Meditations

Narratives of Manhattan Project secrecy

by Alex Wellerstein, published March 29th, 2013

Secrecy suffused every aspect of the Manhattan Project; it was always in the background, as a context. But it’s also a topic in and of itself — people love to talk about the secrecy of the work, and they’ve loved to talk about it since the Project was made public. In the 1940s there was something of a small industry of articles, books, and clichés regarding how secret the atomic bomb was kept. Of course, the irony is… it wasn’t really kept all that well, if you consider “keeping the secret” to involve “not letting the Soviet Union know pretty much everything about the atomic bomb.” (Which was, according to General Groves, one of the goals.)

It’s easy to get sucked into the mystique of secrecy. One way I’ve found that is useful to help people think critically about secrecy (including myself) is to focus on the narratives of secrecy. That is, instead of talking about secrecy itself, look instead at how people talk about secrecy, how they frame it, how it plays a role in stories they tell about the Manhattan Project.

One of many early articles in the genre of Manhattan Project secrecy: "How We Kept the Atomic Bomb Secret," from the Saturday Evening Post, November 1945.

One of many early articles in the genre of Manhattan Project secrecy: “How We Kept the Atomic Bomb Secret,” from the Saturday Evening Post, November 1945.

My first example of this is the most obvious one, because it is the official one. We might call this one the narrative of the “best-kept secret,” because this is how the Army originally advertised it. Basically, the “best-kept secret” narrative is about how the Manhattan Project was sooo super-secret, that nobody found out about it, despite its ridiculous size and expense. The Army emphasized this very early on, and, in fact, Groves got into some trouble because there were so many stories about how great their secrecy was, revealing too much about the “sources and methods” of counterintelligence work.

The truth is, even without the knowledge of the spying (which they didn’t have in 1945), this narrative is somewhat false even on its own terms. There were leaks about the Manhattan Project (and atomic bombs and energy in general) printed in major press outlets in the United States and abroad. It was considered an “open secret” among Washington politicos and journalists that the Army was working on a new super-weapon that involved atomic energy just prior to its use. Now, it certainly could have been worse, but it’s not clear whether the Army (or the Office of Censorship) had much control over that.

Panel from FEYNMAN by Jim Ottaviani and Leland Myrick.

Panel from FEYNMAN by Jim Ottaviani and Leland Myrick.

We might contrast that with the sort of narrative of secrecy that comes up with regards to many participants’ tales of being at places like Los Alamos. Richard Feynman’s narrative of secrecy is one of absurd secrecy — of ridiculous adherence to stupid rules. In Feynman’s narratives, secrecy is a form of idiotic bureaucracy, imposed by rigid, lesser minds. It’s the sort of thing that a trickster spirit like Feynman can’t resist teasing, whether he’s cracking safes, teasing guards about holes in the fence, or finding elaborate ways to irritate the local censor in his correspondence with his wife. All participants’ narratives are not necessarily absurd, but they are almost always about the totalitarian nature of secrecy. I don’t mean “fascist/communist” here — I mean the original sense of the word, which is to say, the Manhattan Project secrecy regime was one that infused every aspect of human life for those who lived under it. It was not simply a workplace procedure, because there was no real division between work and life at the Manhattan Project sites. (Even recreational sports were considered an essential part of the Oak Ridge secrecy regime, for example.)

So we might isolate two separate narratives here — “secrecy is ridiculous” and “secrecy is totalitarian” — with an understanding that no single narrative is necessarily exclusive of being combined with others. 1

"Beyond loyalty, the harsh requirements of security": Time magazine's stark coverage of the 1954 security hearing of J. Robert Oppenheimer.

“Beyond loyalty, the harsh requirements of security”: Time magazine’s stark coverage of the 1954 security hearing of J. Robert Oppenheimer.

