By 1984 the waters had been sufficiently muddied as to the legitimacy of the evidence pointing to a second attack, and with it, incidentally, the legitimacy of the Tonkin Gulf Resolution. The Johnson administration’s case had hung on the assertions of Commander Herrick that the attack took place, and the intercepted North Vietnamese cables apparently ordering warships into action the night of August 4th. But now, thanks to the US News investigation,  Herrick’s assertion had become suspect, due in large part to his own investigation in which he proved that his ship’s sonar had not picked up torpedo screws, but instead Maddox’s own movements. Parenthetically, Herrick’s revelation vindicated suspicions raised by John Galloway over a decade earlier that three North Vietnamese patrol boats, only one of whom carried torpedoes could have launched twenty six of them at the Maddox and the Turner Joy that night.  Even the intercepted cables had raised eyebrows, with experts wondering if they’d been correctly interpreted, or if they were part of an intentional misreading of them in order to justify wider US action in Vietnam.  Nearly twenty years after the fact, journalists and historians, working separately had raised substantive and legitimate questions about the roles played by the US Navy and the Johnson Administration in the Tonkin Gulf.  But in order to put a finer point on the matter, more digging would be required.

Naval historian Edward Marolda no longer believes there was a second attack in the Tonkin Gulf.  But in 1986, when Marolda and Oscar Fitzgerald completed work on Volume 2 of The United States and the Vietnam Conflict: From Military Assistance to Combat, 1959-1965, they lacked access to the detailed information which would ultimately cause Marolda to change his mind.  Nonetheless, using the information available at the time, the two chapters on the Tonkin incident are both methodical and comprehensive. They lay out the Navy’s reasons for its presence in the Tonkin Gulf, how it believed it was supporting US interests in the region, and what went through the minds of on scene commanders as well as their superiors as they tried to grapple with the need for real time information on events which were happening a world away.  It is in this area that perhaps Marolda and Fitzgerald do the greater good.  By exhaustively examining cable traffic, and in particular the communications between John Herrick, the task force commander on scene (even though Maddox was commanded by Herbert Ogier and Turner Joy by Robert Barnhart), and his superiors, including Admiral Roy Johnson, Commander, Seventh Fleet, his superior, Adm. Thomas Moorer, Commander in Chief, U.S. Pacific Fleet, and ultimately Admiral Ulysses S. Grant Sharp, readers get a glimpse at how, for better or worse, decisions were made which had long term consequences for the United States. The work outlines the start of the DESOTO patrols and how they gradually came to be included in the Vietnam effort; one in December, 1962, and six more in 1963.  In 1964, Paul Harkins, Commander US Military Assistance Command Vietnam asked the Navy to expand DESOTO’s intelligence gathering operations in order to help the OPLAN 34A teams attack North Vietnam’s coastal  and island defenses. William Westmoreland, Harkins’ replacement sustained the request, but doubted the effectiveness of seaborne intelligence collection and, according to Marolda, suggested the August patrol be cancelled. Admiral Moorer contended the patrols provided important training for naval personnel, helped the overall intelligence effort in the region and was proof the US was no paper tiger when it came to exercising its right to freedom of the seas.  He convinced the Joint Chiefs of Staff, which approved the patrol (405).

As Herrick’s task force prepared to enter the Tonkin Gulf it passed the boats carrying 34A personnel returning from their attacks on North Vietnamese positions.  On August 1, Marolda tells us, US intelligence sources noted what was believed to have been the first reaction by North Vietnam to the presence of DESOTO patrols operating this soon after the commandos left. A day later, Marolda reports, Herrick was warned he could be attacked. He immediately requested that the mission be scrubbed, that if the intelligence was correct, then continuing was an unacceptable risk. (414) He was overruled by both Admirals Moorer and Sharp. One day later, on August 2 Maddox and Turner Joy  were attacked by a collection of North Vietnamese gun and torpedo boats, which despite all the firing and maneuvering only managed  to register a small hit on Maddox and paid a high price for it.  The military engagement was then followed by a series of cabled engagements between Herrick and Moorer, who ordered the patrol to proceed as a show of US resolve to assert its right of freedom of the seas. Herrick, says Marolda, challenged the order when he cabled back that ‘US ships in the Gulf of Tonkin can no longer assume that they will be considered neutrals exercising the right of free transit. They will be treated as belligerents from the first detection (420).  Marolda also reported the emergence of a sense in Washington’s intelligence community that Hanoi considered the 34A attacks and the DESOTO to be a joint operation. In fact, he said the North Vietnamese filed a complaint with the International Control Commission that the “The US and South Vietnamese administrations sent two naval vessels to shell Hon Ngu and Hon Me islands”. The conclusion was drawn, he said, that the attack on Maddox could have been provoked by ‘enemy incursions into the Gulf of Tonkin’.   Herrick lost the verbal engagement when President Lyndon Johnson intervened and ordered the patrol be resumed.  Herrick complied.  At the same time orders were given to resume 34A operations, and North Vietnamese installations were attacked on the night of August (Marolda 424).  One day later Maddox and Turner Joy would be drawn into an event which ultimately led to an expanded military role for the United States. It’s during the interim as well as the after action period in which Marolda reveals the mixed messages, hidden agendas and possibly genuine concerns about actions and consequences by officers and officials which drove the Tonkin Gulf debate.

