SOON AFTER THE WAR, FERMI’S PERSONAL FINANCES CLASHED with national security policy. The story is a fascinating one, and though it ended reasonably well, it could have soured Fermi’s relationship with the US government. It is perhaps the best example of how the demands of the new national security state came into conflict with traditional ways of doing things. That Fermi, who assiduously avoided any such conflicts throughout his lifetime, found himself in this situation is one of his life’s greater ironies. It can only be appreciated in the context of Fermi’s overall attitude toward financial matters.
FERMI’S INTENSE CONCENTRATION ON PHYSICS NEVER DISTRACTED him from the state of his bank account. When he was in the process of deciding to move to the United States, he took a careful inventory of his and Laura’s assets and noted them in the back of a ledger he later used for physics. Of the two, Laura—he referred to her as “Lalla,” the affectionate nickname her closest friends used—was by far the wealthier. She had some 120,000 lire worth of bonds and about 1,000 shares of stock in various companies. These holdings alone would have made her a very comfortable, upper-middle-class woman. She also held title to the apartment at Via Belluno and one-quarter of the villa in Tuscany they visited every summer before they moved to the United States. Enrico owned the new apartment at Via Magalotti and the garage in the building next to their home at Via Belluno. He also seems to have owed 10,000 lire to Laura’s father, Augusto Capon. When they left Rome in 1938, the family entrusted all their Italian assets to Enrico’s sister Maria, who carefully protected them and sold them off after the war.
When he came to the United States, he meticulously recorded his monthly paychecks in his pocket diaries, which increased from an annual rate of $8,000 per year to almost double that by the time he passed away in 1954. At $15,000 per year, this made him one of the highest paid professors at Chicago. He also noted carefully all out-of-pocket expenses for work-related travel, to Los Alamos, to Hanford, to Oak Ridge, and to various universities and summer schools. He noted consulting fees from his association with DuPont and other US companies eager to have his advice on technical matters. He also recorded in detail the various stock and bond transactions he entered into throughout the postwar years. In another life he would have made a world-class accountant.
Giulio recalls his father as somewhat stingy and gives as an example Enrico’s preference in the winter to keep the heat in the Chicago houses set at a chilly sixty degrees. Whether Enrico was genuinely miserly is debatable. For example, he lent Schrödinger money before the war and graciously wrote off the debt when Schrödinger tried to repay. However, he was certainly no spendthrift. He was always looking for ways to spend less money while, at the same time, always looking for ways to increase his income. Several of those postwar efforts to increase his revenues are noteworthy. One was an ill-fated attempt to replicate in the United States the financial success he achieved with the publication of his Italian physics textbook for high school students. Another was an eight-year effort to get financial compensation from the US government for the use of the slow-neutron patent granted to him and his fellow inventors in July 1940.
IN NOVEMBER 1946 FERMI APPROACHED MACMILLAN PUBLISHERS with a proposal for a translation of his Italian textbook Fisica, designed for liceo students. For the head of Macmillan’s education division, R. L. Knowlton, the idea of a high school textbook authored by the famous Manhattan Project veteran must have seemed too good to be true. Macmillan sent a senior editor, Martin Robertson, on December 9, 1946, and by December 16 the two men were outlining terms for the project. Laura was responsible for the translation into English. She was deeply involved in the Italian edition, which took shape at the Tuscan villa owned by her uncle during summers and holidays in the late 1920s. By March 1947, a “competent” high school teacher, Warren Davis, principal and physics teacher at Alliance High School in Alliance, Ohio, was selected to join the project. The choice of Davis is a bit of a mystery, because he had not yet published anything at this point and indeed never published anything during his career. Fermi had certainly never heard of him. However, after meeting him, Fermi agreed to the arrangement and on March 24, 1947, made a note of the division of royalties from the project. Royalties on the book would be split 60/40, with Fermi getting the larger percentage. Royalties for a laboratory workbook written by Davis would be split 60/40 the other way. Macmillan followed up with a detailed letter outlining the terms of the arrangement, including a review by Davis of all existing high school texts and a proposed completion date of November 15, 1948. This would allow for publication in early 1950.