But the Feynman approach looks perhaps unreasonably jolly when we contrast it to the narrative of J. Robert Oppenheimer and his students, for whom secrecy became something more sinister: an excuse to blacklist, a means of punishment. Oppenheimer did fine during the Manhattan Project, but the legacy of secrecy caught up with him in his 1954 security hearing, which effectively ended his government career. For his students and friends, the outcomes were often as bad if not worse. His brother, Frank, for example, found himself essentially blacklisted from all research, even from the opportunity to leave the country and start over. (It had a happy ending, of course, because without being blacklisted, he might never have founded the Exploratorium, but let’s just ignore that for a moment.)

For a lot of the scientists involved in the Manhattan Project, secrecy ended up putting their careers on the line, sometimes even their lives on the line. In response to (fairly ungrounded) suspicions about Oppenheimer’s student Rossi Lomanitz, for example, Groves actually removed his draft deferment and had him sent into the dangerous Pacific Theatre. This narrative of secrecy is what we might classically call the “tragic” narrative of secrecy — it involves a fall from grace. It emphasizes the rather sinister undertones and consequences of secrecy regimes, especially during the period of McCarthyism.

The original "best-kept secret" story, released on August 9, 1945 (the day of the Nagasaki bombing).

The original “best-kept secret” story, released on August 9, 1945 (the day of the Nagasaki bombing).

So what other narratives are there? Here is a short list, in no particular order, that I compiled for a talk I gave at a workshop some weeks ago. I don’t claim it to be exhaustive, or definitive. Arguably some of these are somewhat redundant, as well. But I found compiling it a useful way for me to think myself around these narratives, and how many there were:

  • Secrecy is essential”: early accounts, “best-kept secret” stories
  • Secrecy is totalitarian”: secret site participants’ accounts
  • Secrecy is absurd”: e.g. Feynman’s safes and fences
    • Common hybrid: “Secrecy is absurdly totalitarian
  • Secrecy is counterproductive”: arguments by Leo Szilard et al., that secrecy slowed them down (related to the “absurd” narrative)
  • Secrecy is ineffective”: the post-Fuchs understanding — there were lots of spies
  • Secrecy is undemocratic”: secrecy reduces democratic participation in important decisions, like the decision to use the bomb; fairly important to revisionist accounts
  • Secrecy is tragic”: ruinous effects of McCarthyism and spy fears on the lives of many scientists
  • “Secrecy is corrupt: late/post-Cold War, environmental and health concerns

It’s notable that almost all of these are negative narratives. I don’t think that’s just bias on my part — positive stories about secrecy fit into only a handful of genres, whereas there are so many different ways that secrecy is talked about as negative. Something to dwell on.

What does talking about these sorts of things get us? Being aware that there are multiple “stock” narratives helps us be more conscious about the narratives we talk about and tap into. You can’t really get out of talking through narratives if you have an interest in being readable, but you can be conscious about your deployment of them. For me, making sense of secrecy in an intellectual, analytical fashion requires being able to see when people are invoking one narrative or another. And it keeps us from falling into traps. The “absurd” narrative is fun, for example, but characterizing the Manhattan Project experience of secrecy makes too much light of the real consequences of it.

As an historian, what I’m really trying to do here is develop a new narrative of secrecy — that of the meta-narrative, One Narrative to Rule Them All, the narrative that tells the story of how the other narratives came about (a history of narratives, if you will). Part of talking about secrecy historically is looking at how narratives are created, how they are made plausible, how they circulate, and where they come from. Because these things don’t just appear out of “nowhere”: for each of these narratives, there is deep history, and often a specific, singular origin instance. (For some, it is pretty clear: Klaus Fuchs really makes the “ineffective” narrative spring to live; Leo Szilard and the Scientists’ Movement push very hard for the “counterproductive” narrative in late 1945; the “best-kept secret” approach was a deliberate public relations push by the government.)

As a citizen more broadly, though, being conscious about narratives is important for parsing out present day issues as well. How may of these narratives have been invoked by all sides in the discussions of WikiLeaks, for example? How do these narratives shape public perceptions of issues revolving around secrecy, and public trust? Realizing that there are distinct narratives of secrecy is only the first step.

