The Man Who Invented the Computer

by Jane Smiley

  Kirkpatrick asked Mauchly questions designed to get him to elaborate on his replies, which he did, without challenging him or giving away the fact that the Honeywell and CDC lawyers had plenty of documentation that contradicted Mauchly’s testimony almost completely. In the meantime, Charles Call took a deposition from Lura Atanasoff in Boulder, where she was now living. Even though the lawyers had been somewhat nervous about how her divorce would influence how she would report events in Ames, she was clear, concise, and in complete agreement with Atanasoff’s version of Mauchly’s visit.

  The lawyers for Honeywell and CDC were exceptionally thorough, and they had plenty of information to work with. The lawyers for Sperry Rand were less fortunate. One day in mid-November 1967, Atanasoff answered the phone. John Mauchly was on the other end of the line. He said that he would like to see Atanasoff and proposed that he come to Atanasoff’s Maryland farm with one of the Sperry lawyers, to discuss the case. Atanasoff was suspicious enough by this time to ask his wife, Alice, to listen in on the extension and take notes. Atanasoff did not immediately agree to the meeting—he called the Honeywell/CDC lawyers and reported Mauchly’s proposal. Once Atanasoff indicated that he would not reveal what he knew about the case, or what the Honeywell/CDC strategy was, it was decided that a meeting might be informative. It was.

  When he called again to set up an appointment for bringing the Sperry lawyer, a man named Lawrence B. Dodds, to Maryland, Mauchly and Atanasoff chatted rather cordially. Mauchly explained to Atanasoff that Dodds was representing Sperry Rand in a case against Control Data and Honeywell, thus revealing to Atanasoff that neither he nor Dodds knew of Atanasoff’s central position in the case Honeywell and Control Data were preparing. Mauchly’s attitude indicated that he had no idea that Atanasoff might be his antagonist in the patent dispute. His immediate reason for calling Atanasoff was that he had been subpoenaed in the patent dispute and had discovered old letters to Atanasoff that he had forgotten in the course of twenty-six years. He had also given a deposition. By this time, Atanasoff had read Mauchly’s deposition, but he didn’t reveal this, just suggested that he would try to get hold of it. In his deposition, Mauchly told Atanasoff, he had said a few things that might make Atanasoff “mad,” for example “that when you got into administrative work you lost interest in computers.” Atanasoff said, “Maybe I did seem to.” Mauchly was surprised that Jean Berry appeared on the list of witnesses, but not Clifford Berry. Then Mauchly speculated that Berry had died recently, since he had seen Berry’s letter to R. K. Richards describing the ABC and stating that only Mauchly had seen the ABC “in full.” Hadn’t Caldwell seen it? suggested Mauchly, referring to Samuel Caldwell of MIT, who had been asked to make an evaluation of the ABC for grant purposes. Atanasoff told him that no, Caldwell had never had the same detailed access that Mauchly had had. Mauchly then complained about Allen Kirkpatrick, the lawyer who had deposed him, who, he thought, “had practically accused me of plagiarizing everything I’ve done.”

  Mauchly arrived for his visit on the morning of December 16, 1967. Dodds appeared an hour later. Mauchly’s manner revealed that he still did not understand Atanasoff’s position in the case, and Atanasoff remained reticent. When Dodds arrived, Atanasoff was straightforward about what information he would give the Sperry lawyer—he would speak generally, but not specifically, about his deposition, and his position on Mauchly’s visit would be clear. “Dr. Mauchly came to Ames on approximately June 15, 1941. He spent considerable time with the machine; he understood it fully, and in substantially every detail. If you don’t like it, that is just too bad, because those were the facts.”

  Mauchly observed, “You are taking a very positive posture which I cannot take. Your memory is better than mine.”

  Gradually in the course of their conversation, it seemed to dawn on Dodds and Mauchly that Atanasoff was not as uninformed about the case as they had thought he was. Finally there was a revealing exchange:

  Mauchly: “Do you contend that I read the book?” (meaning the thirty-five-page description of the ABC)

  Atanasoff (after hemming and hawing): “However, the answer is yes, and you also asked me if you could take a copy home with you. I denied the request, and so you did not take the copy away.”

