The discovery of nuclear fission did not immediately produce an atomic bomb. In December 1938, Otto Hahn and Fritz Strassmann identified an unexpected product of their uranium experiments in Berlin. Lise Meitner, living in exile, and her nephew Otto Frisch supplied the physical interpretation: a heavy atomic nucleus could split into smaller nuclei and release energy. The discovery depended on experimental work, explanation and communication across borders—not on one country possessing every part of the answer. [1] [3]
During 1939, further research made the possibility of a continuing nuclear reaction increasingly important. Leo Szilard and other physicists sought government attention, and Albert Einstein signed a warning to President Franklin D. Roosevelt dated 2 August. Alexander Sachs delivered it on 11 October. Those are different dates: writing the warning, reaching the President and establishing a research organisation were separate steps. Einstein did not subsequently work on the Manhattan Project. [11] [12] [13]
Part 4 followed displaced scholars as they found places to continue working. Part 5 follows what happened when their knowledge and relationships intersected with a discovery of possible military importance. The focus is late 1938 to the end of 1939. We will distinguish an observed scientific process from an unproven application, a feared German capability from evidence of one, and the first American committee from the much larger programme that came later. [10] [15]
Previous chapter: Part 4 — Finding a Place to Work: The Rescue of Displaced Scholars. Next chapter: Part 6 — 1940–1941: Building the Scientific War Effort. Begin the book: Part 1 — Prologue.
A Result That Would Not Fit
The letter at the end of the previous chapter did not announce that someone had built a weapon. It asked for help understanding a result.
On 19 December 1938, Hahn communicated the troubling evidence emerging from his work with Strassmann. Their chemical investigations pointed to barium among the products of experiments involving uranium. Earlier interpretations had looked towards elements heavier than uranium. Barium was a much lighter element. The result did not fit the expected kind of nuclear change. The Max Planck Institute for Chemistry’s account records both the finding and the difficulty of explaining it. [1]
Meitner was no longer working beside Hahn in Berlin. As Part 4 showed, persecution had forced her to leave. Yet the correspondence continued. The political separation of a research community had not completely ended its exchange of knowledge.
That distinction is important to the larger argument. A country can damage an institution without making every experiment inside it impossible. An expelled scientist can lose access to a laboratory without losing the ability to interpret evidence from it. The relationships may survive in altered and unequal forms.
The discovery therefore complicates a simple national story from its beginning. The experimental result came from work in Germany; an essential interpretation involved a scientist whom Nazi policy had driven out. Neither half is sufficient on its own. Removing the German experiment from the account would be inaccurate. Removing Meitner’s contribution would be inaccurate for the same reason.
For the reader, the first question is not yet whether Germany or America was ahead. It is what had actually been established—and what remained to be understood.
An Explanation Across the Border
During the Christmas period in Sweden, Meitner and Frisch considered how the result could be physically possible. Roger Stuewer’s historical research locates their discussion in Kungälv and explains how their different experience of nuclear theory contributed to the interpretation. His account reconstructs the discussion from later recollections; it should not be mistaken for a verbatim record made that morning. [2]
The published argument is firmer evidence of what they concluded. In their joint paper, Meitner and Frisch proposed that the heavy nucleus could divide, rather than merely lose a small particle. They used an existing way of picturing nuclear behaviour—a charged liquid drop—to make sense of the large change and the energy released. Their explanation appeared in Nature on 11 February 1939. [3]
For a non-specialist reader, the essential point is straightforward. Ordinary chemical reactions rearrange the relationships between atoms. Fission changes the nucleus itself. The surprising evidence was not simply that uranium had become chemically associated with something else. It indicated a transformation into much smaller nuclei. Recognising that process explained why the chemical result had seemed so difficult to accommodate. [3]
An explanation is not the same thing as a new observation. It proposes how the observations belong together and what further evidence should be possible. Frisch then obtained experimental evidence for the division of heavy nuclei. His separate report appeared in Nature on 18 February, a week after the joint interpretation. The two publications answer related but different questions. [4]
This is a useful pause in a history often told at great speed. Chemical identification, physical explanation and experimental confirmation were not three interchangeable names for one moment of brilliance. They were contributions that could strengthen one another because people communicated their results and subjected them to further examination.