  1. Both of these might classically be considered “comic” narratives of secrecy, under a strict narratological definition. But I’m not really a huge fan of strict narratological definitions in this context — they are too broad.[]

18 Responses to “Narratives of Manhattan Project secrecy”

  1. Gabriel Henderson says:

    I am reminded of the Obama Administration not willing to discuss in any detail — or even acknowledge the existence of — the use drones against its perceived enemies. “Ummmm, no, I cannot talk about that program — a program I have no knowledge of and could not admit the existence of even if I wanted to.” The maintenance of secrecy once the secret is out . . . simultaneously disheartening and comedic.

  2. Blake says:

    I assume you’ve probably seen it, but just in case, someone uploaded the Oppie-Murrow interview from the 50s. Since I’ve never seen video of him more than about a minute long and not reciting his Vishnu line, I thought it was really interesting. He touches on secrecy related things a couple times.

    • I hadn’t seen that, thanks! I had been looking for this for awhile; I heard it was a special feature on one of the Oppenheimer BBC series discs but never got around to tracking it down.

  3. Bill Higgins says:

    It’s notable that almost all of these are negative narratives.

    Positive narratives might hinge on how well secrecy worked. The biggest fear at the beginning of the project was losing a race with the Germans.

    How much did Germany know about Allied efforts to develop nuclear weapons?

    Historians must know this, but I’m not very familiar with the evidence. (What should I read?) As far as I know, the answer is “practically nothing.” I know even less about Japanese or Italian intelligence on the Bomb.

    “Secrecy is essential” is based on fear of the consequences, if adversaries should learn about the project. “Secrecy is effective” would be a positive narrative, strengthened postwar by evidence that Germany knew little.

    “Secrecy is effective” could also be reinforced, in a backhanded way, by the speed with which the Soviets developed the Bomb, coupled with revelations of multiple Soviet spies. “See? Once their spies stole our secrets, it only took four years for their scientists to catch up!” (There may be controversy today over how much credit is due to the spies, and how much to the scientists, but a participant in U.S. security debates circa 1950 might not care about such nuances.)

    Secrecy kept the Germans in the dark, and they didn’t develop nuclear weapons. Secrecy failed, and the Soviets did develop nuclear weapons. This could support an argument, circa 1950, that the U.S. and its allies need more and better secrecy.

    • I haven’t really seen very much about what the Germans or Japanese did or did not know about the Manhattan Project. I’ve poked around a little myself (it is an interesting question, and one that an evaluation of the efficacy does require knowing something about), but not in a concerted way; thanks for reminding me of it, I will make some inquiries.

      There are hints that it was a non-zero amount of information. Heisenberg’s famous Copenhagen conversation with Bohr, for example, implies he is aware to some degree of Allied interest in nuclear energy. There are accounts that imply that Japanese scientists were suspicious. But suspicion and rumor are not the same thing as verified knowledge, or even real fears. The reaction of the Germans at Farm Hall appears to be genuine surprise that the US and UK pulled it off — to the point of suspecting the announcement of Hiroshima to be propaganda.

      But a lot of this was just luck. It actually wasn’t very hard to figure out that a bomb program was going on, once you thought to ask the question. Georgii Flerov figured it out just on the basis of the lack of publications on fission; a group of Indian physicists figured it out as well just by paying attention (someday I will write about this episode). That this is so isn’t too surprising — the Manhattan Project was just too large, and even the blank spaces it left were conspicuous.

    • J C says:

      The Farm Hall tapes do provide a great deal of insight, because they reveal that the Germans did not know what was (in my opinion) the only significant secret of the Manhattan Project: The Germans didn’t know that we were sure it would work. More specifically, they didn’t know that we had enough confidence in it to put more resources into the Manhattan Project than we put into pushing jet aircraft into production. Indeed, as you note, they show the Germans were so sure it could NOT work that they believed the newspaper story was a plant to get them to give up what they knew.

