John Piña Craven was the chief scientist of the Deep Submergence Systems Project, and Sealab III, since the autumn of 1965, had been his. He told the story of 17 February 1969 in one chapter of his memoir, The Silent War, published in 2001. It is the only account by a senior officer of that day, and the only one in which the word sabotage is used by someone who had the standing to use it. This piece follows it, with what the divers and Walter Mazzone have said against it.
The flooded capsule
Craven’s suspicion began before the deployment. In the weeks after the habitat was moored off San Clemente Island he would stop at the site on his way back to Washington from Hawaii. The two personnel transfer capsules were being soak-tested, closed up and left in the sea for a day or more to see whether they leaked. On one visit he arrived as a capsule was being recovered, flooded, with every piece of electrical equipment inside it destroyed. “Examination of the capsule’s valves and hatches indicated that they were in perfect order and a soak test proved it,” he wrote. A senior officer of his staff investigated and “concluded that a valve had been positioned incorrectly and the culprit had covered it up as soon as the PTC had broken clear of the surface or while still submerged.” There was no time for a repair. The project went on with one capsule.
The divers’ recollection of the same capsule is of a machine patched back into service with its scrubber blowing unbaffled cold helium at the men inside and its heater dead. Keith Moore, who had watched a repair crew at Mare Island complaining that they had not been given time to do the job properly, called it “a disaster.” Fernando Lugo, who was inside the second capsule when it dropped from the gantry on 12 February and who helped fight the fire on the gantry an hour later, “does not suspect sabotage.” Craven does not mention the drop or the fire.
The lift
He flew out to the island by seaplane “some two weeks later,” on the morning Cannon died, and was met at the landing with the news. The senior officer in his chain of command had been ordered back to Washington, which left Craven the senior officer present, with Commander Jackson Tomsky in command of the operation and a habitat on the bottom at 600 feet that was losing its helium and oxygen and would soon flood. There were volunteers to go down and blow its ballast, among them Scott Carpenter and Philippe Cousteau. The doctors were against sending anyone, “even though the cause of Cannon’s death appeared to be known,” and Tomsky agreed with them. He “did not believe that the fatality was in any way connected to the leak, but he recognized that the recovery operation was risky, and that everyone was under emotional stress.”
The arithmetic was short. The helium-oxygen supply would last a few hours. After that the habitat could be kept up on compressed air, but air under nineteen atmospheres is rich enough in oxygen to be set off by a spark, “from, for example, a short generated in a battery that became flooded with seawater,” and the air itself was limited. A submarine was sent for with more, and would be many hours. “The salvage would have to be accomplished inside about ten hours or not at all.”
The floating crane that had put the habitat down had gone back to Long Beach, nearly a hundred miles away. What the barges had was a winch, out of position, with a length of cable, and the habitat’s own lowering cable tied off to a buoy. Craven and Captain Gus Eble of the project staff, who had won the Silver Star rescuing men from the submarine Darter, worked out that two sheaves would carry the winch wire horizontally out over the habitat and vertically down to a splice with the habitat’s cable. The welders made them. The maker’s rating for the winch cable was too low for the load. Eble and Craven calculated separately what it would really take, agreed, estimated where it would part if it did and who it would hit, and Tomsky accepted the risk. The Westinghouse submersible Deepstar 4000, which had been used to inspect the habitat, would watch the lift from below. If the dynamometer on the winch reached a set load before the habitat broke the mud, they would stop. If it came free they would bring it to a hundred feet and send divers down to blow the ballast. Tomsky, before any of it began, took two hours’ sleep. “This ability to act on a self-evaluation during the heat of an operation was one of his finest qualities.”
The helium ran out and they switched to air. The submarine arrived “just before the air supply was about to run out,” and Tomsky went ahead. The strain built; the dynamometer came near its limit; and then somebody noticed that the barge’s freeboard, the height of its deck above the water, was shrinking, because the cable was pulling the barge down. “A few more inches and the water would be pouring over the deck, flooding the barge’s compartments.” Tomsky was about to stop when the cry came up from the Deepstar: “She’s broken free!” The habitat came to a hundred feet, was blown, and surfaced. It was midnight. The men on the barge, Craven wrote, were “choked with the involuntary sob that accompanies the end of a battle won but lives lost.”