  Dodds: “Will you treat us as well as our opponents?”

  Atanasoff: “I do not see why I should place you and your opponents on the same footing. It is obviously to your advantage to prove that there was no development of a computing machine at Ames, Iowa. Your opponents contend the contrary and my interests must lie in that direction.”

  Dodds then asked if anyone else “now alive” had read the manuscript and Atanasoff pointed out that it had gone to various agencies in hopes of funding. Then Atanasoff remarked that he had read the 827 patent that summer—“The 827 patent almost exactly described my own apparatus and its specifications.” Then Mauchly had to be shown the 827 patent (which Atanasoff had a copy of), since he did not remember which one it was. Dodds and Atanasoff sparred a bit about the language of the 827 patent, Dodds saying that the patent didn’t mention “regenerative memory” and Atanasoff pointing out that what was described—“interaction of logic circuits in the computing elements”—was his idea. Dodds acknowledged that this was so. Mauchly kept quiet.

  Alice served lunch. As they got up to go to the table, Atanasoff remarked that the Honeywell/CDC lawyer had encouraged him to find every document and potential witness and remember every detail. Mauchly replied, “Our lawyers don’t want me to remember anything.”

  Sometime later, Atanasoff could not help exclaiming, “Mr. Dodds, in the face of the facts, how do you expect to win this case?”

  Dodds, irritated, replied, “You don’t know anything about how federal judges are likely to act. They may decide the question upon their own impulse instead of fact, law, or reason.” Mauchly and Dodds, it seems, could not help revealing themselves to Atanasoff. Atanasoff, on the other hand, did not reveal that Mauchly and Dodds’s assumption that there had been no witnesses to Mauchly’s work on the computer was wrong. Throughout the interview, Mauchly retained his strange presumption that he and Atanasoff were on the same side. Once Dodds left, Mauchly even remarked that Sperry was paying him a healthy consulting fee for his work on the case and suggested that Atanasoff might try to get the same sort of arrangement, and then he reiterated what he had said before, that the Sperry lawyers had advised him to remember his Ames trip as vaguely as possible. Throughout the rest of the afternoon (Mauchly was not inclined to depart), Mauchly continued to reveal details of the case, things he had seen, bits of advice he had received, royalties he had gotten for the patents, how he had gotten them, what he had done with the money. He showed a friendly interest in Atanasoff’s own career (and evident prosperity), and Atanasoff was left with the feeling “that Dr. Mauchly was genuinely pleased to find that he had not entirely deprived me of living substance.”

  Atanasoff did not at all share Mauchly’s casual attitude toward the suit—like Jean Berry, he had become convinced that there had been foul play in Clifford Berry’s death, and he even persuaded the Honeywell/ CDC lawyers to send a lawyer along with him back up to New York to look into the case. As usual, Atanasoff devoted himself to finding out everything he could, to thinking it through, and to persuading those in charge to see things his way. He did, in fact, talk to the detective in charge of investigating the case into reopening it—he did not think the levels of alcohol and medicines in Berry’s blood (they were low) and the way that he had died (quietly, his arms at his sides) added up to a realistic case for suffocation by plastic bag. He was convincing enough for the immediate investigator, but not enough for his superior, and the case was not reopened. Atanasoff remained uncertain, at least publicly, about the cause of Berry’s death—in subsequent interviews, it was clear that he could see both sides of the issue. Jean Berry was always certain that her husband had been murdered—he was the person who had the clearest information both about what the ABC was
and how it worked, and how much time Mauchly had spent with the machine, what he had done, and what Berry himself had told him.