      The Farm Hall tapes also reveal that once the Germans believed that an atomic bomb would work, they identified the errors in their analysis and calculations and figured out in about a day that a Uranium bomb would be easy to build. So we kept the only secret that mattered during WW II until we used the weapon.

      All of the other secrets concerned technical details that affect how long it takes to develop the device, but have little impact on whether you could ultimately do so. And if I recall correctly, the Russians spent a lot of time checking what they had because they weren’t sure whether the spies had gotten disinformation or real information. Many people forget that we did it in 3 years from the conceptual design in the summer of 1942 that Serber recorded, and the Russians would be starting from about the same point. What was surprising was that it only took them one extra year despite starting with a badly damaged industrial base in 1945.

      • “… figured out in about a day that a Uranium bomb would be easy to build.”

        Assuming you had the highly-enriched uranium, of course, which was well over half the expense of the project.

        (As an aside, even by at least 1947 — perhaps even later — Heisenberg publicly stated he thought that you could use plutonium in a gun-type arrangement, indicating that he was apparently ignorant of the difficulty of using reactor-grade plutonium in a weapon design.)

        I don’t add that to be pedantic, but just to indicate that there are many levels of possible secrets there. A good idea of the critical mass is a necessary precondition to deciding that a bomb is a feasible thing to make, but it’s a long way from devoting the resources to making one. Developing a successful enrichment program before anyone else had proven that it could be done (if one is willing to spend a lot of resources on it) is no small matter. I think we fixate a little too much on the designs as the possible secrets to be gained or lost; General Groves certainly didn’t think they were the important parts, and there were quite “in-the-know” people arguing that even implosion could be declassified as early as late 1945, because knowing about it really wasn’t what kept anyone from getting a bomb.

        That being said, my understanding is that the main factor in the Soviets’ development of the bomb was their original lack of good raw uranium resources (they had very few sources of uranium ore, and most of them were very low grade compared to what the US was using). And you’re right, they didn’t use the espionage information very efficiently, though it’s admittedly still hard to guess from that what their speed would have been without it. Personally I lean towards guessing that with or without espionage they would have gotten a bomb around the same time that they did, but that’s only because I think that getting some kind of bomb is mostly a factor of throwing resources at the problem, as opposed to being terribly clever. (Cleverness arguably gets you better bombs, e.g. a better yield-to-weight ratio.)

        • J C says:

          Excellent points, especially to a pedant!

          I recommend that anyone interested in this question read the Farm Hall book, if only to read how his colleagues slam Heisenberg for getting the critical mass of U wrong (by a large factor) without realizing that they should have checked his work without regard to his fame as a theoretical physicist.

          Also agreed on the pace of the Soviet program. I’ve often pointed out to classes that touch on this topic that developing nuclear weapons requires a major industrial effort on a scale that can only be guessed at even if you compare the size of just the Y-12 plant to a college campus. As you noted, simply knowing that the separation problem can be solved makes it easier to throw money and resources at the problem.

    • GeBeWubya says:

      The ultimate positive secrecy narrative is silence. Those not in-the-know not only do not know the secret, they do not know there is a secret.

  4. J C says:

    Back on your main points about secrecy itself, there were some other ones that must be in the competition for “best kept”. However, none of these were on a scale to rival the massive industrial facilities at Oak Ridge and Hanford whose purpose had to be concealed.

    One would be the mere existence of the Colossus digital computer and other computer-like code-breaking devices at Bletchley Park. The U.S. publicized the ENIAC and even put the MANIAC in movies, but the British machine wasn’t even hinted at (until the classification expired) by people who knew they had built the first computer. What the US knew about breaking Japanese codes (and hence related codes used by others) or the use of a broken “one time” Russian code to catch the Rosenbergs (the VENONA project) are also candidates.

    It would be interesting to see what the KGB archives have to say about what and when the Russians started looking into those projects. Just because Americans didn’t know about them didn’t mean they were actually secret, but what mattered the most was whether the Germans or Japanese or Russians (or Americans) knew their own codes were broken.

