Computing is our duty,
We’re faithful and polite,
And our record book’s a beauty.
—From The Observatory Pinafore
It is only with great difficulty that one can imagine what it was like to be a computer at Harvard Observatory a hundred years ago, not a soulless machine of wire and silicon but a living, breathing young woman. Her name was Henrietta Swan Leavitt and her job was counting stars.
Today this kind of work is done by machine. Arrays of electronic sensors grab images of the sky, long streams of digits for computers to analyze. In the late 1880s, when Harvard embarked on a marathon project to catalog the position, brightness, and color of every star in the sky, the closest things to a modern digital computer were clunky mechanical calculators like the Felt & Tarrant Comptometer or the Burroughs Arithometer, with their rows of clacking buttons, stiff hand-pulled levers, and ringing bells. And there was the human brain. Diligent souls like Miss Leavitt—they actually were called computers—were paid 25 cents an hour (10 cents more than a cotton mill worker) to examine blizzards of tiny dots, photographs of the night sky. They would measure and calculate, recording their observations in a ledger book.
Observatory Hill, 1851 (Harvard College Observatory)
Imagine a sky with the colors reversed, cold black stars sprayed against a firmament of white. These photographic negatives were produced when a telescope was trained at the heavens, its light focused onto a large glass plate coated on one side with light-sensitive emulsion—a forerunner to photographic film. Today, half a million of these fragile plates are stored in a brick building adjacent to the one where Miss Leavitt and the other computers worked. Fearing an earthquake might shatter this database of glass—the astronomical equivalent of the burning of the library of Alexandria—Harvard built the repository as two nested structures. Physically isolated from the building’s exterior shell, an internal matrix of steel beams and flooring rests, the story goes, on an apparatus of leaf springs, like those in a wagon or an old pickup truck.
The result is an invaluable archive of how the sky looked on different nights since the first surveys were done in the 1880s. Among the most precious items in the collection are pictures of the Magellanic Clouds. We know them now as neighboring galaxies, companions to our Milky Way. Back then no one was quite sure what they were. Hunched over the plates in an observatory workroom, Miss Leavitt found the pattern that eventually led to the answer. She discovered a way to measure beyond the galaxy and begin mapping the universe.
Today nearly every scientifically literate person knows, or thinks he knows, that our planet circles an unremarkable star lost among galaxies of galaxies extending billions of light-years in every direction. One can almost hear Carl Sagan intoning the words on public TV. We’ve learned to revel in our insignificance. As far as most astronomers are concerned, only small details are still in dispute: Is the universe 13.9 billion light-years in radius or just 13.8 billion? So much confidence exudes from these discussions that a spectator may forget to ponder the most basic question: How can anyone know for sure?
Suppose two stars seemingly of equal brightness are shining side by side against the dark dome of night. Knowing nothing else about them, one might conclude that they are equally far away. But that would be true only if the stars happened to be emitting, from their nuclear furnaces, the same amount of light. More likely one star is more powerful than the other yet shining from farther away. How much more powerful and how much farther? Barring a breakthrough in interstellar space travel, there seemed to be no way to find out.
The same uncertainty applied to the faint hazes called nebulae, clouds of light. Were they sprawling galaxies, “island universes” shrunken by their great distance? Or were they little gas clouds right here in the Milky Way? With no means of measuring the universe, the question was almost theological. How many angels can dance on a pinhead? How far away are the stars in the sky?
TODAY A VISITOR to Observatory Hill, a low rise on Garden Street about a fifteen-minute walk northwest of Harvard Square, looks in vain for a sign that anything cosmic happened there. Dwarfed by the giant mountaintop observatories at Palomar, Wilson, Cerro Tololo, and Mauna Kea and blinded by the light of the Boston glare, the Harvard telescope, called the Great Refractor, is now in retirement. But when it saw first light in 1847 it was one of the most powerful in the world.