  Each of the Honeywell/CDC witnesses had something different to offer: Sam Legvold had seen Mauchly around the computer in the basement of the physics building; Lura Atanasoff had seen him in her home, with the copy of the description of the ABC in his hands, and pens, and bond paper, with his light on late into the night; R. K. Richards had Berry’s clearly stated correspondence on the issues under question. And then, Atanasoff offered to have several technicians in his Maryland machine shop take the thirty-five-page written description of the machine and build a complete demonstration model. Alice Atanasoff went shopping for the exact outmoded parts that they would need (though the proper 1940-vintage vacuum tubes were hard to find). It was agreed that Atanasoff himself would neither oversee the construction nor participate, just to demonstrate that the description was enough of a blueprint. When it was built, in the summer of 1968, it worked beautifully and did everything Atanasoff said it would. Atanasoff himself was so pleased with it that he built another one for himself.

  Data gathering and record gathering continued through 1968, with the Honeywell lawyers and the CDC lawyers seeking out every document and witness. The Sperry lawyers were not as industrious, and neither was Mauchly—when he appeared for discovery in October 1968, he had only a few papers with him, all, he said, that he could come up with. The Honeywell/CDC lawyers had to remind him that he was legally bound to search out everything that he could find. They questioned him for three days, most particularly about three separate issues—what were the precise concepts he had thought up on his own before his December 1940 discussion with Atanasoff at the annual meeting of the American Association for the Advancement of Science, what had he done and learned in his June trip to Ames, and what had he intended to say in his correspondence with Atanasoff after his visit. Mauchly could hardly remember anything, and he remarked over and over that he had a bad memory. The lawyers could not tell whether he actually could not remember anything or whether he was following instructions from Sperry lawyers. He repeated that he had not used any of Atanasoff’s ideas and that the ABC was “an incomplete machine” and “would not do” what it was intended to do. At least that much he remembered. He did not seem to realize that he was contradicting himself.

  Atanasoff’s deposition, which began eleven days after Mauchly was finished, was the exact opposite of Mauchly’s in many ways—he had plenty of evidence that he had thought about and tried out various concepts before and after his revelation of December 1937, including grant proposals. He also had a clear memory of his own thinking, and of events surrounding Mauchly’s visit (a memory that was corroborated by his witnesses). He was so organized that the lawyers could suggest only very small ways to shape his testimony to make it more forceful or more clear—he had an excellent grasp not only of what he had done with the ABC, but also of what he was doing in the case.

  When Mauchly returned for a second session of questioning in April 1968, he brought a large stash of documents that he had managed to uncover. Unfortunately, they had the effect of supporting Atanasoff’s contentions, not his own—whether or not Mauchly now remembered that the ABC worked, he had written enthusiastically to friends in 1941 that the computer could “perform all kinds of mathematical feats.” Charles Call read the documents while another lawyer deposed Mauchly. Call then told the other lawyer what he found in the documents, and, using this information, the other lawyer challenged Mauchly’s testimony. Mauchly thereupon modified his testimony. As his own biographer remarks, “He made a huge mistake by obfuscating the facts of his Iowa visit.” In ENIAC, McCartney tries to make a case for Mauchly having had ideas about a computing device before he went to Ames—according to a colleague, he invented a “little computing device … [that] used neon tubes as trigger circuits. And he’d done some simple arithmetic work on the desk setup, using those triggers.” But this is a defense of Mauchly that Mauchly did not make for himself, possibly because, according to Mollenhoff, his device was a single neon tube mounted on the lid of a Quaker Oats box that turned on and off.