  5. Sandy says:

    There are a whole set of related things where WW II secrecy was very interesting, and a bunch of issues around allied co-operation.

    Bletchley was a big one. The really big deal there was breaking the Enigma machine and building the bombes to crack it. The Poles actually did that back in the early 30s but kept it secret, even from their allies, until ’39. Then they told the Brits & French; the Brits took their stuff, improved it, and built Bletchley. They also kept it secret, even from allies. The Poles were never told about Bletchley, nor were allies like Russia, China & the Free French. The Americans were not let in on it until ’42.

    The British “tube alloys” project, and related Canadian work on uranium mining & refining, started well before the Manhattan Project. Once the US got involved, the scientists involved were all sent there. Some of them complained that the security was asymmetric; they were expected to reveal everything they knew to the Americans, but were not cleared to know what the Americans were discovering.

  6. In the second, 1955 edition of my favorite book, Philip Wylie’s 1942 best seller /Generation of Vipers,/ there’s a footnote about the fact that he made a not-especially-veiled reference to the atomic bombardment of cities in his 1942 book. Apparently it became a standard question at every lecture and public appearance for people to ask him how he knew about atomic bombs in 1942. His response was dismissive: “H.G. Wells knew about atomic bombs.”

    This suggests another meta-narrative you may want to add to your list: “Secrecy is impossible.” Possible additional examples would be the old mafia rule of thumb that “Three people can keep a secret if two of them are dead.” A perhaps better example would be that the Soviet ambassador received his copy of The Pentagon Papers before the President himself did, and used to show it off to visitors to his office. Or, for that matter, the “secret” bombing of Cambodia in the 1970s that was, come on now, hardly a secret to the Cambodian people who were being bombed, or to any reporter who talked to the survivors over the next week, or the “secret” CIA drone program that has been in the news practically since the first day it started.

    There’s even a Manhattan Project connection to the “secrecy is impossible” meta-narrative: the Germans didn’t know how successful the Manhattan Project was being, but it was in fact impossible to keep it secret that we were working on it and in a big way: one thing that was pointed out at the time was the fact that every single US atomic physicist of any note suddenly stopped going to conferences and publishing research papers at the same time, and that was something that couldn’t be hidden.

  7. paul says:

    There’s something implicit here that should, I think, be made explicit, namely that certain kinds of secrecy also involve almost complete transparency to security operatives. If you read some of the transcripts of the Oppenheimer hearings, for example, it becomes clear that he lived his Manhattan-project life with a security officer never more than a dozen or so yards away. Constant reports of Oppenheimer’s activities (and presumably those of everyone else on the project) flowed to the upper echelons and to some set of archives. That enormous flow of information is part of the totalitarian narrative, but also in some ways separate from it.

  8. Darrell Dvorak says:

    Three comments: First, thanks to Alex for framing a fascinating “nuclear secrecy” topic. I’m not schooled enough to say whether his framing is unique, but it is nevertheless fundamental.

    Second, on the positive side of the ledger, Manhattan secrecy at least delayed the Soviets’ acquisition of nuclear weapons. Imagine the outcomes, especially in Europe, if they had even a temporary monopoly.

    Third, as a professional historian, Alex is sensitive to the power of “narratives” to skew historical accounts. As a non-professional with a personal interest in the facts of a slice of the Manhattan Project narrative, to me his probing is invaluable.

  9. […] Narratives of Manhattan Project secrecy | Restricted Data – […]

  10. […] How people talk about the secrecy that surrounded the Manhattan Project (Nuclear Secrecy) […]

  11. […] power to notice and put together regarding the bomb effort. It was not quite so perfectly secret as we often talk of it as being. We know it was possible to put some of the pieces together, because the Soviets did it, and even […]

  12. […] of nuclear weapons was unthought except inside classified circles. This is a side-effect of the narratives we tell about Manhattan Project secrecy, which emphasize how extreme and successful these restrictions on […]