Computers at Work: Astronomical labor at the HCO at the turn of the century

The women whose assistance, administration, and computing supported astronomy are now great subjects in the history of science. Far from serving an auxiliary role, these computers standardized the skies, and laid the foundation for modern astrophysics. While we know big names who pioneered new theories through their calculations, many of these women remain anonymous, their work part of a collective effort towards the sciences. This exhibit samples brief moments of that effort. Once referred to shallowly as “Pickering’s harem,” they now appear all at once as brilliant observers, administrators, and scientists.

Detail from plate AM3695. HCO Plate Collection.

Observatories tend to impart a mystical aura for those who work within, perhaps for some reason to do with deep cosmic secrets. A journalist writing for the Boston Evening Transcript in 1881 fully embraced this feeling, describing the observatory dome bathed a “dim, religious light.” In contrast, writing for the New England Magazine the following year, Helen Leah Reed described the observatory library, and the women assistants who tended the books and worked at the tables, with industrious cheer, “well-lit…workrooms, as they are.” The contrast between the dark and mystical observation room and the bright, productive work-floor reveals the public attitudes towards women scientists of the era. Denied the privilege of telescopic observation and a direct lens to the cosmos, they were relegated to industrious spaces with clear models of production. They were not hired to be philosophers; they were assistants. However, their collective work fit into a new type of standardized observational astronomy, using photography, a technology that restructured the way astronomers studied the skies.

The photographic revolution of the mid-nineteenth century had exploded by the time Edward Pickering assumed directorship over the Harvard College Observatory (HCO) in 1877, to such an extent that it became feasible to document and analyze the sky with more detail and resolution than eye based observations. Glass plates were exposed at night by (men) astronomers, and in the daytime, (women) assistants reduced values, computed magnitudes, and organized their findings into catalogues and publications.

This division of labor reflected many of the gendered stereotypes epidemic to women in the workplace. Historians of late-nineteenth century industrial labor use the term “women’s work” to discuss the types of labor that suited women who needed a wage, and also had to tend to domestic duties.  Not only could a woman take work home to complete in “off” hours, she could often be paid less than a man for the same work. Pickering did have qualms about paying his computers (who were often college educated) less than a dignified wage, but still relegated most of their intellectual productivity into computing and administrative work. Attendant to “women’s work” was a paternalistic attitude that aimed to protect the women from harsh working conditions. In the case of astronomy at HCO, Pickering infamously restricted all women but Annie Jump Cannon from cold and windy nighttime telescopic observations (interestingly, she was permitted to do so in the dead of winter), and he also reduced daytime working hours for observatory staff in the stifling summer months.

Observatory computer room and staff, 1881. Harvard University Archives.
Standing (L-R): Edward Pickering, Williamina Fleming; Seated (below β Aurigae graph): Antonia Maury; Remaining unknown.

Still with such restrictions, the computers found many areas on the job in which to contribute to astronomical theory. Most notably, Williamina Fleming, Antonia Maury, and Annie Jump Cannon devised stellar spectra schemes while classifying new stars, and Henrietta Leavitt used her analysis of Cepheid variables to determine galactic distances. But the collective work done by the women reached much deeper, translating the massive supply of photographic data into standardized measurements, providing the foundations for developing astrophysical theories.

The following items only lightly touch on a few of the many tasks accomplished by these many women, only too few of them named here. While many of the computers dwell silently in the archives, their voices emerge in a recitation of their work. Those who left memoirs, primarily Williamina Fleming and Cecilia Payne-Gaposchkin (although she was not a computer), speak both about the difficulties for working-women in a traditionally masculine workplace, but also about their excitement and enthusiasm for their work. The turnover rate for a computer was steady — it seems Pickering hired one or two new assistants each year to keep a roster of around twenty — but those who had found the job had done so out of passion. Whether to turn their astronomical education into a career, work with family, follow a legacy, or simply find a source of income amenable to their busy lives, the women at the HCO found the opportunity to engage with the current of astronomy, and contribute to the building of the universe.

