The Second Attack

2005

The answer first. NSA historian Robert Hanyok wrote that the second Gulf of Tonkin attack, on 4 August 1964, did not happen. He wrote in the agency’s classified journal; the NSA itself had “officially maintained” that it did. In 2003 the U.S. invaded Iraq after officials said it held banned weapons, and the Iraq Survey Group did not find the stockpiles. What each official knew, and said, differs from one name to the next. The records, not the story, sort mistake from lie.

2 Aug 1964

A patrol, a raid, a first shot. The destroyer Maddox began a Desoto signals patrol on 31 July 1964; the night before, South Vietnamese boats had raided two islands under OPLAN 34A, a U.S.-run program. McNamara told senators the Maddox was on “a routine patrol” and opened fire “after three warning shots.” The Senate staff, reading the ship’s log in 1968, found “no indication” of warning shots and that the Maddox “fired the first shot.” Three North Vietnamese boats attacked it on 2 August; Hanoi admitted that.

6 Aug 1964

3 and 6 August. On the 3 August tape McNamara tells Johnson the raids “undoubtedly led them to connect the two events.” Johnson answers: “Well say that to Dirksen.” No record of that day’s briefings of Congress has been found. On 6 August, after Senator Morse asked about the raids, McNamara said: “I say this flatly.” In 1968 the Senate staff concluded he “misled the Committee” on whether the Navy knew of them, while saying it did not suggest he “consciously misled” it. He told senators then that the Maddox “did know what 34A was.”

4 Aug 1964

4 August, morning. At 9:43 a.m., before any attack was reported, McNamara told Johnson on the tape that the ship was “allegedly uh to be attacked tonight.” He recommended retaliation “after a second attack on our ships”; Johnson wanted to “pull one of these things” on “one of their bridges or something.” Hanyok found the alert rested on a mistake: the Marine signals unit at Phu Bai read messages about damaged boats being towed for repair as preparations to attack. NSA headquarters did not correct it.

4 Aug 1964

4 August, night. In darkness and rough weather the Maddox and the Turner Joy reported torpedoes and fired at radar returns. James Stockdale, a navy pilot overhead, had “the best seat in the house” and saw no boats. Hanyok writes that the officers came to realize the sonar torpedo sounds were the ships’ own rudders reflecting the turbulence of their propellers. Captain Herrick, the task force commander, cabled within the hour: “ENTIRE ACTION LEAVES MANY DOUBTS EXCEPT FOR APPARENT ATTEMPTED AMBUSH AT BEGINNING.”

4 Aug 1964

Doubt reaches Washington. Herrick’s cable of 1:27 p.m. said many reported contacts and torpedoes “appear doubtful.” The records disagree on when the Pentagon had it: about 3 p.m. (the administration’s chronology), “around 1400” (Hanyok), 1:27 itself (Prados). Johnson had agreed to a strike over lunch, which ended at 2:50. McNamara said in 1968 that about 2:45 Herrick cabled that the ambush was “bona fide,” and that Sharp then called twice to say it had happened. Gore replied that Admiral Moorer was still asking for confirmation at 7:06. The National Archives says McNamara “did not report” the doubts to Johnson; Hanyok says Johnson required verification because of them.

4 Aug 1964

Five reasons. This is the White House’s own chronology of 28 August, on the 4:47 p.m. meeting. They concluded an attack had happened on five points; two rested on one NSA translation that two boats had been “sacrificed.” Johnson’s memoir has the North Vietnamese skipper reporting “sacrificed two comrades”; why NSA wrote “boats” “is not clear,” says Hanyok. The strike order went out at 5:19. At 6:15 USIA director Rowan asked whether anyone knew “for a fact”; McNamara: “We will know definitely in the morning.” At 6:45 Johnson told congressional leaders that “nine torpedoes were shot at our destroyers.”

