The Man Who Invented the Computer

by Jane Smiley

  Mauchly was just far enough behind the economic downturn when he got his PhD that he had a difficult time finding a job (Atanasoff, who found his job at Iowa State in 1930, had to take a pay cut along with all the other faculty members as the Great Depression worsened in the early thirties). Mauchly spent one year as a research assistant at Johns Hopkins and then was hired by Ursinus College, a four-year liberal arts school outside of Philadelphia founded by the German Reformed Church. At the time Mauchly taught there (and his teaching responsibilities were heavy), the college was associated with the United Church of Christ. Mauchly’s students were, for the most part, not planning to be engineers or physicists. Mauchly was the only member of the physics department—his job was to give his premed students their required course in physics and to bring returning high school teachers up to date. His salary was comparable to Atanasoff’s—$2,150. He also had a wife and two children—he had married a mathematician, Mary Walzl, in 1930.

  Like Atanasoff, Mauchly engaged his students in his research interests—he put them to work correlating rainfall with the rotation of the sun, but the calculations were stupefyingly tedious and time-consuming. In 1940, while maintaining a full teaching load, Mauchly had constructed the Harmonic Analyzer.

  After Mauchly’s lecture, Atanasoff hurried to the front of the room. The two scientists, both talkative and enthusiastic by nature, hit it off, and they discussed their projects for about half an hour, Mauchly describing his Harmonic Analyzer and Atanasoff describing the ABC. Mauchly showed so much enthusiasm that Atanasoff invited him to visit Ames and have a look at the machine. But Atanasoff was cautious about disclosing technical details because back in Ames, Iowa State was already beginning the patenting process, and Atanasoff was well aware that he could run into trouble if he divulged too much. Patenting was on his mind—the Atanasoffs were to meet Clifford Berry in Washington, D.C., a day or so later. After chatting with Mauchly, Atanasoff and his assistant spent four days looking through patent documents, reassuring themselves that their ideas were new and had never been patented before.

  Atanasoff felt that the ABC was such a success that a significantly larger investment on the part of Iowa State was warranted and would pay off in the future. After the Christmas break, he met with college officials in order to persuade them to hire Richard Trexler, a patent attorney from Chicago with an excellent reputation. Iowa State officials balked, but Atanasoff talked them into it, and Atanasoff sent Trexler the third and last copy of his thirty-five-page description of the machine. Trexler seemed to think there would be no trouble patenting it, but Iowa State officials still did not understand the possibilities. It was only after Atanasoff received a $5,330 grant from the Research Corporation of New York at the end of March that Iowa State began to realize that the machine in the basement of the physics building might have a worthwhile, and lucrative, purpose. College president Charles E. Friley was impressed by the size of the grant—about double Atanasoff’s yearly salary and equivalent to almost $85,000 in 2010 dollars (although it was still only about 1 percent of what IBM would ultimately spend on the Aiken Mark I).

  For the next six months, Friley and Atanasoff negotiated the terms of an agreement to divide up profits that might accrue to Atanasoff’s ideas. The college was stingy, to say the least—Friley wanted 90 percent of any income and, following college policy, did not want to give Clifford Berry any portion of the profits. The penalty for Atanasoff of refusing to sign would be that the college would withhold the Research Corporation of New York grant. Atanasoff, never one to be bullied, persisted until the college agreed to give him half of the profits, after expenses. From Atanasoff’s portion, Berry would receive 10 percent. The parties signed a final contract in July 1941.

  There were other naysayers—Howard Aiken, having begun to work with IBM, was committed to the Mark I and not enthusiastic about the ABC. Warren Weaver, from the University of Wisconsin, who visited concerning Atanasoff’s war-related project, also reaffirmed his belief that analog was the way to go. Samuel Caldwell, of MIT, also visited the defense project, and though he was impressed by Atanasoff’s machine, he was himself working on Vannevar Bush’s Differential Analyzer (soon to be called the Bush-Caldwell Analyzer).

