Does nuclear secrecy make us more secure? New book offers counterargument
The revolutionary discovery of nuclear fission in December 1938 helped launch the atomic age, bringing with it a unique need for secrecy regarding the scientific and technical underpinnings of nuclear weapons. This secrecy evolved into a special category of proscribed information, dubbed “Restricted Data,” which is still in place today. Historian Alex Wellerstein spent over ten years researching various aspects of nuclear secrecy, and his first book, Restricted Data: The History of Nuclear Secrecy in the United States (University of Chicago Press), was released earlier this month.
Wellerstein is a historian of science at the Stevens Institute of Technology in New Jersey, where his research centers on the history of nuclear weapons and nuclear history. (Fun fact: he served as a historical consultant on the short-lived TV series Manhattan.) A self-described “dedicated archive rat,” Wellerstein maintains several homemade databases to keep track of all the digitized files he has accumulated over the years from official, private, and personal archives. The bits that don’t find their way into academic papers typically end up as items on his blog, Restricted Data, where he also maintains the NUKEMAP, an interactive tool that enables users to model the impact of various types of nuclear weapons on the geographical location of their choice.
The scope of Wellerstein’s thought-provoking book spans the scientific origins of the atomic bomb in the late 1930s all the way through the early 21st century. Each chapter chronicles a key shift in how the US approach to nuclear secrecy gradually evolved over the ensuing decades—and how it still shapes our thinking about nuclear weapons and secrecy today.
Along the way, we meet such pivotal figures as Vannevar Bush and James Conant, as well as famous Manhattan Project scientists like Robert Oppenheimer, embedded journalist William Laurence, and notorious Soviet spies Klaus Fuchs and Julius and Ethel Rosenberg. Wellerstein delves into the establishment (and eventual dissolution) of the post-war Atomic Energy Commission, the emergence of the Cold War, and how attempts to reform the system failed (due in part to partisan politics), leaving the US with an outdated nuclear secrecy policy that is arguably not especially effective.
“One of the things that makes American nuclear secrecy so interesting is that it sits at a very interesting nexus of belief in the power of scientific knowledge, the desire for control and security, and the underlying cultural and legal values of openness and transparency,” Wellerstein writes in his introduction. “These at times mutually contradictory forces produced deep tensions that ensured that nuclear secrecy was, from the beginning, incredibly controversial, and always contentious, and we live with these tensions today.”
Ars sat down with Wellerstein to learn more.
Ars Technica: Why is there still so much interest in this period of US history?
Alex Wellerstein: I think there is an inherent draw toward nuclear weapons because of their power and their persistence. Even if we got rid of all of them tomorrow, we would still be fascinated with their history and their development because they represent some level of the maximum that we as clever creatures can accomplish, for both good and for ill. The Cold War feels like it’s further away from us, but we still live in a world that’s shaped by it, and it really wasn’t that long ago. We had the historical conditions that led to these states building massive nuclear arsenals, doing so with huge amounts of secrecy around them, knowing that if they ever use these weapons, it could potentially be catastrophic. I think it’s very telling about human beings and the types of creatures we are.
Ars Technica: A major theme running throughout your book is the tension between the need for secrecy and the ideal of free and open science.
Alex Wellerstein: You cannot make these weapons without relying heavily on advanced scientific input. Scientists typically share an ideology that developed over the 19th and 20th centuries about what it means to be a scientist. It usually does not mean that they want to be technicians; they actually look down on technicians and engineers—people they see as being simply transactional in their knowledge, people who are just fulfilling a role. The scientists, especially the physicists, look at their job as being explorers of the natural world. They often identify not along national lines but on professional lines. They see themselves as being apart from the world to some degree.
So you have these conflicting desires, even within individual people. I spent a lot of time in the book talking about Leo Szilard. I like him as a character because he was, in certain ways, really conflicted. He believed in the openness of science. He did not believe that military secrecy is a good thing; he thought that it would be misused. He believed scientists need to have total freedom of movement. At the same time, he also was terrified of the Nazis. So he had to try to come up with ways of reconciling these two impulses, which ultimately left him pretty unsatisfied because there really isn’t a great way to reconcile them.