7 Aug 1964

88 to 2. At the 6:15 meeting Rusk read out a draft resolution calling the attacks “unprovoked”; Hanyok says a resolution was drafted in June and shelved; the Pentagon Papers say one had been under consideration since May. At the same meeting CIA Director McCone told Johnson the North Vietnamese were “reacting defensively to our attacks on their off-shore islands.” The Pentagon study says the second incident was “promptly exploited,” yet shows “little indication of a deliberate provocation.” Senate 88 to 2, House 416 to 0, on 7 August. Morse and Gruening voted no.

20 Feb 1968

1968. After three months the Senate Foreign Relations staff concluded there was “considerable evidence” the patrol aimed “to provoke the North Vietnamese,” and “ample grounds to question whether North Vietnamese boats were there at all.” A captured North Vietnamese commander “strongly denies” any attack. Under Secretary Nitze showed Fulbright a cable called “conclusive”; the staff never saw it. McNamara called talk of a plot “monstrous”; Fulbright answered that “nobody has been suggesting a conspiracy.” McNamara’s memoir, Hanyok notes, still found the evidence for an attack convincing.

2005

What the NSA historian found. Hanyok read 122 intercept reports; NSA’s summaries had used fifteen. He found the NSA staff, who “believed that the attack happened,” kept out nearly 90 percent of them: “a conscious effort,” he wrote. He found no pressure from above, which “simply cannot be supported,” and said the “how” and “why” “remain unknown.” Of Johnson he wrote that why he ordered the strike so fast “is not totally clear,” and that “it also has been suggested” he used the warning “as an excuse” to get a resolution passed.

26 Aug 2002

Iraq, August 2002. Cheney spoke in Nashville on 26 August. In 2008 the Senate Intelligence Committee found that “many senior policymaker statements in early and mid-2002 claimed that there was no doubt,” while most intelligence reports “expressed room for doubt about whether production was ongoing.” It found statements on possession “generally substantiated,” but those on ongoing production “reflected a higher level of certainty than the intelligence judgments themselves.” In the same speech Cheney said analysts had once put Saddam five or ten years from a bomb, and “he had been much closer than that.”

9 Jul 2004

The Estimate. After senators asked Tenet for it in September 2002, the intelligence community wrote its classified estimate in October. In 2004 the Senate committee found its key judgments “overstated, or were not supported by” the reporting, and “no evidence” of political pressure behind that. Three senators said the report “fails to fully explain the environment of intense pressure”; the CIA ombudsman said the “hammering” was the worst of his 32 years, over terrorism links. The Energy Department agreed reconstitution was under way but doubted the tubes; State’s intelligence bureau rejected the case, and the committee agreed with State.

Jan 2003

Curveball. Almost all information on Iraq’s supposed mobile germ-weapons labs came from one source, code-named Curveball, handled by a foreign service that gave U.S. analysts no direct access. Pentagon collectors “did nothing to validate” him; CIA officers doubted him from 2000; in April 2002 a foreign service wrote that his behavior was “typical of individuals we would normally assess as fabricators,” though it was also “inclined to believe” much of his reporting. The 2005 commission, “not authorized” to examine how policymakers used intelligence, found “poor tradecraft and poor management,” and said it was “hard to deny” analysts worked where skepticism was not encouraged.

28 Jan 2003

The President’s words. In October 2002 Tenet told the Deputy National Security Advisor that the “President should not be a fact witness” on uranium from Africa; the line came out of the Cincinnati speech. On 28 January 2003 a version went into the State of the Union, credited to Britain. The Senate found no CIA official told the White House to cut it, and that Tenet “should have taken the time to read” the speech. In 2005 Bush said: “much of the intelligence turned out to be wrong. As President, I’m responsible for the decision to go into Iraq.” The record read here does not show Bush was told of Tenet’s call.

5 Feb 2003

5 February 2003. Powell told the Security Council: “every statement I make today is backed up by sources, solid sources.” The 2004 Senate report found much of the CIA material for the speech “overstated, misleading, or incorrect,” and that the CIA “should have alerted” him about the biological sources. A CIA deputy’s e-mail the day before said “this war’s going to happen regardless of what Curve Ball said or didn’t say.” A division chief recalled telling Tenet, “something to the effect of,” that the reporting “has problems”; Tenet says Curveball was not discussed. The Senate found no evidence Tenet was told of the doubts.