The distinction also changes what we mean by intelligence in this book. The achievement was not simply the possession of a clever idea. It involved recognising a difficulty, trusting carefully established evidence enough to reconsider an expectation, and connecting forms of expertise that did not all reside in one person.
The Paper Travelled Before It Was Printed
Frisch later recalled drafting the joint paper and reading it to Meitner over a long-distance telephone connection between Copenhagen and Stockholm. She interrupted with comments and suggestions. The American Institute of Physics preserves this recollection. It offers a practical glimpse of collaboration after displacement: even preparing the argument for publication required arrangements across the separation. [5]
The same recollection explains the choice of the word fission. Frisch consulted the biologist William A. Arnold about the term used when a cell divides. A word from one field helped name a process in another. The analogy supplied a name; the physical evidence still had to support the scientific claim. [5]
Meanwhile, Niels Bohr learned of the interpretation before leaving Europe. He reached New York on 16 January 1939. Discussions involving Bohr and his travelling companion Léon Rosenfeld spread the news in American scientific circles; Bohr and Enrico Fermi discussed the discovery at the Washington conference on theoretical physics on 26 January. Stuewer’s reconstruction supplies this sequence. [2]
The dates explain an apparent puzzle. American researchers could be discussing the discovery in January even though the Meitner–Frisch paper appeared in February. Scientific communication was not confined to the date printed at the top of a journal page. Conversation, correspondence and meetings could precede formal publication.
That does not make publication unimportant. It distinguishes the circulation of a claim from the creation of a durable public record. A report passed through colleagues can prompt an experiment; a published argument allows a wider community to inspect what its authors actually asserted. Both matter, but they leave different kinds of evidence for historians.
At Columbia, John Dunning later remembered a discussion on 25 January about the implications of the new results. A paper entitled The Fission of Uranium, published on 1 March, listed Herbert Anderson, Eugene Booth, Dunning, Fermi, G. Norris Glasoe and Francis Slack at Columbia’s Pupin laboratories. The author list is a contemporary reminder that the American response involved a working group, not a solitary recipient of news from Europe. [6] [7]
We can now see another consequence of the scholarly relationships followed in Part 4. They did more than help someone secure employment. They also made it possible for information to reach people who had the equipment, colleagues and experience to investigate it.
The Second Question: Could the Process Continue?
Showing that a nucleus could split did not establish a self-sustaining chain reaction. In recollections preserved by the American Institute of Physics, Dunning and Anderson distinguish the observation of fission from the next question: whether the process released neutrons that could contribute to further reactions. Fermi and Szilard were among those pursuing that issue. [6]
A chain reaction, at the level needed for this history, is a sequence in which one event can help produce further events of the same kind. Demonstrating an isolated event and demonstrating that a sequence can sustain itself are different achievements. Neither, on its own, establishes that an effective military device can be produced.
The distinction matters because a later result can make an earlier question appear trivial. From the perspective of 1939, the additional evidence still had to be obtained. It was not enough to know that energy had been released in an experiment and assume that every proposed application would follow.
The investigation was international. On 18 March 1939, Nature published Hans von Halban, Frédéric Joliot and Lew Kowarski’s report, Liberation of Neutrons in the Nuclear Explosion of Uranium. Here, “explosion” referred to the nuclear event under investigation, not to a completed atomic weapon. Reading the historical title without that distinction would give the reader the wrong stage of development. [8]
On 1 August, Physical Review published Neutron Production and Absorption in Uranium by Anderson, Fermi and Szilard at Columbia. Even its title identifies the research problem as an investigation of what was produced and absorbed, not an announcement that a wartime weapon was ready. The journal record supplies the authorship and publication date; it is not evidence that a national production programme already existed. [9]
This is the developing situation into which the warning to Roosevelt would enter. The concern was no longer detached speculation about an unknown source of energy. There were experiments and published findings. But an increasingly serious possibility remained different from an established engineering capability.