Ars Technica: Many Jewish physicists fled Nazi Germany and occupied countries and ended up working on the Manhattan Project. That experience couldn’t help but color their perceptions.
Alex Wellerstein: It’s not even that many as a proportion, but their impact is extremely disproportionate. That’s not a coincidence. If your project requires people who really take the threat seriously, there’s nobody who takes the threat of a Nazi atomic bomb more seriously than Jewish refugees from Nazism. Somebody who’s a native-born American might say, “Well, I don’t know what the odds are that it’s possible.” These are the people who are going to say, “It doesn’t matter if there’s a low chance because the consequences are unimaginable. This is not some petty squabble with a fool. This is a genocidal experience. And if you don’t get your act together, it’s going to come for you, too.”
It’s also part of the answer to why so many of the spies were Jewish—because of the history, especially in New York City, of Judaism and communism. This was a time in which many of the Jews felt that the United States and the capitalist world were not doing enough to combat fascism, so [Stalin] looked like a viable alternative.
Ars Technica: Just how extensive were the spy networks?
Alex Wellerstein: We’ve learned a lot more about the spies in the last 15–20 years, including the size of the Soviet spying effort, partially through the release of the Venona transcripts, which are intercepted Soviet decrypts. There were so many communications from World War II that were decrypted after and revealed the existence of these spy networks. And there has been at least one major case of a former Soviet agent grabbing all of his old books and [defecting] to the United States, which gave the codenames of everybody who was in the Venona transcripts.
As far as we can tell, there were zero spies for Nazi Germany in the Manhattan Project, zero spies for Japan, zero spies for Italy. The Soviet spy apparatus in the United States consisted of several hundred people in different roles, maybe 10 of whom were connected in some way to the Manhattan Project. Of those 10, two or three at Los Alamos actually knew very much. And of those, Klaus Fuchs was the only one with deep connections. Through him, you got David Greenglass and Julius and Ethel Rosenberg. Fuchs could give the Russians detailed diagrams, with measurements, of every part in the bomb.
There were probably at least 10,000 scientists working on the Manhattan Project out of a total labor pool of 500,000 people, so having one crucial spy is not very surprising. They were so preoccupied with other things, that really wasn’t what they were worried about. In the book, I describe it as an impossible mandate because it’s literally impossible to imagine screening out those people and not having one of them slip in under the radar. With Fuchs, they didn’t screen him at all because he was part of the British delegation, so he basically got a free pass. If they had looked closely at him, they probably would have raised some questions about what he was doing and what his political views were.
There was an event for the 70th anniversary of the Manhattan project at the Atomic Heritage Foundation a couple of years ago. A physicist named Ben Bederson spoke; he was bunkmates with David Greenglass in the same special engineering detachment. He said, “Oh, Greenglass was obviously a [communist], he talked about it all the time. I tried to get transferred out of his bunk because he was so annoying. He never tried to recruit me, but he assumed that because I was from the same part of New York as he was, that I had similar opinions because I was Jewish. If anybody had ever asked me once, ‘Are there any communists in here?’ I’d have said, ‘Obviously, David Greenglass.’”
Ars Technica: While researching your book, did you learn anything that really surprised you?
Alex Wellerstein: One was the fact that in the US we still have this parallel separate system for nuclear weapons secrets that is different from any other kind of secrecy. “Restricted Data” was a specially created category for nuclear weapons in 1946 because they really just were not sure what to do with this new concept. So we still have a very 1940s style system. There’s a lot of reasons one could imagine for saying, “Maybe we don’t need to treat nuclear weapons as a totally parallel system from everything else in the world. Maybe that’s not the best way—maybe we’re in some ways inflating the value of this information by doing that.”