It had arrived, some liked to say, on the tail of a comet—the March Comet of 1843, which burned so bright that it was visible in broad daylight, a signal to some that the Day of Reckoning was at hand. (A group called the Millerites had used biblical passages to predict that Jesus would make his Second Coming sometime between March 21, 1843, and March 21, 1844. The comet was right on schedule.) Those with a scientific bent felt a deeper kind of awe. Where did the comet come from and when might it return? For answers they could turn to the observatories at Cincinnati or at Yale or Williams colleges. But Harvard didn’t have a good enough telescope to properly study the phenomenon. Even the Philadelphia High School was better equipped.
It was an embarrassment Bostonians vowed to correct. Twelve acres called Summer House Hill had recently been purchased by Harvard as an eventual site for a large telescope, but little progress had been made. Now the project began in earnest. Well-to-do citizens took out “subscriptions,” some $25,000 worth, to build the best observatory in the world. To ensure as stable a platform as possible, the building was constructed around a massive granite pier rooted 26 feet into the ground and rising from the bedrock to the observatory floor. Beneath a 30-foot dome was placed the Great Refractor. The mahogany-veneer tube was outfitted with a lens, 15 inches across, that had been ground by master craftsmen Merz and Mahler of Munich, who were told to make it at least as powerful as the one the Russians had recently purchased for the Imperial Observatory. The space race had begun.
The Great Refractor (Harvard College Observatory)
The first astronomer to peer through the glass was rendered nearly speechless: “It is delightful,” he wrote, “to see the stars brought out which have been hid in mysterious light from the human eye, since the creation. There is grandeur, an almost overpowering sublimity in the scene that no language can fully express.”
With this new tool, astronomers quickly discovered the inner ring of Saturn and, outfitting the telescope with a photographic plate, took the first picture of a star.
2
On a clear night high in the mountains when the air is cold and dry, the brightest stars shine some two hundred and fifty times brighter than the faintest ones, those that can barely be discerned with the naked eye. The ancient Greeks divided this stellar multitude into six categories. The brightest lights were said to be of the first magnitude while the dimmest were of the sixth.
This rough gauge has been refined over the centuries so that each step now means an increase in brightness of about two and a half times. The actual figure is closer to 2.512, conveniently making a fifth magnitude star 2.512 × 2.512 × 2.512 × 2.512 × 2.512 or 100 times dimmer than a star of the first magnitude. A sixth magnitude star is about 250 times dimmer than that, and a seventh magnitude star about 600 times dimmer. (In the original, rather roughshod system, all the brightest stars had been bunched into magnitude one. Measured more precisely, some of them have ended up with magnitudes of less than one, and the brightest with magnitudes of less than 0. Blazing Sirius is about –1.4.)
Centuries ago, with his simple spyglass, Galileo had amplified his vision enough to see stars as faint as the eighth magnitude. The Great Refractor extended the reach to the fourteenth magnitude, resolving images some 400,000 times dimmer than what could be seen from Earth with 20-20 eyes.
With the ability to see farther than ever before, Harvard embarked, in the late 1870s, on the kind of exhaustive search that would become its hallmark, precisely cataloging the brightness of every star in the sky. The observatory was now being run by a young physicist named Edward Charles Pickering, who had made his mark as a professor at the Massachusetts Institute of Technology by establishing the first curriculum in the country where students could confront the ideas of physics head-on—in laboratory experiments, poking at nature and carefully recording the results. He got an early taste of astronomy when he served on two government expeditions to observe total eclipses of the sun. When he was hired in 1876 to take over the observatory, he was thirty years old.
Until this time astronomy had focused on trying to establish two primary details about every star: its position and its motion through space. Pickering was struck by how very little good data had been gathered on two equally important characteristics: a star’s precise brightness, a clue to its distance, and color, a clue to what chemicals it contained. Pickering was a fastidious measurer. He occupied himself on hikes in New Hampshire’s White Mountains by measuring out the terrain, using an instrument he had fabricated himself. His mission, he decided, and that of the observatory, would be to amass mountains of data, about which others could theorize.