A comparator and loupe, for magnification. HCO Plate Collection.

The primary tool-kit of a computer consisted of a few key items: the glass-plate photograph, magnifying and comparing device (to figure apparent brightness), a log book, and, in the case of a speedy observer like Cannon, a second set of hands. In addition to magnitudes and classification, the computers would also analyze these plates as part of an informal astronomical plate reference program for remote astronomers. Pickering described his plan in HCO Circular 123 to “furnish copies of [observatory] photographs at cost to whoever might desire them.” The result of this call was torrent of reference work. Requests came in by letter or telegram, which the assistant brought to the plate stacks, returning copies of relevant photographs with annotations and reductions. The extent and scope of this work is poorly documented, but likely permeated the entire discipline at the turn of the century.

Observatory data analysis by woman computers, c. 1880-1920. Harvard University Archives.
(R-L): Ida Woods, Williamina Fleming (standing), Eve Leland; Remaining unknown

An example of reference work comes from Ida Woods, an astronomer awarded for her prolific discoveries of novae. The observatory received a letter in January 1925 from B. Asphind in Sweden, requesting plates from 1901 that might show the “Wolf Planet” — also known as Wolf’s Asteroid. Woods spent around two weeks searching these plates before submitting her report to Asphind, along with photocopies annotating potential asteroids (these had been checked by her colleague Mabel Gill). The task was not over, however, and Asphind corresponded with HCO concerning positions of the Wolf and other asteroids for many years.

Ida Woods’ notes on the Wolf Asteroid, and a photo copy of plate B28570. Project PHaEDRA.

Woods and Gill both worked at the HCO in the 1910s and 1920s, when the observatory had affirmed its position in the astronomical community as a “clearinghouse” for astronomical data. With so many requests flowing, most plates ended up in multiple hands for multiple reasons. Searching for her proud nova, Woods found an “excellent multiple image plate” of the region (but no nova) that had been a favorite of Williamina Fleming’s, fifteen years before. This plate, AM3695, might seem impossible to decipher, but those immersed in the glass plate aesthetic knew well how to tell one point of light from another.

Detail of Plate AM3965, on which Williamina Fleming searched for the star Algol, and Ida Woods for a nova. HCO Plate Collection.

The opportunity for such close analysis of plates led many of the women to discoveries in their own right. As mentioned above, Woods received awards for her discovery of novae, and these were published crediting her observation. However these credit lines often were not identical to the author line, leading to moments of indirect credit such that Solon Bailey, in HCO Bulletin 714, published under his name that Woods had made the discovery. Pickering occasionally did the same, but more frequently prioritized the worker’s names, both in font size and page positioning.

Detail of title page to HCO Annals Vol. 28 part 1, featuring work discussed by Antonia Maury

Even computer work as mundane as reductions and conversions received gestures of credit in publication, but the praises usually only fell on the project leader. A catalogue of stars found using the Meridian Circle, in production since 1870, was published in 1896 as part of the Harvard College Observatory Annals, Vol. 12 part 2, with a brief mention of Anna Winlock and R. G. Saunders, “who rendered important aid in the computations through their entire progress.” However, we know from logbook analysis as part of Project PHaEDRA (see below) and other historical scholarship, that at least four other women participated in the project: Selina Bond, Mary Farr, M. E. Ray, and a certain Miss Merrill.

That these women are not credited in this project likely results from the circumstances of their employment. While Winlock, Bond, and Saunders belonged to families associated almost dynastically with the HCO, these other women seem to have sought employment simply for the wage. Pickering offered $500/year (around 25 cents/hr) for entry level assistants, with a reduced wage if they should choose to work out of the home. In general, “outwork” was an important feature of women’s labor in the late nineteenth century, freeing space at the observatory, and maximizing the computing time of workers. Often this work fell on top of women’s expected (and unpaid) duties at home, a struggle that Williamina Fleming discusses in her diary. Working at home, these women were less imbricated in observatory life, and stood less prominently in the minds of their supervisors, appearing instead as cogs in the machine.