Sound judgement had to hold those two facts together. Dismissing the concern because the final application was unproven would ignore the new evidence. Describing the application as already achieved would erase the work and uncertainty still ahead.
When Sharing Knowledge Became a Strategic Question
The exchanges that accelerated understanding also raised a difficulty. Information useful to one laboratory could be useful to another government’s researchers. The means by which scientists checked and developed knowledge could also spread knowledge with possible military consequences.
Henry DeWolf Smyth’s official 1945 account describes efforts in spring 1939, associated with Szilard and colleagues, to obtain voluntary restraint in publication. It records the failure to achieve complete agreement and the continuation of open publication, alongside some voluntary withholding. This was not yet the comprehensive secrecy later associated with the wartime project. [10]
The report was written after the wartime programme had succeeded. It is evidence for the administrative history, but its retrospective position matters. The participants of early 1939 did not possess its knowledge of the outcome. Their disagreement should not be rewritten as a contest between those who understood an inevitable future and those who failed to understand it.
There was a genuine tension. Sharing evidence allows others to challenge mistakes and repeat observations. Restricting it may limit access by potential opponents, but can also obstruct the exchanges through which the original claim becomes more reliable. The choice involves costs on more than one side.
This episode does not establish a universal rule that scientific openness is dangerous, or that secrecy is always protective. It reveals a change in the questions researchers were asking. Alongside “Is the result correct?” appeared “Who might use it, and should we change how it circulates?”
In the preceding chapters, politics had changed who could work in an institution. Here, the prospect of war was beginning to influence how work moved between institutions. The relationship between knowledge and power was changing at another point.
There Had Already Been an Approach to Government
The celebrated Einstein letter was not the first attempt to bring the subject to American officials. Smyth records that George Pegram of Columbia arranged a meeting between Fermi and Navy representatives in March 1939. According to that account, the Navy expressed interest and asked to be kept informed; the meeting did not produce the larger commitment the scientists sought. [10]
This matters because a history built around one famous letter can hide earlier attempts, partial responses and unsuccessful channels. Government attention is not a switch that remains entirely off until the best-known person speaks. Different officials can know different amounts, and being informed does not necessarily mean assuming responsibility.
There is also a difference between explaining a scientific possibility and presenting a decision that an institution can make. A laboratory may want support for further investigation. An official may need to know who will supervise it, which budget can meet the cost and what evidence would justify continuing. Those questions do not replace the science, but they shape whether the institution can act on it.
We encountered a similar problem in academic rescue. A scholar’s qualifications did not create an appointment without someone willing to arrange and support it. Here, the recognition of a potentially important result did not create a government programme without an administrative connection.
The analogy has limits. Providing refuge and investigating military possibilities are not morally equivalent activities. What they share is an organisational problem: information about a need has to reach someone able to take a relevant action.
The Letter With Two Dates
The letter signed by Einstein was dated 2 August 1939. The Franklin D. Roosevelt Presidential Library identifies 11 October as the date Sachs personally delivered it to Roosevelt. The outbreak of war in Europe intervened. More than two months separated the date on the document from its presentation to the President. [12]
That interval should remain visible. It prevents a familiar but misleading compression in which Einstein writes, Roosevelt immediately decides and a great industrial programme springs into existence. Each of those actions requires its own date and evidence.
Szilard was central to the effort to secure the warning. The National Park Service’s account also identifies Eugene Wigner and Edward Teller among the physicists involved in approaching Einstein. His signature brought unusual public and scientific authority to the message. It did not make him the director of the research being discussed, and he did not become a member of the later Manhattan Project. [13]
Sachs supplied a different connection: access to the President. The library describes him as a longstanding economic adviser to Roosevelt. The warning therefore travelled through people with different kinds of influence. Scientific expertise, recognised authority and political access were not interchangeable contributions. [12]
We need not invent Roosevelt’s private thoughts to understand the significance of that arrangement. The documents establish that the message reached him through a particular route. Its subsequent treatment can then be examined through the reply and the committee that followed.