Sep 2005

“Painful.” In September 2005 Powell told ABC’s Barbara Walters the speech was a “blot” on his record. He said Tenet “believed what he was giving to me was accurate,” but that some in the intelligence community “knew at the time that some of those sources were not good” and “didn’t speak up.” He said he had never seen evidence linking Baghdad to September 11. He said he supported Bush’s decision for war, because the regime was “in violation of all those U.N. resolutions.”

23 Jul 2002

Downing Street, July 2002. The Sunday Times printed this minute in 2005. The Iraq Inquiry names “C” as Sir Richard Dearlove, head of MI6. He told it the remark was about attempts “to join up terrorism and Iraq,” and accepted he “might well have used the word ‘fitted’.” Rycroft, who wrote the minute, checked his notes and took no action on Dearlove’s challenge. In 2005 Blair said “the facts were not being fixed in any shape or form”; Bush said that the claim “we had made up our mind” to use force was “nothing farther from the truth.”

30 Sep 2004

Nothing there. The Iraq Survey Group searched from June 2003. Its report called the prewar view “misapprehensions of weapons, which were not there.” The Senate summary says it found no nuclear restart, no biological production after 1996, and no chemical caches, though some pre-1991 chemical munitions turned up. The 2004 Senate report records that Iraq “had repeatedly lied” about earlier programs, and that UN inspectors found no evidence of germ-weapons work at a suspect site while analysts “discounted the UN’s findings.” The ISG reported Saddam wanted to end sanctions “while preserving the capability to reconstitute” his weapons.

Apr 2008

2008: two readings. The report found statements that Iraq and al-Qa’ida were partners “not substantiated,” and statements that Saddam was prepared to give weapons to terrorists “contradicted by available intelligence information.” Chairman Rockefeller called the claim of an operational partnership “fundamentally misleading,” and wrote that for unsubstantiated statements “the Committee renders no judgment as to why.” Vice Chairman Bond and the minority called it “a partisan exercise”: policymakers’ statements “were substantiated,” and “it was the intelligence that was faulty.” In 2005 Bush said many intelligence agencies around the world had judged Saddam had such weapons.

10 Oct 2002

77 to 23. On 10 October 2002 the Senate voted 77 to 23 for the Iraq resolution and the House 296 to 133. Senator Robert Byrd, who had voted on the Tonkin Gulf Resolution, said the administration’s claims for it “were simply not true, and history is repeating itself.” Senator John McCain replied that blaming that resolution for the Vietnam tragedy was “simplistic,” and that it “was simply window dressing.” One difference on the record: this time senators asked for the intelligence estimate before the vote.

2011

The dead. For Vietnam, the National Archives file holds 58,220 U.S. military fatal casualty records, from 1956; its counts are “not official statistics.” In 1995 Hanoi’s news agency said 1.1 million Communist fighters and nearly 2 million civilians died, and 223,748 South Vietnamese soldiers. For Iraq: the WHO-backed survey counted 151,000 violent deaths, March 2003 to June 2006 (104,000 to 223,000); a 2013 survey estimated 405,000 excess deaths to June 2011 (48,000 to 751,000); Iraq Body Count documents 200,065 to 226,628 civilian deaths. Rockefeller counted over 4,000 Americans by 2008. The counters differ in scope and are not added here.

2008

What each record shows. Hanyok found a deliberate effort inside NSA to keep out contrary intercepts, believing the attack real, but not who ordered it. The Senate staff found McNamara “misled the Committee” on the Navy and the raids, not that he knew. Nobody read says what Johnson was told of Herrick’s doubts. On Iraq, the 2004 and 2005 bodies found analysts’ errors and bad management; the 2005 commission could not ask how officials used the intelligence; the 2008 majority found some statements “contradicted,” and no judgment as to why. The line between mistake and lie is still argued.