The second Gulf of Tonkin attack: an NSA historian's case that it never happened

Two dates that must stay separate

The Gulf of Tonkin controversy concerns two reported encounters, on 2 and 4 August 1964. Robert Hanyok's released NSA history argues that the second attack did not occur. That conclusion must not be expanded into a claim that there was no separate encounter on 2 August.

The distinction matters because the events became compressed into a political justification. The Gulf of Tonkin Resolution spoke of repeated attacks and authorized presidential action. A later historical inquiry must recover the differences that a decisive public account can flatten.

Intelligence does not interpret itself

Hanyok examined signals intelligence and its handling around the reported second attack. His historical argument concerns how information was selected and interpreted, not simply whether a message existed.

A received signal, a translation, an analytic judgment and a report to decision-makers are different stages. An error or assumption at one stage can acquire authority when it is repeated at the next. Seeing several documents repeat a conclusion does not necessarily mean that several independent observations support it.

The consequence outlasted the uncertainty

NARA's account of the resolution describes the disputed second incident and the later conclusion that it had not happened. The resolution itself remains available to read as a primary document. Its asserted factual foundation and its grant of authority can therefore be examined together.

The wording of an authorization can be certain even when the underlying reports were not. That is a general problem in fast-moving decisions: an unresolved operational picture may become the premise for an action whose consequences cannot be reversed simply by correcting the record afterward.

The resolution is evidence of what was asserted

The resolution's opening language describes deliberate, repeated attacks. As a source, it establishes the premise Congress put into its authorization. It does not independently verify the operational reports behind that premise. This is a useful distinction whenever a legal or political text is cited as proof of an event: the text may reliably record an official assertion while leaving the assertion's accuracy open to examination. The later intelligence history and the contemporary resolution consequently answer different questions. One concerns what the evidence supported; the other concerns what government formally said and authorized. Reading both avoids asking either document to do a job for which it was not created.

Read the history beside the resolution

Hanyok's article is a retrospective NSA history published in Cryptologic Quarterly, not a contemporaneous ship's log. Its institutional origin and later release make it significant, but its arguments still depend on the evidence and interpretation it presents.

Read it alongside the text of the resolution. One shows a later attempt to reconstruct what the intelligence supported; the other shows the public language under which Congress acted. Keeping those documents distinct produces a more precise account than saying only that Tonkin was a lie. It identifies which incident was disputed, what kind of evidence was reexamined and why the difference between uncertainty and asserted certainty mattered.