From the start, the experience was not a happy one. By early October 1947 Laura was translating the second volume, but Fermi had heard nothing from Davis. A letter he sent to Davis at that time must have prompted a package of material, because the next letter in the file is from Fermi to Davis in late January 1948, in which Fermi—clearly unhappy with what he had read—suggests that he mark up the sections on fluid dynamics with a series of editing “codes” for Davis to consider. These codes give a flavor of the difficulties Fermi encountered in Davis’s manuscript. He would mark “1” for grammatical errors, “2” for incorrect physics, “3” for material that was too obscure, “4” for an unnecessary deviation from Laura’s translation, “5” for material that should be deleted, “6” for material that required additional illustrations, “7” to indicate where the use of letters would simplify the text, and “8” for questions in the text that should be changed or deleted. Fermi was not impressed with what he had read, yet he still had the patience to see if the collaboration would work.
In early August 1948, Fermi wrote another letter to Davis. The high school teacher revised the opening chapters, but they still were unsatisfactory to Fermi, who marked them up for further revision. He also felt that Davis had added enormously to the word count, perhaps doubling Laura’s translation. By the end of August 1948, Fermi took the time to read the drafts more carefully and his verdict was even more negative: “Unfortunately, I find that the manuscript is still in a very unsatisfactory shape, and I am afraid that it will be so even after the corrections I am suggesting are incorporated into it. I am afraid that the trouble is much more serious than I realized.”
Realizing the seriousness of the problem on their hands, the editors at Macmillan decided to step in and review the manuscript that Davis had presented to Fermi. They made some minor changes to one chapter, which Fermi seems to have approved, although he must have cringed at the editors’ decision to replace the term velocity—which has a very precise meaning in physics—with the term speed, which has a precise, but different, meaning. By this time, the process had become quite confusing, with Davis, aware of the looming deadline, sending a spate of draft chapters to Fermi and to Macmillan for review and editing. It took Fermi a few months to review and revise chapter four, and he submitted it to Macmillan in mid-March 1949, now some five months past the deadline. Macmillan agreed it was a major improvement over Davis’s draft.
The next letter in the file is a draft of correspondence from Fermi to Davis, which must have been sent some time later. It is a disavowal of the entire project. Fermi criticizes Davis for not following the original book more closely. He observes that the language in places seems more suited for “younger children.” The draft of the letter ends with Fermi’s concern that he no longer could afford to continue the collaboration, but this sentence was crossed out. Because the actual letter he sent does not exist in the file, it is difficult to know what the final version said.
Some four years after the original deadline, in November 1952, Macmillan wrote a letter to Fermi acknowledging his decision to end the project. The long-suffering editor, Martin Robinson, put it this way:
I am sincerely sorry that the plan did not work out to everyone’s satisfaction, and personally regret any misunderstandings that may have developed. Needless to say, I still hold you and your work in the highest esteem and hope that if you find the time to prepare a college text or a reference work, that you will give us an opportunity to show you that we can really make good books without so much struggling.
The project’s disintegration must have been agonizing for everyone involved and its end, some relief, albeit tinged with regret, for Macmillan.
For a host of reasons it is a great pity that the project failed. Obviously, it would have made a significant amount of money for all parties concerned—Macmillan, Davis, and the Fermis. Sticking to the translation provided by Laura might well have proved the best course for everyone. Back in Rome, Amaldi took over the project of updating and revising the Italian textbook, which continues to be a best-seller. When Edoardo passed away, the publishing project passed to his son, Ugo, named after Edoardo’s father, who continues to update it. It has sold more than three million copies to date and is still the basic textbook for high school physics in Italy.
Fermi was interested in the money, but he was more interested in getting the textbook right. He had very high standards and could not bear the thought of a slipshod textbook being published in his name, irrespective of the financial opportunity involved. Davis seems to have been out of his depth and certainly not on the same wavelength as his coauthor. High school students to this day are the poorer for Davis’s failure.
FERMI’S 1940 POCKET DIARY CONTAINS A SEVEN-DIGIT ENTRY FOR July 3—“2,206,634.” It represented a milestone for Fermi. The day before, the US Patent Office issued Fermi and his Rome colleagues a patent for the slow-neutron technique. Gabriello Giannini had filed the patent on their behalf on October 3, 1935, almost a year after the patent was filed in Italy. In the interim, the application had worked its way through the bureaucracy, emerging on July 2, 1940, as a fully registered US patent. Fermi learned of his good fortune the next day and scribbled down the number of the patent for his records.