  In some ways, John Mauchly remains the most mysterious and contradictory figure of all of our computer innovators. There is no evidence that he was coldly calculating in any sense of the word. His efforts in every direction seem to have been expansive, impulsive, and inclusive rather than cool and directed. When J. Presper Eckert’s second wife remarked that Mauchly could not have put together ENIAC without Eckert, but that Eckert would not have thought of it without Mauchly, she was portraying Mauchly as a certain type of genius—a disorganized dreamer full of inspiration that comes from nowhere. However, everyone, including those who knew him through Atanasoff, remarked on his sociable nature and his aptitude for conversation, and so the evidence is that his ideas did come from somewhere—from others, if only in embryonic form. This, too, accords with certain theories of creativity (most particularly those delineated by Malcolm Gladwell in Outliers)— that “genius” is a social phenomenon, that ideas grow out of human intercourse, that certain communities produce a wealth of talent because of certain mores of interaction. One such habit, Kati Marton would say, was the way Jews in Budapest of the 1920s loved to linger in cafés, smoking and talking—perhaps the world in which John von Neumann came to believe that some ideas should not be possessed and patented by individuals. Mauchly was a connector extraordinaire—every story about him attests to that; in each description of him, even when he begins by asserting something (for example that he remembers nothing of a particular event) he soon comes round to remembering it much in the way his interlocutor does. As Atanasoff discovered in December 1940, Mauchly was by nature an enthusiast, and in 1941, when his colleagues at Iowa State were skeptical of his computer, the very person Atanasoff needed to support his own confidence in his machine was an enthusiast who seemed at least somewhat knowledgeable.

  Atanasoff seems not at all like Mauchly—he was well organized and well directed above all things. As a “problem finder,” he had a special talent for formulating specific questions that required solution—as at Bikini Atoll, for example—and then using available materials to come up with the best available solution, if not necessarily the ideal one. But like Mauchly, his talents thrived on social interaction—as he taught his students, he learned from them; as he directed their work, he came up with ideas for his own. His quest for the computer grew out of his understanding of a general need belonging to his community of mathematicians, physicists, and engineers. He was stimulated by everything from the slide rule he got from his father as a boy to methods of house construction he employed in retirement. What had already been discovered and invented served Atanasoff as a springboard to other things. At the same time, he was good at progressing from level to level—at learning from Charles Babbage’s unfortunate experience not to try to invent the universal machine before you have gotten the specific one to work. Atanasoff was Mauchly and Eckert rolled into one—he had grand mathematical ideas and he had specific engineering ideas. He understood what both kinds meant and he understood how the two fit together. And then he was blessed with a perfectly congenial partner, Clifford Berry, whose building process was smooth, thoughtful, and efficient. Like Atanasoff, it did not occur to Berry to try a zillion things at once or to drop one project and begin another before the first was completed.

  Sperry had one advantage at the trial, the considerable one of the reluctance of courts to overturn patents already granted. But Honeywell and CDC had one, too—Mauchly was required to prove a negative, to prove that he had not been influenced by the time he had spent in Ames, and to prove that he was neither lying about his memories nor simply failing to remember things that he had thought and done. McCartney maintains that the case “boiled down to one scientist’s words against the other,” but in fact many of the words the Honeywell and CDC lawyers were using against Mauchly were his own. And the Honeywell lawyers went about their case in an Atanaso
ffish way—exhaustively. They put all their documents together in electronic legal files, which gave them excellent organization and ready access. They seemed to realize, as the Sperry lawyers did not, just how useful computers could be.

  The Honeywell and CDC lawyers also understood that they had to supply expert witnesses for the trial who could explain to the judge, Earl R. Larson, what the issues at stake were—how computers worked, how the ideas in the ABC were linked to ENIAC. Larson had an excellent reputation as a scrupulous judge whose opinions were rarely appealed and even more rarely overturned. He intended to preserve this reputation in what he soon came to understand was one of the most important intellectual property cases of the twentieth century. He knew that he had plenty to learn and that he had to take the time to learn it. The Honeywell/CDC lawyers hired Isaac Auerbach Associates (the same Isaac Auerbach who had asked Mauchly at the panel Call witnessed about his visit to Iowa State), and Auerbach supplied him with three computer experts who had worked on both ENIAC and EDVAC. The witnesses had several jobs: corroborating and explaining Atanasoff’s testimony to the judge; informing the judge of the relevant history of the computer; reading and explaining the thirty-five-page report on the ABC; corroborating that the model newly constructed from the old plans was as it was said to be; and ultimately tracing connections from ABC to ENIAC and EDVAC.