This week is an Archives Week! Which means I’ll be spending the whole week in an archive, doing research, with the aim of posting something amusing for you, dear reader, each and every day of the week. This might seem ambitious, but the best part about working on nuclear history is that each and every archival box contains something so surreal it would knock Dali’s socks off. No doubt every giant bureaucracy produces its Kafkaesque moments, but mixing them with the potential to wipe out entire nations makes them into something sublime.
I’m camping out all week in the Legislative Archives, which are housed in the downtown DC National Archives building. Veteran researchers know that the downtown National Archives are not where most of the research records are kept (most were moved, some time back, to the Archives II facility in College Park). But the Legislative Archives are one of the exceptions, so I get to spend the week in the same building that the tourists all go to when they want to see original copies of the Constitution.
I’ll be looking primarily at the records of the Joint Committee on Atomic Energy (JCAE). The JCAE was created in the Atomic Energy Act of 1946 to serve as the Congressional oversight committee for the U.S. Atomic Energy Commission (AEC). What this means, in a nutshell, is that the JCAE was a group of Senators and Representatives who spent most of their time being very unhappy with the AEC, with the ability to make life difficult for the AEC. They were immensely powerful, as Congressional committees go, and, because of the national security implications, incredibly secretive. They were also, on the whole, neither scientists, professional bureaucrats, or military men, so they often provided a, shall we say, “unique” take on the major nuclear issues of the day.
So keep an eye on this spot, and you’re sure to be rewarded with some amusing archival finds, just in time for the holidays!
Incidentally — so today’s post isn’t just a total bore — I tried to come up with some quantitative data on how secret the JCAE really was. (Though the heavy presence of graphs might indeed make it even more of a bore…)
Here is a graph (created using data from the Lexis-Nexis archive which I previously discussed here) of the number of open and secret (“Executive” hearings) that the JCAE held over its 14 Congressional sessions (1947-1976):
So there are some definite trends here — in the early days, there were a LOT of secret hearings. However this data (which just counts hearings) is a little misleading, since 1. some of the “published” hearings are actually multi-day affairs (and the classified hearings never are), and 2. not all hearings are equal. Should we consider an hour-long secret hearing to be the same as a five hour open hearing?
With a little more digging, and a script that extracted page numbers from the Lexis Nexus data, I came up with this variation on the above graph. This one shows the number of “secret” versus “open” pages:
Now this tells a different story — we can see there is still quite a lot of secrecy early on, but the amount of public hearings goes up quite drastically.
One thing I should also point out: for the first Congressional session, the “secret” meetings were not always compiled verbatim, but as summaries. The “public” sessions are always verbatim, though. So there’s some under-counting of secret words/pages for the first session.
Still, even this graph masks a particular tweak of the data: “closed” sessions of the JCAE are printed on double-spaced pages with a typewriter, and the “open” sessions are printed in that particular high-density style of the Government Printing Office. In other words, a page is not necessarily a page. I did some very rough counting and found that on average, a line of text in both the “open” and “closed” hearings contained around 10 words. The “open” hearings, though, have around 53 lines per page, while the “closed” hearings have more like 25.
So using these rough numbers, here’s a graph of the word counts between “secret” and “open” hearings of the JCAE:
“Oh,” you say, “well they are not so secret after all!” Well, that’s still about 6.5 million “secret” utterances, which is quite a lot, if you ask me, even if the percentages aren’t quite as dramatic as before. And of course none of this discussion takes into account the content of the speech. I can thoroughly attest to the fact that much of those “open” utterances are pretty boring. And much if not most of them are utterances by people testifying in front of the JCAE. By contrast, the “secret” hearings contain lots of sessions where the JCAE are just talking amongst themselves.
Some of the trends are interesting:
- In the first three Congressional sessions (1947-1952), roughly 24%, 41%, and 74% (!!) of all of their utterances were classified.
- This latter number (74%) is of course somewhat stunning: they spoke nearly three times as many “secret” words on the record in 1951-1952 than they did “open” words. Skimming the titles of the “closed” hearings that year, they were mostly concerned with the status of the H-bomb project (which tested its first prototype in late 1952), and the expansion of the nuclear complex under AEC Commissioner Gordon Dean. More puzzling is why there weren’t more “open” hearings that session — it’s the all-time low in general. (It might be a database/indexing error, but I don’t see any reason to suspect that, other than it being anomalous.)
- In all of these visualizations, it’s striking how more “open” the JCAE gets after the “Atoms for Peace” years. The reasons for this are in part because huge swaths of the AEC program get declassified (including reactor technology in general) and the JCAE spends a lot of their time debating how best to promote a civilian nuclear power industry, as well as endless discussions of Civil Defense and other ostensibly “open” topics.
- It’s especially striking how the JCAE seems to essentially stop having classified sessions at all after 1962. This could be, again, some kind of indexing error, it could mean that they just stopped recording these sessions, or it could be that they never declassified those sessions and thus they don’t show up in Lexis Nexus. I haven’t looked into it, myself, but that’s in part because I’d never tried graphing it this way before.
Anyway, with all of that numbing your mind, be sure to show up here tomorrow for more exciting updates on my archival questing. It’s not quite Indiana Jones, but what is, these days?