Thus began a thirteen-year saga, which eventually embroiled the five named inventors—Fermi, Amaldi, Pontecorvo, Rasetti, and Segrè, as well as D’Agostino and Trabacchi, both of whom were promised shares in any royalties—in extensive and complex litigation against the US government.
IT MAY HAVE BEEN CORBINO’S IDEA TO APPLY FOR A PATENT ON THE slow-neutron idea, but Fermi and his colleagues eagerly adopted it as their own. Corbino saw the commercial potential for a technique that could be used to develop new radioactive isotopes for medical use. He could hardly have anticipated that slowing down neutrons would enable physicists to split the uranium atom, much less create a chain reaction. Though the fission weapon did not rely on slow neutrons, slow neutrons created the world’s first controlled chain reaction and were an essential part of the vast reactor facility at Hanford that churned out plutonium for America’s nuclear arsenal. The slow-neutron technique had enormous potential civilian power generation applications as well. To understate the case, the slow-neutron patent was valuable intellectual property.
As the Manhattan Project developed, Fermi, Segrè, and Pontecorvo saw the growing value of their patent, but seeking compensation for its use by the US government was problematic. First, Fermi and Segrè were overwhelmed with work on the bomb and had little time for the administrative or legal effort required to pursue the matter. Second, the true value of the patent was apparent only to those who were cleared to know what was happening within the Manhattan Project. Giannini, even though a US citizen, was not cleared, and of course Amaldi, D’Agostino, and Trabacchi were citizens and residents of Italy, at that point an enemy nation. A third complicating factor was the policy developed haphazardly during the project that all intellectual property arising from it would be signed over to the US government. Fermi and his Manhattan Project colleagues filed applications for at least twelve patents between 1944 and the end of the war related to their work, the most famous being patent 2,708,656, filed jointly by Fermi and Szilard, for the “neutronic reactor.” All of these were effectively the property of the US government and remained classified until the 1950s. It would be another ten years, in the 1960s, before the patents were granted. The slow-neutron patent was filed before the United States entered the war, but pursuing compensation while the war was on would, in the inventors’ view, run afoul of the way the US government was handling intellectual property related to nuclear weapons. A fourth issue was the uncomfortable fact that Amaldi, D’Agostino, and Trabacchi were all Italian citizens living in Rome and thus enemies of the US government.
Nevertheless, Fermi and Segrè made several ineffective attempts to discuss compensation for the slow-neutron patent with the relevant Los Alamos lawyers, Navy captain Robert Lavender and his assistant, Ralph Carlisle Smith. After the war, and as the most business-minded of the group, Segrè was chosen to liaise with Giannini and Lawrence Bernard, the Washington, DC, lawyer Giannini chose to represent the team, but the end of the war only further complicated matters.
The Atomic Energy Act of 1946 rendered the entire nuclear industry a monopoly of the US government. Previously, Giannini and Bernard negotiated with Captain Lavender in Vannevar Bush’s Office of Scientific Research and Development (OSRD). OSRD proposed a variety of compensation schemes, initially offering a lump sum settlement of $900,000. However, with the 1946 act’s creation of the AEC, the jurisdiction of the dispute shifted to the commission’s office responsible for patent work.
The change of jurisdiction introduced a myriad of new delays. The AEC would not engage in serious negotiations before establishing its own set of policy guidelines, which took several years to complete. The slow-neutron patent was but one of many the AEC had to consider. When Bernard finally filed for compensation in October 1948, Giannini told him to suggest $1.9 million: $1 million as a lump sum payment and nine annual payments of $100,000. Bernard did as he was told but also suggested that the inventors would settle for the original $900,000 offered by Bush in 1946. Bennett Boskey, a lawyer in the Office of the General Counsel of the AEC, was assigned to examine the claim.