The valve
While that was going on, eight men were in the two deck decompression chambers, three in one and five in the other, beginning a week of coming up. Mazzone was watching over them. Craven’s account is this: Mazzone “suddenly noticed that the mixture going into their breathing masks in the chamber was overly rich in oxygen. He found the problem. A valve from an oxygen tank was open when it should have been closed. He closed it. A short time later he found it open again. He closed it again and tagged it with a wire placed in the valve that would break if the valve were opened again. An hour later the tag was broken. It was closed and retagged and Tomsky assigned a guard to watch over the decompression chamber for the remainder of the weeklong decompression.” Tomsky told nobody on the salvage side until it was over. Too much oxygen under pressure is itself a poison.
After the lift Craven and Tomsky sat up and went over it. “We agreed that there had to have been multiple acts of sabotage. This was the only plausible explanation of the failure of the PTC, the absence of baralime in Cannon’s canister, and the persistent opening of the oxygen valve to the breathing masks of the men in the decompression chamber.”
Washington
In Washington Craven ran his own inquiry and consulted the project’s psychiatrists. “One of them identified a suspect, someone possibly afflicted with psychiatric problems that could have given rise to this murderous behavior. We agreed, however, that it would be a violation of that person’s privacy rights to implicate him in specific acts without due process. So all of the evidence was turned over to the Office of Naval Intelligence.”
The board of investigation in San Diego named Tomsky and Chief Paul Wells as parties; George Bond asked to be named too, “apparently to assuage feelings of guilt,” and the board found no fault in him. “Naval intelligence reported that there was insufficient evidence for concluding that sabotage had played any role in the SeaLab disaster.” Wells, “a thirty-year man with an unblemished record and one of the most upstanding chiefs in the naval establishment, took sole responsibility,” though, as Craven notes, the documentation recorded that he had filled the canister. Another court of inquiry was sitting at the same time, into the surrender of the Pueblo. “There were many who felt that in the SeaLab disaster a finding of sabotage would be an additional public relations fiasco for the Navy. In any event, the incident was ‘resolved’ by a letter of reprimand issued to Commander Tomsky and Chief Wells. Both men resigned from the Navy.” Craven urged Tomsky to demand a court-martial and clear himself; Tomsky “preferred not to further discomfit the Navy, and he regarded the outcome as no more than a battle injury incurred in his service to his country.” Mazzone, who received a reprimand too, said in 2000 that he had been told at the hearing by the officer running it, “I don’t know why you are here but I need you,” to settle what the valves did.
What can be said
Craven’s chapter is a memoir written thirty years on, and it differs from the other accounts in places: he has two aquanauts going out and four more staying in the capsule, where every other account has four men in it and two going out; and his cause for the leaks, valve packings installed the wrong way round, is not the one the divers and the project’s engineers give, who describe gas coming from the cable penetrators. Neither point touches the valve, which he had from Tomsky, or the finding by Naval Intelligence, which is the only official word on sabotage that exists.
Three explanations of the empty canister have been offered since 1969, and they are not compatible. The board’s was that it had been left unfilled, by Wells or, as was also suggested, by an unnamed helper. Craven’s was that somebody emptied it. Kevin Hardy’s, in 2024, is that it was never empty when it went down, that it washed out in the half hour it hung in the sea outside the capsule, and that Cannon was killed by the current leaking from the habitat’s main cable; that argument is in Four rigs. Nobody was ever charged with anything, and Craven did not name the man the psychiatrist suspected. The one point on which Craven, Bond and every diver who has spoken agree is the one the board decided against: that Paul Wells did not leave a canister empty.
Sources: John Piña Craven, The Silent War: The Cold War Battle Beneath the Sea, Simon and Schuster, 2001, chapter 9, “Humans as Marine Mammals”; the interview with Walter Mazzone by Jan Herman, 18 March 2000; Kevin Hardy, “SEALAB III (1969): The Divers’ Story,” Journal of Diving History, 2024, for Moore and Lugo. All are linked from the entry’s sources.