Outwork, and computer work more generally, fell into a hierarchy whereby one head computer (most famously Williamina Fleming) would communicate with Pickering and the other male astronomers. In the case of Anna Winlock’s meridian circle reductions, she arranged her teams’ calculations, questions, and ideas into regular letters to the project lead William Rogers, who had moved from his position at HCO to a new professorship at Colby College. No longer employed by Harvard, but still with permission to use their equipment and photographs, Rogers found it easier to hire these workers out of the home rather than contest with Pickering at the HCO for their daytime labor.  Winlock’s correspondence to Rogers betrays nothing beyond work-related problems, but the pressures must have been intense. Selina Bond, daughter of the first director and assistant at the HCO for over 40 years, wrote Pickering in 1897 after the project’s completion, that she “felt the need for rest and change” after so many years engaged in computer work.

Correspondence book from Anna Winlock and Selina Bond to William Rogers. Project PHaEDRA.

Perhaps the most dreaded role assigned to women at the observatory was that of editor for the institution’s publications. It was not an enjoyable task for women with big ideas to parse through others’ incorrect and lazy work. Both Williamina Fleming and Cecilia Payne-Gaposchkin cursed the task, although Fleming was more decorious (“revised…I hope, for the last time.”) While Payne-Gaposchkin appreciated the opportunity to work with words, she felt that “tabular matter is often an editorial nightmare.”

Williamina Fleming worked in a period in which photographic analysis, a new and unexplored method, impinged on the tradition of eye/telescope skywatching. HCO Annals Vol. 24, published in 1890 and containing a stellar catalogue calculated from plate values, fell under controversy when old-school astronomer Seth Carlo Chandler accused it of gross error. Fortunately, Fleming’s diligent eye corrected (or rather, vindicated) the many alleged errors. In the “Corrected Copy” held by the Wolbach Library, Fleming’s fine hand appears with a set of numbers every few pages. There were errors — human errors — but only a handful in a selection of twenty thousand values. Fleming had not only vindicated the photographic method, but also her attention to detail and the skill of her computing team.

Detail of page in HCO Annals Vol. 24, with Williamina Fleming’s revisions.

These assistants received too little credit then, but are finding proper recognition as we further understand the scope and impact of their effort. However, many still, for the most part, remain nameless. Historian Keith Lafortune has published a comprehensive list of accounts found on the HCO payroll in the years 1877-1919, which names over eighty women at the observatory. The biography of Henrietta Leavitt is certainly inspiring, but what more could Hester Levis say? How did Mollie O’Reilly or Edna Hawes experience the workplace, and contribute to astronomy? And who is Miss Merrill?

Project PHaEDRA has the potential to answer these questions! While the librarians at Wolbach are collecting basic metadata, there are also pages for transcription and study available at Smithsonian Transcription Center and our finding aid, respectively. Have a look, find a new set of initials, and maybe we can uncover who these workers were and what worlds they built.

Observatory Group [photographic group portrait], 1913. Harvard University Archives.

The following is a description of this photo from HCO historian Barbara Welther. This is a small sample of the women known to have worked at HCO in this era:

“At the far left of the photograph is Margaret Harwood (AB Radcliffe 1907, MA University of California 1916), who had just completed her first year as Astronomical Fellow at the Maria Mitchell Observatory. She was later appointed director there, the first woman to be appointed director of an independent observatory. Beside her in the back row is Mollie O’Reilly, a computer from 1906 to 1918. Next to Pickering is Edith Gill, a computer since 1889. Then comes Annie Jump Cannon (BA Wellesley 1884), who at that time was about halfway through classifying stellar spectra for the Henry Draper Catalogue. Behind Miss Cannon is Evelyn Leland, a computer from 1889 to 1925. Next is Florence Cushman, a computer since 1888. Behind Miss Cushman is Marion Whyte, who worked for Miss Cannon as a recorder from 1911 to 1913. At the far right of this row is Grace Brooks, a computer from 1906 to 1920. Ahead of Miss Harwood in the front row is Arville Walker (AB Radcliffe 1906), who served as assistant from 1906 until 1922. From 1922 until 1957 she held the position of secretary to Harlow Shapley, who succeeded Pickering as Director. The next woman may be Johanna Mackie, an assistant from 1903 to 1920. She received a gold medal from the American Association of Variable Star Observers (AAVSO) for discovering the first nova in the constellation of Lyra. In front of Pickering is Alta Carpenter, a computer from 1906 to 1920. Next is Mabel Gill, a computer since 1892. And finally, Ida Woods (BA Wellesley 1893), who joined the corps of women computers just after graduation. In 1920 she received the first AAVSO nova medal; by 1927, she had seven bars on it for her discoveries of novae on photographs of the Milky Way.”