For this book’s central question, the episode is especially revealing. Some of the people displaced by European persecution were not merely carrying expertise into a receiving country. They were also trying to communicate a danger to its government. The contribution was partly scientific and partly the work of getting a scientific concern taken seriously.
A Warning Was Not a Claim of Certainty
The letter distinguished the prospect of a powerful energy source from the less certain prospect of extraordinarily destructive bombs. It proposed continued contact between the administration and researchers, attention to supplies, and support that could extend work beyond university budgets. It also reported concern about German activity involving uranium. These were requests and assessments, not an announcement that a usable weapon already existed. [11]
The distinction between the two prospects is not a minor qualification. It shows that a serious warning can contain uncertainty about the precise form of the danger. The practical question is whether the possible consequences justify investigation before all uncertainty has disappeared.
It is possible to reason badly in either direction. Someone may dismiss a warning because it does not prove that the feared outcome will occur. Someone else may turn the same warning into a claim that the outcome is already inevitable. Neither response follows simply from evidence that a possibility has become scientifically credible.
The letter’s report of German activity must be handled with the same care. It is evidence of what Einstein’s warning communicated to Roosevelt. It is not, by itself, independent confirmation of how far any German weapons effort had progressed. Assessing an opponent’s actual capability requires evidence about that opponent’s work, not just the fears expressed by those watching it.
We can therefore separate two questions. Was the warning influential in American decision-making? And was every inference about German progress correct? The answer to the first does not settle the second. A belief can help cause an action even when the situation to which it refers remains uncertain.
This is one of the clearest examples of cause and effect in the series so far. The cause of a policy decision need not be an enemy’s completed achievement. It can be a judgement that the possibility of such an achievement is serious enough to warrant a response.
That observation does not establish that every subsequent decision was justified. It identifies the kind of reasoning at the point we have reached. To examine later commitments responsibly, we must follow the additional evidence, costs and choices that entered afterwards.
19 October: The President Replies
Roosevelt’s response, dated 19 October 1939, thanked Einstein and stated that a board had been convened to investigate the uranium question. It would bring together the head of the Bureau of Standards and representatives of the Army and Navy; Sachs would cooperate with it. The reply is short. Its significance lies in assigning the question an institutional destination. [14]
The government was no longer being asked merely to receive information. It had identified a body to examine the possibilities. That is a meaningful change, but it is not the same thing as authorising every future facility or choosing every later objective.
Lyman Briggs, director of the National Bureau of Standards, chaired the Advisory Committee on Uranium. It met on 21 October with military representatives and participating scientists and intermediaries. The Department of Energy’s administrative history identifies it as an early federal step towards the later atomic programme. It also describes the limited scale of the initial effort. [15]
There is a reason to resist writing “the Manhattan Project began” over this moment and moving on. Such wording can make a tentative investigation look like the organisation it eventually became. It gives the first committee the resources and certainty of a later stage that had not yet been created.
The Department of Energy’s account of Einstein’s letter makes the distinction explicit: the October decision was one among a sequence that eventually produced the Manhattan Project. It was important because it opened a route for government involvement, not because it completed the route. [16]
From this point onwards, the history needs new kinds of evidence. Correspondence about a discovery remains relevant, but so do committee reports, funding decisions and the allocation of responsibility. Once a government takes an interest, the question becomes how that interest is converted into sustained work.
A Report Was Not Yet a Programme at Scale
The committee’s report to Roosevelt was dated 1 November 1939. Smyth’s account records that it considered atomic power and bombs as possibilities and recommended further investigation. His administrative chronology then places the first transfer of Army and Navy funds in a memorandum of February 1940. A recommendation, a payment and a working research effort were distinct events. [10]
We stop at the threshold rather than follow those funds into the next chapter. By the end of 1939, the warning had helped create an official channel. It had not resolved all scientific uncertainties or assembled the later industrial system.
