Pinafore makes it clear that orbits and positions are no longer the most interesting part of astronomical research. Upton’s writing transforms the rousing chorus of British superiority at the end of the play into a song of praise for a photometer, an instrument that measures the brightness of stars.66

  PART II

  Mass Production and New Fields of Science 1880–1930

  The engineer, the astronomer, the mathematician, the electrician, form a mighty and always increasingly important army of male labourers.

  Olive Schreiner, Woman and Labor (1911)

  CHAPTER SIX

  Looking Forward, Looking Backward: Machinery 1893

  As the machine is truer than the hand, so the system, which does the work of the master’s eye, turns out more accurate results.

  Edward Bellamy, Looking Backward (1888)

  ON A TYPICAL WORKDAY in 1881 or 1882, Rhoda Saunders would begin her calculations in the Harvard Observatory computing room by picking up her pen, uncorking her bottles of ink, one black and one red, and opening her computing book. All of these objects would have been familiar to Edmund Halley in the seventeenth century or Nicole-Reine Lepaute in the eighteenth, but each had been subtly changed by industrialization. Saunders’s pen had a preformed steel nib that easily outlasted the hand-cut point of a goose quill. Her bottles of ink were commercially produced and varied little from batch to batch. The lines in her computing book, printed by a mechanical press, were straight and true, without any of the wiggles or cramped margins found on the hand-ruled sheets of Nevil Maskelyne or Maria Mitchell.