Good old-fashioned astronomy is what he wanted. No big bangs, no black holes, no dark matter—this was way before all that. Space was still flat and of no more than three dimensions. Understanding the universe meant charting little lights as they moved across the sky.
He started with stellar brightness. In the past, astronomers had made some progress along these lines with a German instrument, the Zöllner astrophotometer, which compared a star’s brightness to the glow of a kerosene lamp. Focused through a pinhole and reflected by a mirror into the visual field of a telescope, the dot of lamplight appeared as a tiny sun hovering beside a star. The observer adjusted the instrument, stepping down the brightness of the artificial star until, in his judgment, it matched its companion. Then its magnitude could be recorded. (Some of these demanding measurements had been carried out by an observatory employee named Charles Sanders Peirce, who came to be known as one of the most brilliant and eccentric philosophers of all time.)
Pickering felt that a definitive survey should rely on a standard more universal than the brightness of lamplight. He devised an instrument with an arrangement of lenses and mirrors that would allow any star within view to be lined up side by side for comparison with the North Star, which was set, somewhat arbitrarily, at magnitude 2.1. Once astronomers learned how to use the new device, they could knock off as many as one star a minute. Eventually Harvard measured and cataloged forty-five thousand of them.
That was barely a beginning. Within the gaps between stars there were surely many more, so dim and far that they did not register on the retina of an eye, even one fitted with such powerful lenses. To see farther, light from these faint sources would have to be gathered during a time exposure, accumulated on a photographic plate attached to the end of a telescope. Mounted on a rotating platform and driven by mechanical clockworks, the telescope could track a star as it arced across the sky, pooling its light photon by photon, chemically etching an impression.
The astronomical leverage this provided was stunning. From Earth the Pleiades appear as a subtle glow engulfing seven bright points of light—the “Seven Sisters” of Greek mythology, pursued by Orion. Galileo had already seen through his telescope that the sisters were joined by dozens more. A three-hour time exposure, taken in Paris, revealed that the constellation included more than 1,400 stars.
More stars still could be unveiled by mounting telescope and camera as far above sea level as possible, cutting through miles of atmospheric distortion. After unsuccessful attempts to establish stations at Pikes Peak in the Colorado Rockies and Mount Wilson in Southern California, Pickering decided to try the high reaches of Peru. He dispatched an expedition led by a trusted colleague, Solon I. Bailey, who established a temporary post atop a peak that, it was decided, would now be called Mount Harvard. Bailey hadn’t reckoned on the length of the annual rainy season and was forced by the clouds to find a clearer site, finally settling in a remote town called Arequipa. This time the location seemed perfect. Pickering arranged to have an observing station sent piece by piece, from Boston Harbor around the tip of South America. Included among the cargo were the components of the 24-inch Bruce Telescope (named for the heiress who paid for its construction, Catherine Wolfe Bruce).
For all Pickering’s hopes, the project got off to a bad start. His brother William, as headstrong and arrogant as Edward was modest and reserved, was placed in charge of Arequipa, mismanaging the operation and scandalizing the world astronomical community when he began dispatching outlandish scientific reports to that great academic journal the New York Herald. Ignoring his assignment to study the stars, he trained the telescope on Mars, enthusiastically describing huge mountain ranges rising above giant rivers and lakes extending hundreds of square miles—a geography that remained stubbornly invisible to any eyes but his own.
While Edward Pickering concentrated on damage control at home, Bailey was sent back to Peru to retake the observatory. Before long, the Arequipa station was shipping crate after crate of photographic plates north to Cambridge, the first pieces of what would become a mosaic of the entire southern sky.
With so much new information to digest, astronomers were soon overwhelmed. They were faced with an embarrassment of riches now familiar throughout science, a burgeoning glut of undigested data begging to be categorized. That is where the computers came in.
3
“A great observatory should be as carefully organized and administered as a railroad,” Pickering once observed. “Every expenditure should be watched, every real improvement introduced, advice from experts welcomed, and if good, followed, and every care taken to secure the greatest possible output for every dollar expenditure. A great savings may be effectuated by employing unskilled and therefore inexpensive labor, of course under careful supervision.”