That distinction should not be turned into an automatic verdict that the early response was either adequate or disastrously slow. Judging its pace requires attention to the evidence available then, the alternatives considered and the consequences of different choices. The later existence of a successful weapon cannot simply be inserted into the information available to the first committee.
Nor is a larger budget by itself a complete explanation of success. Resources have to be connected to useful questions, capable investigators and ways of learning from results. The emerging problem was how to organise work whose importance might be enormous while much about its practical application remained unresolved.
It is here that the phrase “America assembled a team” begins to become an institutional question rather than a description of talented people living in the same country. Who could decide which investigation mattered? Who could provide continuity? Who could join work taking place in separate laboratories without pretending all the uncertainties were already settled?
Those questions are the bridge into the next stage of the main series. They cannot be answered by adding more famous names to the end of the Einstein letter.
What Germany Had Not Lost—and What It Had
The episode also requires us to refine the German side of the comparison. The chemical evidence that set this development in motion came from a German laboratory. That fact rules out a total-collapse account in which Nazi persecution had already made important scientific work inside Germany impossible. [1]
At the same time, the explanation involved Meitner after her forced departure. A crucial professional relationship continued through correspondence rather than through the shared working conditions it had previously possessed. Germany’s remaining capability and the damage done by persecution are not mutually exclusive findings.
The right question is therefore not whether Germany had either retained all its science or lost all of it. It is which capacities remained, which people were excluded, and how the changed relationships affected what could be done. That requires more precise evidence than a national label attached to a discovery.
A further distinction concerns intention. A finding made in a country is not necessarily a finding produced because its government correctly anticipated its later military use. To demonstrate that claim, we would need evidence about the research’s direction and support, not simply the eventual usefulness of the result.
The same standard applies to America. The presence of displaced scientists did not prove that the American government had previously planned every contribution they might make. The assistance networks of Part 4, the laboratories of this chapter and the committees at its conclusion performed different roles at different moments.
This is how the original thesis becomes stronger through qualification. Rather than a contest between two inventories of intelligence, we can now follow changing connections between people, evidence and authority. Some were broken by persecution. Others were rebuilt through refuge. Still others had to be created before research could become a coordinated national undertaking.
The Warning Had Arrived. The Larger Organisation Had Not.
Across this chapter, the same emerging possibility passed through several forms. It was first a problem in experimental evidence, then a physical explanation, then a subject of further investigation, then a strategic concern, and finally a matter assigned to a government committee.
Each transition required something the previous stage did not supply on its own. A puzzling result needed interpretation. An interpretation needed testing. A concern needed an audience able to act. Official attention needed a structure through which work could continue.
Nothing in that sequence allows us to say that persecution was necessary for discovery or that the later weapon was inevitable. It shows people maintaining and rebuilding connections despite damage, then confronting choices created by what their work revealed.
Nor should this nuclear story be mistaken for the whole scientific war. It is one closely documented route through which research became a question of national capability. The next stage will widen the view to the organisation of wartime science and the international cooperation that a country acting alone could not supply.
For now, we leave the narrative at the end of 1939. Roosevelt had received the warning. A committee had met. Researchers still needed answers, support and decisions about what should happen next.
Getting a warning onto the President’s desk was an achievement. Building an organisation capable of responding to it was another.
Continue to Part 6 — 1940–1941: Building the Scientific War Effort. The next chapter follows contracts, laboratories and British–American exchanges as scientific warnings and discoveries begin to acquire an organisation capable of acting on them. Radar and penicillin widen the story beyond uranium, while the British atomic assessment returns us to the question of industrial commitment. The sequence remains main series → prequel → sequel to the present → final hub.
Earlier chapters: Part 1 — Prologue · Part 2 — The comparative thesis · Part 3 — The purge before the war · Part 4 — The rescue of displaced scholars.
Sources and Chapter Notes
The principal narrative covers December 1938 through November 1939. The brief reference to funding in February 1940 marks the boundary with the next chapter. Publication dates are distinguished from the dates on which information circulated; the Einstein letter’s date is distinguished from its delivery. Oral recollections are identified as retrospective, not treated as verbatim contemporary records. Scientific discussion is limited to the concepts needed to understand the history, not the design or construction of a weapon.