There is an alternative argument, which is that secrets don’t control nuclear weapons very well. It seems obvious to most people, and certainly did to me when I started this, that knowledge is power. Nuclear weapons are sort of infinite power, so their knowledge should be infinitely important, right? But the counterargument—and Oppenheimer was one of the first to really put this out there in a strong policy-framed way—is that secrecy is about control of a certain type of information, what philosophers might call “explicit information,” stuff you can write down. You can restrict tons of knowledge just by not letting your experts go to another country and show them how to do stuff.
But that is only a small percentage of what it takes to actually make a weapon, specifically a nuclear weapon. As a result, it might not be the thing you want to focus on to control these weapons. You might want to focus on controlling the processes to make the fuel because that turns out to be the necessary thing. I can draw for you a beautiful sketch of how to make a thermonuclear weapon, but it’s not going to help if you don’t have the fuel—and you don’t because we restrict that.
You could get rid of all of the secrecy tomorrow and the world would not measurably become more dangerous, because it’s other things that are actually keeping these weapons from spreading. To me, it’s still a pretty radical idea because it not only goes against our intuitions about the bomb, but it also goes against what we tell ourselves about the way in which technology functions. It’s not the equation that gives you the technology; it’s the overall socio-political, human system that causes it to exist in the first place.
Ars Technica: What are some lessons from our past history of nuclear secrecy that we can apply today?
Alex Wellerstein: I’m not convinced that all this secrecy has gotten us a whole lot of security. The rhetoric of nuclear weapons in general, especially in the US, is about control. It’s about our desire to put shackles on this force because we recognize that if it fell into the wrong hands, or if things got out of our control, the consequences would be unacceptably large. The problem is that we can easily fool ourselves into thinking we have more control than we do. Secrecy can do that in two ways.
One is you can think, well, if this information isn’t known to somebody, then they can’t use it. This is what the cryptographers called security through obscurity. They hate it because it assumes that your enemy doesn’t know it—yet your enemy probably does. There’s no reason for you to assume that you’re not compromised in some way, so you’d be better off setting up a system that does not require your enemy not to know something because knowledge is extraordinarily easily to transmit, even in the 1940s. Now we’re in the information age, and you leak knowledge just by leaving the house.
Also, secrecy can prevent independent auditing and can lead to a kind of self-censorship. If you start talking online about anything related to nuclear weapons, people say, “Oh, I think we’re on the list now.” They’re joking, but it does reflect an institutionalized paranoia, [a sense] that this isn’t really for us, and we can’t comment on it. Only people who are on the inside really know about it. The people on the inside feel this way, too. They dismiss those on the outside as not being informed enough. This can lead to a false sense of control because you have people on the inside who are not really in dialogue with a broader, critical world. You have a lot of groupthink going on. So nobody is really in control of this thing, and yet nobody is owning up to that fact.
Ars Technica: There have been a lot of Hollywood portrayals of nuclear weapons over the decades in film and television, some quite powerful and iconic. But apart from the short-lived series Manhattan several years ago, the subject doesn’t seem to resonate with our 21st century culture as much as it used to.
Alex Wellerstein: We think in really metaphorical terms, but we don’t have a lot of great instant cultural metaphors to draw on anymore. We can say, “Oh, it’s a Dr. Strangelove situation, or it’s a WarGames situation.” But we don’t have something for our current modern-day situation.
We take seriously the threats that we can visualize. We also have almost no cultural language for talking about a non-kumbaya version of nuclear disarmament. The only way in which people think about nuclear disarmament is a bunch of hippies getting together and saying, “Oh, it’s gone.” And then we just live happily after. The disarmament people don’t believe that. That’s not their goal, either. But we don’t have anything that looks like a model for a world in which people don’t feel the need to have nuclear weapons.
I don’t think you can go to Hollywood and say, “Give us new imagery.” It works the other way around. Dr. Strangelove worked not because it told people in the 1960s how to think about nukes but because it reflected how people thought about nukes in a very potent and clever way. Unfortunately, at the moment, nobody’s really thinking about nukes, so they don’t really get a lot of representation. They are out of sight, out of mind, but they’re still there. People ought to think about them a little bit more than they do. In order to get another WarGames or Dr. Strangelove, you need a shift in the culture first. And I don’t know how to do that.