Edward Pickering (Harvard
University Portrait Collection)
Imagine trying to find people to do such precise work for 25 cents an hour—what amounted to the minimum wage. Today the job would probably have to be farmed out to star-counting sweatshops in Asia. For the late nineteenth century, computing wasn’t such a bad deal. Seven hours a day, six days a week, the job paid $10.50 a week and included a month’s vacation. Not many men were interested in the tedious work, so the positions went mostly to women. (The tradition was a long time in passing. As recently as the early 1960s, Brookhaven National Laboratory hired Long Island housewives to pore over the tangled images of subatomic particles, looking for patterns that might foretell a new physics.)
Recognizing that his housekeeper, Williamina Paton Fleming, was overqualified for mopping floors, Pickering hired her as one of his first computers. (Abandoned by her husband after emigrating from Scotland, she was grateful enough to name her son, born that year, Edward Pickering Fleming.) She eventually became curator of the collection of photographic plates, doubling her salary, and was in charge of classifying stars according to their spectra, the colors revealed when their light was refracted through a prism. This was another of the observatory’s ambitious efforts, resulting in a monumental work called the Henry Draper Catalogue, named for an accomplished and wealthy amateur astronomer who had taken the first photograph of a nebula. Funded by Draper’s widow, the compendium provided employment for two other computers, Annie Jump Cannon and Antonia Caetana Maury.
This women’s work sometimes resembled bookkeeping more than scientific research. Pickering tried to make it reasonably stimulating and treated his computers with respect. But he was intent on the observatory’s getting its money’s worth.
“He seems to think that no work is too much or too hard for me no matter what the responsibility or how long the hours,” Mrs. Fleming complained in a diary. “But let me raise the question of salary and I am immediately told that I receive an excellent salary as women’s salaries stand.”
If he would only take some step to find out how much he is mistaken in regard to this he would learn a few facts that would open his eyes and set him thinking. Sometimes I feel tempted to give up and let him try some one else, or some of the men to do my work, in order to have him find out what he is getting for $1,500 a year from me, compared with $2,500 from some of the other assistants.
Does he ever think that I have a home to keep and a family to take care of as well as the men? But I suppose a woman has no claim to such comforts. And this is considered an enlightened age! … I feel almost on the verge of breaking down.
When she asked him for a raise, Pickering agreed to pass on the request to the university president. Money was always tight. This was before the era of government-funded big science, and observatories were dependent on the charity of rich benefactors, and on people with a monastic dedication to the craft. Pickering worked as hard as any of them, administering by day, stargazing by night. When the sky was cloudy he would do calculations late into the evening, sometimes with an assistant reading to him for entertainment (Shakespeare was a favorite). Considering the long hours, his $3,400 annual salary came to much less than two dollars an hour. (He and his family also got to live in the less-than-luxurious director’s residence on Observatory Hill.) No one was in this for the money.
“An astronomer is a sorry soul,” began a chorus in The Observatory Pinafore, a parody of Gilbert and Sullivan’s comic operetta H.M.S. Pinafore written by one of Pickering’s assistants.
He must open the dome and turn the wheel,
And watch the stars with untiring zeal,
He must toil at night though cold it be,
And he never should expect a decent salaree.
Most of the time, the computers seemed to enjoy their jobs, making light of the low pay and somewhat Dickensian working conditions. In The Observatory Pinafore, one of them, “Josephine,” sings of her toil in the “dark and dingy place, all cluttered up and smelling strong of oil,” apparently from the furnace that had recently replaced fireplaces for taking the edge off the New England cold. At another point in the story, a whole chorus of computers breaks out in song:
The Observatory Pinafore (Harvard College Observatory)
We work from morn till night,
Computing is our duty,
We’re faithful and polite,
And our record book’s a beauty.
It is tempting to imagine Henrietta Swan Leavitt joining in the song. But that couldn’t be. Though written in 1879 the musical was not actually performed until New Year’s Eve 1929. By then she had already died.