[1] Max Planck Institute for Chemistry. The Discovery of Nuclear Fission. Institutional account of the December 1938 chemical findings, their earlier research context and the subsequent interpretation.
[2] Stuewer, Roger H. (2013). An Act of Creation: The Meitner–Frisch Interpretation of Nuclear Fission. In Traditions and Transformations in the History of Quantum Physics, Max Planck Research Library, chapter 9, especially sections 9.4–9.5. Historical reconstruction with references to correspondence and recollections.
[3] Meitner, Lise; Otto R. Frisch (11 February 1939). Disintegration of Uranium by Neutrons: A New Type of Nuclear Reaction. Nature, 143, 239–240. Primary text reproduced by German History Intersections. The article text, rather than the reproduction page’s introductory geographical summary, is used here.
[4] Frisch, Otto R. (18 February 1939). Physical Evidence for the Division of Heavy Nuclei under Neutron Bombardment. Nature, 143, 276. The separate experimental report; publication details checked against the journal record.
[5] American Institute of Physics, Moments of Discovery. The Discovery of Fission: Frisch’s Recollections. Recorded recollections about collaboration by telephone, the interpretation and the naming of fission. Recollections are not a contemporary transcript of the 1938 discussions.
[6] American Institute of Physics, Moments of Discovery. The Discovery of Fission: Dunning and Anderson’s Recollections. Evidence of the questions discussed at Columbia and the distinction between observing fission and investigating a continuing reaction.
[7] Anderson, H. L.; E. T. Booth; J. R. Dunning; E. Fermi; G. N. Glasoe; F. G. Slack (1 March 1939). The Fission of Uranium. Physical Review, 55, 511. Used for the contemporary publication record, authorship and Columbia affiliation. Names also checked against the Columbia Department of Physics historical account.
[8] von Halban, H.; F. Joliot; L. Kowarski (18 March 1939). Liberation of Neutrons in the Nuclear Explosion of Uranium. Nature, 143, 470–471. Contemporary research report; the title refers to the nuclear process, not a completed bomb.
[9] Anderson, H. L.; E. Fermi; Leo Szilard (1 August 1939). Neutron Production and Absorption in Uranium. Physical Review, 56, 284. Journal metadata used for title, authorship, affiliation and publication date; no claim of a detailed technical reanalysis is made.
[10] Smyth, Henry DeWolf (1945). Atomic Energy for Military Purposes, Chapter III: Administrative History up to December 1941, especially sections 3.1–3.6. Official account published in 1945 and reproduced by Atomic Archive. Used for publication restraint, the March 1939 Navy approach, the November report and the timing of initial funding; not treated as a complete or neutral account of every participant’s motives.
[11] Einstein, Albert (2 August 1939). Letter to President Franklin D. Roosevelt. Primary text reproduced by German History in Documents and Images. Used to distinguish the document’s proposals, predictions and reported concerns from independently established enemy capabilities.
[12] Franklin D. Roosevelt Presidential Library and Museum (13 October 2011). Found in the Archives: Einstein Letter. Confirms the letter’s 2 August date and Sachs’s delivery on 11 October 1939. National Archives document description and archival identifier.
[13] National Park Service. Manhattan Project Pioneers: Albert Einstein. Distinguishes the warning initiative from participation in the later Manhattan Project.
[14] Roosevelt, Franklin D. (19 October 1939). Response to Albert Einstein. Primary reply reproduced by Atomic Archive and attributed to the Franklin D. Roosevelt Presidential Library and Museum.
[15] U.S. Department of Energy, Office of History and Heritage Resources. Committee on Uranium. Administrative history of the committee’s formation, chairmanship, first meeting and later reorganisations.
[16] U.S. Department of Energy, Office of History and Heritage Resources. Einstein’s Letter, 1939. Historical account identifying the October response as an early decision in a longer process, rather than the immediate creation of the complete Manhattan Project.