As Herrick and his taskforce resumed the patrol, Marolda says Herrick fired off another cable to the effect that evaluation of intelligence from various sources indicated DRV considered the patrol directly linked with 34A operations.  The DRV considered US ships present as enemies because of these operations and have already indicated a willingness to treat us in that category.  After the events on the night of August 4, Herrick again cabled Moorer hoping to be a voice of moderation and restraint.  “The review of action makes many reported contacts and torpedoes fired doubtful”, he said.  “Freak weather effects on radar and overeager sonar men may have accounted for many reports.  No actual visual sightings by Maddox….Suggest complete evaluation before any further action taken” (440).  That was followed by another communiqué in which Herrick concluded “the entire action leaves many doubts except for apparent ambush at beginning”.  That same message, however, included reports culled from the captain of Turner Joy, who said his ship had been fired upon and that at least one of the attacking craft had been hit by her gunfire. While consensus by on scene commanders was still gelling, Admiral Sharp reported that based on separate intelligence sources there was no mistaking the enemy’s hostile intentions.  These reports, say Marolda, included North Vietnamese accounts of aircraft falling into the sea and damage to an American vessel.  Yet another intelligence source indicated that the North Vietnamese had lost two vessels in action.  The snippets of independent information were coalescing and were pointing to the need for retaliatory action.  All that was needed was consensus by on scene commanders, and that, according to Marolda came as preparations for the retaliatory raids –to be dubbed Pierce Arrow-were underway.  Captain Barnhart aboard Turner Joy said he was attacked by two PT boats; it took Captain Herbert Ogier until August 6 before he stated he believed at the time that Maddox was under attack.  And Commander Herrick finally fell into line when he stated that “certainly a PT boat action did take place.” That was enough for the Commander of the Seventh Fleet, the Commander in Chief of the US Pacific Fleet, and the Commander in Chief of the Pacific. President Johnson and his advisers in Washington held the same view, and the US went to war.

At the time of its publication in 1996, Edwin Moise’s Tonkin Gulf and the Escalation of the Vietnam War was hailed as the most thorough study of its kind. Its size and level of detail can discourage a reader.  But its real potential for discouragement would be to those readers who hoped proof of the second Tonkin Gulf attack might be revealed.  Using the first five chapters to explain the role of the Navy and the Desoto patrols in the Gulf of Tonkin, Moise devotes three chapters to a reconstruction of the events of August 2 and August 4. Burnishing that chronology with weather reports, cable transcripts, navigation charts, after action reports, interviews of crew members, pilots, intelligence officials and even data mined from interrogations of North Vietnamese prisoners of war,  sources that the Johnson administration used to make the case for a second attack, Moise makes an alternative case; that the August 4 battle was fictional. However, he suggests, as did Anthony Austin over twenty years ago, that didn’t stop the US Navy from passing it off as fact, and portraying it as such to the Johnson administration.  With regard to the weather, Moise confirmed what was already known; that strange things happened to a ship’s radar and sonar operations in the Gulf of Tonkin.  But his interviews of crewmen on the Maddox, in particular Ensign Richard Corsette, who commanded a forward battery was especially telling.  “I know the way our radar was acting”, Corsette asserted, “My firm belief was that everything I locked onto was weather.” (109) Moise returns frequently to the weather issue as it helped set the stage for a series of confusing events on a confusing night.  Information gleaned from sonar man third class David Mallow also casts the events of August 4 in a suspicious light.  Returning to his station as general quarters was sounded, Mallow began reporting noise spokes to the ship’s Combat Information Center.  He did not, Moise writes, give any interpretation as to the source of those noise spokes.   Such an interpretation, or lack thereof, said Moise, was confirmed by two of Mallow’s superiors, who concluded the decision to pronounce the sound as torpedoes was made higher up the chain of command (126).  The ‘higher up’ in this case, asserted Moise, was Commander Herrick.  Herrick reported that when the news from the sonar room was relayed to the bridge that noises had been detected, he said they sounded to him like torpedoes (126).  In time, said Moise, Herrick came to realize that what he was actually hearing were the sounds of his own ship’s propellers.  But at the moment, careful reflection was a luxury no one, least of all John Herrick could afford.

Moise uses chapter six of his work to relate the confusing sequence of events which played out over two hours of firing and maneuvering by the Maddox and the Turner Joy.  Firing at least 300 rounds, mostly by Turner Joy, the two ships, several miles apart would take aim at a target in the darkness, open fire, maneuver to acquire another target, fire again, and then maneuver even more to evade a host of torpedoes supposedly launched at them.  Moise’s debriefing of the crew continued to uncover their own doubts as to just what they were firing at.  Patrick Park, a gun director on the Maddox claimed that the only target he was sure of on his fire control radar that night was the Turner Joy.  Moise wrote that Park told him “there couldn’t have been a canoe out there” (135).  But the Turner Joy seemed to have had no misgivings about what she was shooting at.  In the early stages of the engagement Moise reported Turner Joy’s detection system, more modern and more automatic than Maddox’s showed contact after contact, which Turner Joy fired at, claiming to have sunk two ships.  Douglas Smith, at the time an ensign assigned to Turner Joy as a gunnery liaison officer told Moise he was sure that what he was shooting at weren’t phantoms (135).  But not everyone was so sure. Robert Barnhart, commander of the Turner Joy was growing suspicious about the authenticity of the attackers.  According to Moise, Barnhart reached a moment of truth when sonar reported an enemy contact so close that a torpedo could not have missed his ship.  Rather than take evasive action, Barnhart ordered the Turner Joy to maintain course.