  In the meantime, Mauchly had access to various projects that were progressing around Philadelphia and Washington, D.C. According to Scott McCartney, Mauchly took his students on one or more field trips to nearby Swarthmore College and was shown a vacuum-tube system for counting cosmic rays. What impressed him was the speed with which the vacuum tubes reacted—he observed that a vacuum tube could distinguish between two inputs only a millionth of a second apart. He saw that vacuum tubes could be much faster than switches or keys in counting, but he still had the model of a desk calculator in mind. The vacuum tubes were for input speed; he had not conceived of a binary system, in which the states of “on” and “off” would produce a logic system.

  In 1939, Mauchly had seen an IBM encryption machine at the New York World’s Fair that “used vacuum tube circuits for coded messages.” Like Atanasoff, Flowers, and Schreyer, he began tinkering with various bits and pieces; he ordered tiny neon bulbs from General Electric as an experimental substitute for vacuum tubes—they were cheaper to buy and cheaper to run. But there is no evidence that he had a larger system in mind, and his only product was the Harmonic Analyzer, which may be thought of as similar to Atanasoff and Brandt’s Laplaciometer. By the time Mauchly drove to Ames, his ideas do not seem to have jelled into a systematic theory about how an electronic calculator would have worked. Although Mauchly’s biographer declares that he was a “ferocious record-keeper,” he offers no citations of records that Mauchly kept during this period, or for any date before 1943.

  Faced with repeated evidence that Iowa State did not understand or particularly value himself, his assistant, or his invention, and that other well-known inventors were committed to their own ideas, Atanasoff reacted with pleasure to Mauchly’s enthusiasm. Mauchly first planned to make the eleven-hundred-mile trip west from Philadelphia in the spring but then put it off until summer. In a letter dated May 31, Atanasoff welcomed his imminent visit and suggested that he drop a line giving a date. Mauchly did drop that line, saying he would arrive either the evening of the thirteenth of June or the evening of the fourteenth, but Atanasoff failed to communicate this information to Lura, so when Mauchly arrived on the evening of the thirteenth, just as Lura was cleaning up after dinner, Lura was both surprised and put out—she had expected to get ready for the visit the next day. And to top it off, Mauchly had his six-year-old son with him. He rather impolitely asked for food and then unceremoniously handed the child over to Lura to take care of for the next four days. Lura was put on alert, and she did not like what she saw.

  Atanasoff did like what he saw, though, because Mauchly was eager for information and seemed receptive to Atanasoff’s ideas—just what a man who was enthusiastic about his project and underappreciated would be looking for. In the four days that Mauchly and his son were in Ames, Mauchly, Atanasoff, and a revolving set of onlookers spent hours in the basement of the physics building. The rest of the time, especially at home, Atanasoff and Mauchly talked incessantly about the machine—how it worked, what the principles behind it were, what Atanasoff’s system consisted of. Mauchly seemed impressed—he carried the green-covered thirty-five-page description around with him and asked to borrow it and take it back to Pennsylvania. Atanasoff would not allow this, but he did allow Mauchly full access to it while he was in Ames and also allowed him to investigate the computer carefully with Clifford Berry. Sam Legvold, who was working in the next room on Atanasoff’s defense project, later remembered that Mauchly had hands-on access to the ABC and even helped Berry do a few repairs—Legvold saw him touching parts and carrying parts around.

  At one point, Mauchly also asked Lura for a stack of bond paper. Lura may originally have been offended by Mauchly’s insensitivity as a houseguest, but she became alarmed by some of his other activi
ties. Lura was a busy seamstress who often stayed up late sewing. She noticed that Mauchly was up late, too, because the light in his room was on, and she suspected that he was writing. She feared that he was not only taking an interest in the ABC but planning and working to steal her husband’s ideas. While Mauchly was in Ames, she warned Atanasoff not to talk as freely and in such detail as she witnessed him doing, and Atanasoff acknowledged her caution. But he did allow Mauchly free access to the computer, and he did answer his questions. He later said that his impression at the time was that Mauchly had neither the background nor the knowledge that would make him capable of stealing the ideas, or understanding them well enough to be able to reproduce the ABC back in Philadelphia. For Atanasoff, the temptation to show off his invention was too great, certainly in part because even while the president of Iowa State was attempting to secure the future profits of the ABC, Atanasoff’s colleagues there were almost universally either skeptical about or indifferent to what he was doing.

