Mary Edwards sat down at the kitchen table, picked up her quill and began a long series of sums. Every few minutes, she stopped, flicked open a book of tables and began to run her finger down the columns until she reached the figure she needed for the next step in her calculation. Over and over, she added, subtracted and looked up figures until the columns began to wobble and the numbers jiggled on the page. It was time to stop. She couldn’t afford to make mistakes.
In 1784, Edward’s husband had died, leaving her with two children, a pile of debts and notice to quit their home. But she was smart, knew a thing or two about mathematics and had a reputation as a reliable number cruncher. While other women scraped a living making lace or stitching gloves, Mary Edwards was a home computer, preparing the astronomical tables sailors relied on to navigate their way across the oceans.
FOR a young country clergyman, John Edwards was beginning to do well for himself. In 1774, he had just been made curate in a little village in the English county of Shropshire. Curates didn’t earn that much and Edwards had both a young family and an expensive hobby to support. He was more interested in telescopes than theology, and his attempts to make better instruments were proving costly. No matter – he had ways of supplementing his salary. He took on paying pupils and he was a paid “computer” for the Nautical Almanac, a collection of data that enabled sailors to plot their ship’s course accurately. The first had a drawback: teaching ate into his time with his telescopes. The second suited him better, for as computer historian Mary Croarken has discovered, he left most of the work to his wife.
Advertisement
By the 1770s, the Nautical Almanac had become an indispensable aid to navigation. The almanac was the brainchild of Nevil Maskelyne, the Astronomer Royal and a member of the Board of Longitude, a body set up in the early years of the 18th century to solve the problem of finding longitude at sea.
Published annually, the almanac contained all the data a sailor needed to pinpoint his ship’s position – a sort of ready reckoner for calculating latitude and longitude. The key parts of the almanac were the “lunars”, the predicted distances between the moon and the sun and certain other stars at specified times on a given day in Greenwich – that is, at longitude zero. To a navigator these were as useful as a clock keeping Greenwich time. To find longitude, he needed to know the difference between the time aboard ship – which he could check by the sun – and the time at Greenwich. Every hour difference is equivalent to 15 degrees longitude. And to find the time at Greenwich, all the navigator had to do was to measure the distance between the moon and a suitable star and look the figure up in the almanac.
Before the almanac, voyagers could work out where they were by the lunar distance method, but they needed impressive mathematical skills and detailed knowledge of planetary motion. Even then, the calculations took around 4 hours. Maskelyne’s great innovation was to employ a team of human computers to take the slog out of the job by calculating lunar distances at Greenwich in advance. This reduced the shipboard part of the job to a set of simple astronomical measurements and half an hour of easy sums.
The first of Maskelyne’s almanacs appeared in 1767. It was a success, and the Board of Longitude wanted more. But it was a lot of work. The daily lunar distances vary from year to year so a ship leaving on a long voyage might require tables for several years ahead. Maskelyne needed more computers to do such a vast number of calculations. He began to recruit a network of clergymen and schoolmasters, men with a working knowledge of maths and a basic grasp of astronomy. In 1773, Maskelyne was introduced to John Edwards. With his passion for telescopes, and an obvious need to increase his income, he seemed ideal.
The following year, Edwards had a new job as a preacher with a much larger salary and a house in the Shropshire town of Ludlow. On top of that, he was being paid for computing, on average, six months’ worth of lunar tables a year. It was just as well. Edwards was spending more and more time and money on his telescopes. “He needed the money to buy the metals to make the mirrors for his telescopes,” says Croarken, a visiting fellow at the UK’s University of Warwick. Maskelyne might have wondered how Edwards found the time between preaching, teaching and carrying out experiments. It was simple: his wife Mary did the work. “John’s name was on the payroll, but Mary was clearly doing most of the computing from the start,” says Croarken.
So how did an 18th-century clergyman’s wife come by these mathematical skills? John might have taught her. But given how quickly and easily Mary slipped into the role of computer, Croarken thinks she was good at maths even before she married. Perhaps she had brothers and sat in on their lessons. Perhaps her father was keen on the numerical puzzles and riddles popular at the time, and had maths books around the house.
Mary didn’t have to be brilliant at maths, just capable and careful. Maskelyne had done the complicated work in advance, leaving the computers to do little more than long sequences of additions and subtractions. For every entry in a table, Mary might have to look up 12 figures in astronomical tables and perform 14 operations on the data – and then repeat it for every day of the month. “It was very, very boring work,” says Croarken. But Mary was quick and rarely made mistakes.
In 1784, John’s passion for telescopes proved fatal. Experimenting with a new mix of metals to improve the reflectivity of his mirrors, he inhaled a lungful of arsenic fumes and died. He was 36. Suddenly, Mary and the children faced poverty and homelessness. John had run up debts. His salary and the house went to his successor. To Mary, the obvious way to support her family was to carry on computing. But would the Board of Longitude employ a woman?
Mary wrote to Maskelyne. By now, the Astronomer Royal knew the family quite well. He had visited on trips to see his sister, who lived nearby. Maskelyne must have known that for the past decade it had been Mary who had been doing the bulk of the computing. And she was good – perhaps the best of his computers. The work kept on coming.
Struggling to make ends meet, Mary took on more and more work for the almanac, until eventually she was computing a whole year’s worth of tables. And while other computers took several months to deliver a two-month chunk of tables, Mary could turn it around in three or four weeks.
Mary’s reputation for reliability and accuracy was her greatest asset. When the computers got too far ahead – with 10 years’ worth of tables in hand – the Board of Longitude stopped the work. Mary asked for compensation for lost income – and got it. When work started again, Mary was put back on the payroll. In 1811, when Maskelyne died and a new Astronomer Royal was appointed, the steady stream of work slowed to a trickle. Again, she petitioned the board – and again they stepped in to protect her livelihood.
In her researches among the Royal Greenwich Observatory archives, Croarken has uncovered more than a woman’s struggle to support her family. She has also found a Nevil Maskelyne at odds with the character of popular mythology. Maskelyne was a key player in the story of longitude. While humble carpenter John Harrison is portrayed as the hero, famously labouring to create a chronometer that would keep accurate time at sea, Maskelyne has been vilified as an arrogant, elitist astronomer determined to promote the lunar tables at Harrison’s expense. But both the chronometer and the tables solved the longitude problem.
“Maskelyne admired Harrison’s chronometers,” says Croarken. “But they were too expensive to be practical.” Until the price came within everyone’s reach, lunar tables were a better bet – cheap, reliable and practical. It was for the same reason that Maskelyne employed Mary Edwards. He may have been moved by her predicament. “But he kept her on because she was good at it, and got the job done,” says Croarken.