The Vela Incident — Our Worst Ally

The Vela satellites were built for exactly one purpose: to catch anyone testing a nuclear weapon in the atmosphere in violation of the 1963 Partial Test Ban Treaty. They carried instruments called bhangmeters, tuned to a signature that is close to unmistakable. A nuclear airburst produces a sharp flash, then a brief dimming as the fireball is swallowed by its own shock-heated air, then a second and longer flash.

By September 1979 the Vela fleet had recorded that double flash forty-one times. Every single one was a nuclear detonation.

At dawn on September 22, 1979, satellite 6911 recorded the forty-second, over the South Atlantic near the Prince Edward Islands.

What Washington concluded in private

The White House was alerted immediately and the President was briefed. Three candidates were considered: the Soviet Union, South Africa, and Israel, with the strongest suspicion falling on Israeli-South African cooperation. That suspicion had a history. In 1977 the Carter administration had asked Menachem Begin directly whether Israel was helping South Africa build nuclear weapons, and got an answer that denied the specific charge while stepping around the broader one. Carter decided at the time that “we shouldn’t push [the issue] any more for now.”

Carter’s diary, published in 2010, tracks where his administration actually landed. On the night of the detection: “There was indication of a nuclear explosion in the region of South Africa, either South Africa, Israel using a ship at sea, or nothing.” Five months later, on February 27, 1980: “We have a growing belief among our scientists that the Israelis did indeed conduct a nuclear test explosion in the ocean near the southern end of Africa.”

The intelligence pointed the same way. A CIA-sponsored panel of scientists concluded the flash was probably a nuclear test. A later CIA assessment put the probability at “90% plus.” Richard Garwin, among the most eminent physicists of the era, wrote weeks afterward that he would “bet 2 to 1 in favor of the hypothesis” that a weapon had gone off.

Physical corroboration accumulated from instruments that had nothing to do with the satellite. The Arecibo Observatory in Puerto Rico registered an ionospheric disturbance consistent with a nuclear blast. U.S. Navy hydroacoustic sensors picked up a signal consistent with an explosion near the surface.

What Washington said in public

The administration convened a White House science panel under Jack Ruina of MIT. In May 1980 it produced an alternative: a small meteoroid had struck the satellite and broken into particles that caught the sunlight in a way that imitated a nuclear double flash. The panel hedged on whether that was what had actually happened, but concluded the signal was “probably not” nuclear.

That finding contradicted the CIA, contradicted a number of the scientists who had looked at the data, and did not account for the ionospheric and hydroacoustic evidence.

A senior United States intelligence official called the White House study a “whitewash” influenced by “political considerations.”

Why the answer mattered so much

The political stakes were not abstract, and they were not only about embarrassment.

A confirmed Israeli atmospheric test would have been a violation of the Partial Test Ban Treaty by an American ally, carried out with an apartheid regime then under international sanction, in an American election year. It would also have collided with American law. As the Symington and Glenn case documents, statutory triggers tied to nuclear proliferation carried automatic consequences for foreign aid, and those triggers depend on the executive branch making a formal determination.

A finding of “probably not nuclear” meant no determination had to be made. The incentive to reach that finding was concrete, statutory, and immediate.

Where the evidence has gone since

The scientific record has moved steadily in one direction, and it is not the White House’s.

Peer-reviewed analysis by Lars-Erik De Geer and Christopher Wright, published in Science & Global Security in 2017 and 2018, examined the radionuclide and hydroacoustic evidence and concluded the event was a nuclear explosion. Avner Cohen, the leading historian of Israel’s nuclear program, has written that a scientific and historical consensus formed that the Vela event was a test and “had to be Israeli.”

South Africa’s own conduct added to it. After apartheid ended, South Africa dismantled its arsenal and acknowledged having built six weapons, while stating that in 1979 it did not have the capability to test a device of the kind the Vela signal indicated. That points away from South Africa acting alone.

None of this is a confession. Israel has never acknowledged the event, consistent with the ambiguity policy set out in the Dimona case. No American administration has ever attributed it to Israel, and the official position, established by the Ruina Panel and never revised, remains that the flash was probably not a nuclear explosion.

Why it matters

In every other nuclear case in this archive, the United States avoids a confrontation by not looking closely or by declining to make a formal finding. Vela is the one where it looked.

The detection system built for precisely this purpose registered precisely this signal. The CIA put it at ninety percent plus. Independent instruments on two continents corroborated it. The President wrote in his own diary that his scientists increasingly believed Israel had done it.

And then the White House told the public that a satellite had probably been hit by a rock.

That is the difference between declining to know something and asserting the opposite of what you know. One of the government’s own intelligence officials named it at the time. Forty-six years and a great deal of declassified material later, the scientific consensus has arrived where Carter privately was in February 1980, and the official American position has not moved at all.

Sources

National Security Archive (George Washington University), “The Vela Incident: South Atlantic Mystery Flash in September 1979,” December 6, 2016, and “The Vela Flash: Forty Years Ago,” September 22, 2019 — declassified CIA assessments and the internal “whitewash” criticism
President Jimmy Carter, diary entries of September 22, 1979 and February 27, 1980 (published 2010)
Lars-Erik De Geer and Christopher M. Wright, “The 22 September 1979 Vela Incident,” Science & Global Security 25 (2017) and 26 (2018) — peer-reviewed radionuclide and hydroacoustic analysis
Bulletin of the Atomic Scientists, “Double Flash: Forty Years Ago, the Carter Administration Covered Up a Presumed Israeli Nuclear Test,” 2019
Foreign Policy, “Blast From the Past,” September 22, 2019
Wilson Center, “Revisiting the 1979 VELA Mystery: A Report on a Critical Oral History Conference”
Avner Cohen, on the scholarly consensus that the event was a nuclear test and had to be Israeli

Note on source tier: The detection, the CIA’s “90% plus” and “probably a nuclear test” assessments, the internal “whitewash” criticism, and Carter’s private belief are all established in declassified U.S. government documents published by the National Security Archive and in Carter’s own published diary. The attribution to Israel is a different order of claim: it is the assessment of the CIA at the time, of named scientists, of subsequent peer-reviewed analysis and of leading nuclear historians, but it has never been officially confirmed by the United States or acknowledged by Israel, and the official American position remains that the event was probably not nuclear. That gap between the evidence and the official position is the documented subject of this article, and the attribution itself is presented as a strongly supported conclusion rather than an established fact.