An Ivy League lawyer, Boskey vigorously defended the AEC’s monopoly on nuclear science and wrote his report accordingly. It was, as scholar Simon Turchetti puts it, “not good news” for the inventors. Boskey questioned whether the slow-neutron method was essential for the production of fissionable material. Undaunted by the facts and not a scientist, he rejected the claim that the basic R&D for the Manhattan Project relied on slow neutrons. He also painted a demonstrably false picture suggesting that the work of Rutherford and Chadwick held priority over the work of Fermi’s team in 1934. He also claimed the production of radioactive isotopes did not rely on the slow-neutron method. In short, he systematically rejected every one of the patent’s claims, an astonishing performance given that a trained US patent examiner had actually granted the Italians a patent almost two decades previously.
To muddy the waters even further, he darkly noted the 1947 treaty ending hostilities between the United States and Italy entitled the United States to ignore any property or commercial claims made by Italian nationals like Amaldi, D’Agostino, and Trabacchi. To top it off, he argued Fermi’s membership in the GAC made him an employee of the US government and therefore it was a crime for Fermi to litigate against the US government. Irritated, Fermi contemplated resigning from the GAC to clear up the conflict but was able to persuade the AEC that his impending departure from the GAC in January 1951 would solve the problem.
Boskey was smart enough to realize that this first, highly aggressive salvo would shake the confidence of the inventors and he was right. However, their confidence took a greater hit when one of the inventors, Bruno Pontecorvo, took the occasion of a summer holiday in Italy to disappear with his wife and reappear in Moscow. Intelligence services quickly discovered his defection, which was publicly reported in October 1950 by Reuters.
THE PROCESS OF LITIGATION AGAINST THE AEC STIMULATED THE FBI to review the backgrounds of all the inventors. Their attention came to rest on the left-leaning Pontecorvo, all of whose siblings were members of the Communist Party in Italy. Bruno was known to be sympathetic to Communist causes, as well. Reports on the FBI research were passed along to British authorities at MI5 and MI6. British security officials determined that although there was nothing definitive to cause concern it would be wise to place Pontecorvo in a less sensitive position. The unmasking of Klaus Fuchs in 1950 as a Soviet agent landed Pontecorvo under even greater scrutiny. The exact circumstances that caused him to defect when he did are not known, but it is possible that he was tipped off by the British liaison to US intelligence in Washington, Kim Philby, who had been working as a double agent for the Soviet Union for decades.*
The news of his defection came as a shock, especially to the other litigants, who now felt that their position had been terribly compromised. Segrè, perhaps the closest to Pontecorvo, decided it would be best for everyone if he resigned as the liaison with the legal team, and Rasetti, who had found a position at Johns Hopkins in Baltimore, took his place. Negotiations dragged on until, eventually, the sides agreed on a disappointing $300,000 lump sum payment, far less than they expected and far less than they felt they deserved. After paying out the approximately $180,000 of legal fees, each litigant was left with a little less than $30,000. Pontecorvo obviously never got his share.
At that time, $30,000 was a significant sum, enough money to buy a grand house in the Hyde Park section of Chicago. The inventors had, however, been expecting millions. For Fermi, the whole misadventure was an embarrassment. He indicated privately that he would probably never have pursued litigation at all were it not for the loyalty he felt for the other members of the team. The litigation put him in an extremely awkward position with the AEC, which was paying him a consulting fee as a GAC member and as a Los Alamos consultant.
Nevertheless the outcome seems unjust. The patent was granted prior to US entry into the war, well before the Manhattan Project was organized. The enormous power wielded by the AEC during the immediate postwar years was sufficient to determine the outcome irrespective of the merits of Fermi’s position. The odds were stacked against the Italians and at some deep level they must have known it. Along with Szilard’s chain reaction patent and the Fermi-Szilard patent on the nuclear reactor, the slow-neutron patent remains one of the most important patents in the history of atomic energy. The commercial value of these three patents is quite literally incalculable. The Italians settled not because they wanted to but because, given the circumstances, they had to.
When Fermi received his payment, he made sure to invest immediately in securities, under the heading “The Patent Fund.” As with all of his investments, he tended to the Patent Fund carefully.
* It remains unclear whether Pontecorvo was an active Soviet agent. Frank Close studied the issue carefully and discovered that Philby knew of the FBI’s interest in Pontecorvo. Within weeks, the physicist defected. See Close, Half Life. Others, particularly in Italy, believe he was not working for the Soviets.