Sir

In his review of two biographies of Carl Sagan, by William Poundstone and by Keay Davidson (Nature 401, 857 ; 1999), Christopher Chyba deals extensively with “Davidson's accusation that the young Sagan wilfully and illegally revealed classified information…” and further states: “This is a serious and specific legal allegation which Davidson does not substantiate”.

The classified project involved, Project A119, entitled A Study of Lunar Research Flights (SECRET), was conducted at Armour Research Foundation (ARF) while I was manager of physics research. I was also leader of the project, so Sagan reported directly to me. I, therefore, feel obliged to extend the historical record beyond the Davidson biography by offering some additional, first-hand comments.

A119 was one of a series of projects conducted at ARF under my direction from 1949 to 1962, all concerned with the global environmental effects of nuclear explosions and related phenomena. Some time before May 1958, we were asked by the US Air Force to add a small, fast-track project to investigate the visibility and effects of a hypothetical nuclear explosion on the Moon. I was told the Air Force was very interested in the possibility of a surprise demonstration explosion, with all its obvious implications for public relations and the Cold War.

Whether the project was motivated by a desire for the United States to impress the world (and the Soviet Union in particular), or by fear that the Soviet Union itself might try the stunt, I cannot say. It was emphasized, however, that the most sensitive aspect of the project was, as with many other Department of Defense projects, its very existence. Hence it was given a separate name and work on it was classified as secret. Our task was to assess what gross visible and other phenomena might be generated by the explosive release on the lunar surface of a given number of kilotonnes of energy.

We were also asked to try to determine what legitimate scientific data might be gathered from such an event: for example, about lunar chemistry. The cost to science of destroying the pristine lunar environment did not seem of concern to our sponsors — but it certainly was to us, as I made clear at the time.

In staffing A119, I realised that we needed expertise in planetary physics and asked Gerard P. Kuiper to act as our consultant. Kuiper agreed, and in time suggested that I hire a graduate student from Yerkes/University of Chicago called Carl Sagan, who needed a job. I gave Sagan the assignment of mathematically modelling the expansion of an exploding gas/dust cloud rarifying into the space around the Moon. This was preliminary to attempts to calculate the visibility of such a cloud from Earth. Sagan had difficulty with the problem and consulted Kuiper several times before I brought in additional help. Sagan soon suggested that he should try to see how a nuclear explosion might be used to detect organic molecules on the Moon. I agreed to a brief effort in that direction.

Nine monthly progress reports, all classified as secret and including all of Sagan's work, were issued by ARF to the Air Force Special Weapons Center under Project A119 from May 1958 to January 1959. According to Armour (now the Illinois Institute of Technology Research Institute) archives, they were all destroyed after October 1987.

I did not know until the biographies were published that Sagan had sent an "unclassified" (by whom?) manuscript about his work on A119 to any unauthorized person — let alone to five people, as Chyba remarks. Both ARF's and Sagan's obligations under the A119 contract, as Sagan's signed security agreement would have clearly informed him, required Air Force clearance of any such revelations. As his boss at the time, I would have had to take forward any such request, and Air Force permission would have been extremely unlikely in those long-ago, very tense times.

In his review, Chyba argues that classified documents can contain unclassified titles or subsections. Perhaps so, but that misses the point. Fortunately for the future of lunar science, a one- or two-horse race to detonate a nuclear explosion never occurred. But in my opinion Sagan breached security in March 1959 when he revealed the ARF's classified projects on “possible lunar nuclear detonations” in his application for a Miller fellowship.