  Subsequent controversy about whether Mauchly or Atanasoff should be given credit for the invention of the computer (or, to be precise, the invention of the calculating device that led to the invention of the computer) has revolved around the question of whether the ABC was operational at the time Mauchly visited Ames. Those who give the credit to Mauchly say that it was not. Those who give the credit to Atanasoff say that it was. But one thing that John Gustafson and his fellow reconstructors discovered when they built the replica and delved into the history of the ABC was that Atanasoff’s friend in the statistics department, Professor George W. Snedecor, “would send problems over to Atanasoff and the ABC would solve them. Then the secretary, Clara Smith, would check the results on a desktop calculator. And they would be correct.” The ABC was functional.

  In the summer of 1941, Mauchly entered a course at the Moore School of Electrical Engineering at the University of Pennsylvania, a much more prestigious and well-connected institution than Ursinus College. The course was given at the behest of the Department of War and was designed as a cram course in electronics for young scientists in other fields. Mauchly, thirty-six, was hoping that he would learn some things that would move his weather project forward. It was there that he met his partner-to-be, J. Presper Eckert, age twenty-two, just graduated from the Moore School in engineering. Like Atanasoff, Berry, and Mauchly, Eckert had a long history of high-energy fiddling, but unlike the others, he was a child of privilege. His father was a Philadelphia developer who hobnobbed with such celebrities as Ty Cobb and Douglas Fairbanks, Jr. Young Eckert went to school—the William Penn Charter School—in a chauffeur-driven limousine. Father and son were both well traveled—Pres, as he was known, had already visited the Pyramids, among other exotic locales. At an amusement park in Paris, he got the idea for a project that won the Philadelphia Science Fair when he was twelve—a four-by-six-foot pond-like tub with magnets resting in the bottom. He steered a model sailboat across the surface of the water using a steering wheel connected to the magnets. He built radios and music systems and installed them around Philadelphia, including a system for a cemetery that, according to Scott McCartney, “masked the unnerving sound of gas burners in the nearby crematorium.” His connections around Philadelphia gave him access to such innovative communications companies as Philco, RCA Victor, and others. He was a member of the Engineer’s Club of Philadelphia and spent time with Philo T. Farnsworth, an inventor of the television who settled in Philadelphia in 1931.

  In 1937, according to Scott McCartney in ENIAC, Pres Eckert scored second in the nation on his math SAT and was accepted to MIT, but his mother and father prevailed on him to stay home and go to the Wharton School. Within a few months, he transferred from Wharton to the Moore School, to study engineering. But he was not a good student—he did only what he wanted to do and occasionally fell asleep in class. According to McCartney, upon being awakened by the dean of the Moore School and asked, “If you’re going to come to class, why can’t you stay awake?” Eckert responded, “Why?” Every day, he wore a clean, pressed, monogrammed white linen shirt to class. After he graduated in the spring of 1941, Eckert joined the same ten-week cram course as Mauchly, and the two were assigned to be lab partners. Mauchly was the oldest student in the class, and one of two PhDs. Eckert was the youngest. At the end of the summer, Mauchly was hired away from Ursinus by the Moore School to teach physics, a replacement for other faculty who were leaving to join the war effort. According to McCartney, Mauchly’s hiring was not a sign that the University of Pennsylvania was impressed by him or considered him promising, only that he was the only available candidate.