— Barbara Welther. “Pickering’s Harem.” Isis 73 v.1 (1982)

Wooden plate stand, c. 1900. HCO Plate Collection. This, or a similar, stand can be seen in the “Observatory computer room and staff” photograph, above.

A ‘flyspanker,’ used to measure dot size. HCO Plate Collection. The initials DH stand for Dorrit Hoffleit, a computer in the 1930s who later earned fame for her astronomy teaching at Yale.










Archival Sources:

“A Night at the Harvard Observatory.” Boston Evening Transcript. December 23, 1881.
Bailey, Solon. “Harvard College Observatory Bulletin 714.” Harvard University. Cambridge, MA, 1920. John G. Wolbach Library, Center for Astrophysics.
Fleming, Williamina. “‘Diary of 1900.’ Chest of 1900.” Harvard University. Cambridge, MA, 1900. HUG 1396.2, Box 59. Harvard University Archives.
Harvard College Observatory. “Harvard College Observatory Observations, Logs, Instrument Readings, and Calculations,” n.d. John G. Wolbach Library, Center for Astrophysics.
———. “‘Selina Bond to Pickering (28 June 1897).’ Records of Director Edward C. Pickering, 1864-1926 (Inclusive).” Harvard University. Cambridge, MA, 1864. UAV 630.17. Harvard University Archives.
Payne-Gaposchkin, Cecilia. Cecilia Payne-Gaposchkin: An Autobiography and Other Recollections. Cambridge, Great Britain: Cambridge University Press, 1984.
Pickering, Edward Charles. “Harvard College Observatory Circular 123.” Harvard University. Cambridge, MA, 1907. John G. Wolbach Library, Center for Astrophysics.
Reed, Helen Leah. “Women’s Work at the Harvard Observatory.” The New England Magazine, April 1882.
Welther, Barbara L. “Pickering’s Harem.” Isis 73, no. 1 (1982): 94–94.
Bailey, Solon. The History and Work of Harvard Observatory, 1839-1927. Harvard Obseravtory Monographs 4. New York, NY: McGraw-Hill Book Company, 1931.
Bigg, Charlotte. “Photography and the Labour History of Astrometry: The Carte Du Ciel.” In The Role of Visual Representations in Astronomy: History and Research Practice, 90–106. Göttingen, Germany: Verlag Harri Detsch, 2000.
Grier, David Alan. When Computers Were Human. Princeton, NJ: Princeton University Press, 2005.
Haley, Paul. “Williamina Fleming and the Harvard College Observatory,” The Antiquarian Astronomer, no. 11 (June 2017): 2–32.
Jones, Bessie Zaban, and Lyle Gifford Boyd. The Harvard College Observatory: The First Four Directorships, 1839-1919. Cambridge, MA: Harvard University Press, 1971.
Lafortune, Keith. “Women at the Harvard College Observatory, 1977-1919: ‘Women’s Work,’ the ‘New’ Sociality of Astronomy, and Scientific Labor.” Master of Arts, University of Notre Dame, 2001.
McEachern, Maria. “Every Star Speaks for Itself.” Galactic Gazette (blog). Accessed July 20, 2018.

Alex McGrath

Archives Assistant at Wolbach Library
Alex is the Archival Assistant at Wolbach.