  The other PhD in the course was Arthur W. Burks, originally from Duluth, Minnesota, whose PhD, from the University of Michigan, was in philosophy (though his BA was in physics and mathematics). He had completed his dissertation, “The Logical Foundations of the Philosophy of Charles Sanders Peirce,” on a brilliant but troubled and even tragic contemporary of William James whose work is much better appreciated today (in part thanks to Burks) than it was during his own lifetime. In the summer of 1941, Burks was twenty-five. He was hired to teach at the Moore School that fall, like Mauchly. He eventually joined the ENIAC team (ENIAC stood for “Electronic Numerical Integrator and Computer”—Mauchly added “and Computer” after visiting Ames), and, like Mauchly, found a wife, Alice, among the women mathematicians who were computing firing tables. Alice had gotten her BA from Penn in 1944 in mathematics.

  The teaching load at the University of Pennsylvania was lighter than that at Ursinus and left Mauchly time that he planned to use improving his Harmonic Analyzer. In October, he wrote Atanasoff, specifically asking, “Is there any objection, from your point of view, to my building some sort of computer which incorporates some of the features of your machine? For the time being, of course, I shall be lucky to find time and material to do more than make exploratory tests of some of my ideas, with the hope of getting something very speedy, not too costly, etc.” Mauchly was also looking toward the future—he asked in the same letter whether “in the event that your present design were to hold the field against all challengers, and I got the Moore School interested in having something of the sort, would the way be open for us to build an ‘Atanasoff Calculator’ … here?” And he reported that Irven Travis, the man who had designed an analog “analyzer” on the model of the Bush-Caldwell Analyzer at the Moore School had entered the navy and departed. Mauchly was quite familiar with Travis’s machine and had discussed it in depth with Travis. Travis later reported that he had discussed his variation on the Bush-Caldwell Analyzer with Pres Eckert when Eckert was his student. Before leaving for the navy, Travis had already considered the idea of building a computer on the scale of Aiken’s at Harvard—he had done a study for General Electric that estimated the cost at about half a million dollars. GE did not want to spend that kind of money, but Travis did give Mauchly a bibliography of material about it. Atanasoff responded cautiously, more cautiously than he had acted in June. He wrote, “Our attorney has emphasized the need of being careful about the dissemination of information about our device until a patent application is filed. This should not require too long, and of course I have no qualms about having informed you about our device, but it does require that we refrain from making public any details for the time being.” He went on to say that with these considerations in mind, he had refused an invitation to describe the machine at the meeting of the American Statistical Association.

  By the summer and fall of 1941, Turing’s work on the Bombe and the Enigma code (which the British referred to as “Ultra”) had profoundly impressed his colleagues at Bletchley Park, and he had also impressed Winston Churchill. The code breakers had been successful: so many German supply ships were sunk in the late spring that the British authorities worried that they had handed the Germans irrefutable evidence that the cipher was broken. As Konrad Zuse had seen, though, the Germans simply decided that such a thing was impossible and continued using Enigma. After May, the work at
Bletchley Park met with a few small obstacles, but by the autumn of 1941, the British were confident that they could decode any German naval communication, and if the British navy used their knowledge wisely, they could severely limit the vulnerability of British forces to German naval operations.

  There was, however, another more complex encoding system that the Germans were working with, which the English decoders at Bletchley Park called “Tunny.” When Alan Turing grew famous in the 1980s, almost all of the information concerning the importance of Tunny and its solution at Bletchley Park was still secret. These secrets were finally revealed in 2006 with the publication of Colossus by B. Jack Copeland and colleagues. While Enigma was used by the German navy, Tunny was used by the German High Command, including Adolf Hitler. After June 1941, Tunny was produced by a more complex encoding machine, the Lorenz Schlüsselzusatz (“Extra Keys”). The security surrounding the breaking of the Tunny codes at Bletchley Park would shape computer history but would remain top secret until the 1990s, long after the death of Alan Turing and long after most historians and students had come to what turns out to be a misunderstanding of the progress of World War II.