There was no torpedo attack and from that point on Barnhart’s confidence about the attack took a serious hit.  It had already stretched the bounds of plausibility aboard Maddox.  Herbert Ogier, the ship’s commander conducted a quick assessment, which Moise covered on page 140; both destroyers reportedly dodged 26 torpedoes detected by the sonarman on the Maddox.  Sonar on the Turner Joy could detect none of the torpedoes fired at the two vessels, while the radar on the Maddox couldn’t locate the ships allegedly firing the torpedoes.  Eventually, Ogier realized that the number of noise spokes reported by sonar had become ridiculous, and that what had been interpreted as torpedo noises had to be something else.  He ordered an end to evasive action, and the incident came to an end (140).    In conferring with task force commander Herrick, the two concluded that the Maddox’s own propeller noise was probably being misinterpreted as torpedoes.  Moise’s work also utilized evidence already taken from aircraft carrier pilots, who were directed by the Maddox and Turner Joy to locations where the destroyer’s radar said the North Vietnamese boats were.  No boats were sighted by any of the pilots, and the film from one of the plane’s photo-reconnaissance system confirmed it.  What the pilots could also confirm, said Moise, was bedlam and confusion coming from the radio communications aboard the Maddox and Turner Joy.  Orders were given then countermanded.  Bearings and vector instructions were frequently interrupted by two or three other voices announcing a torpedo bearing.  Things became so confusing, said Moise that someone from either the Maddox or Turner Joy’s Command Information Center actually gave pilots coordinates for an attack on the destroyers themselves.

Moise also took a hard look at the intercepted North Vietnamese radio messages which the US government interpreted as unequivocal proof that the second attack had taken place.  The cables, according to the Defense Department were orders dispatching patrol boats T-142 and T-146, and Torpedo Boat T-333 to attack Maddox and Turner Joy.  But Moise’s reinterpretation, plus his research into the types of ships referred to in the cable suggests something altogether different.  T-333 was the least damaged of the three torpedo boats that had attacked the Maddox on August 2; T-142 and T-146 were Swatows, lightly armed coastal patrol boats.  If the mission referred to in the cable was an attack on two destroyers more than 20 miles out to sea, sending two patrol boats which had no weapons capable of doing serious harm, and a torpedo boat with a damaged engine and out of torpedoes (they only carried two) did not make much sense. (113). Moise suggested  the cable’s intention is even further perplexing given the fact the torpedo boat in question had put ashore on the afternoon of August 2 and remained there until at least August 5 (113). And if none of that raised sufficient doubts about the authenticity of the second attack, Moise provided readers with summaries of the interrogations of North Vietnamese naval personnel.  Despite what official US Navy records indicate, no North Vietnamese naval personnel captured during the war suggested under interrogation that any of that country’s vessels had been sunk on the night of August 4 or that any combat had taken place that night.  The Navy pinned its case on the data provided by Captain Nguyen Van Hoa, a North Vietnamese naval officer who specialized in military law.  Moise reports that under questioning ,  Captain Hoa mentioned the attack on August 2, but had no information about an engagement involving the loss of North Vietnamese boats on August 4 (194). He did mention an incident in which three torpedo boats were lost in an engagement with the US Seventh Fleet.  Somewhere in Washington, Moise argued, a US naval officer misinterpreted Nguyen statement about the three lost torpedo boats as a reference to the attacks in August, 1964 (194).

Yes, three North Vietnamese torpedo boats had been lost in action against the US Navy, but in July, 1966, not August, 1964. The information, claims Moise, was there in Hoa’s interrogation transcripts.  His chronology was off, and who could blame him.  He was a military law specialist, not a line officer, and his knowledge of torpedo boat activities must have been limited.  Moise argued the only way he could have know about the sinkings was the way most sailors learn about things; scuttlebutt.  And even that, claimed Moise, would have been limited to who was responsible for the loss of boats T-333, T-336 and T-339, which were part of Torpedo Boat Squadron 135 (195).  Moise argued the Navy either suppressed or ignored more valuable intelligence from Tran Bao, one of nineteen survivors captured in the July, 1966 incident, and who had been deputy commander of Torpedo Boat Squadron 135 in 1964.  The simple math of it, concluded Moise, was that there was no way North Vietnam could have lost three torpedo boats to the US Navy in 1964.  There were only twelve torpedo boats in the entire North Vietnamese Navy, a gift from the Soviet Union in 1964. There had been no additions or subtractions until 1966. The prisoners listed the boats by number, mentioned the crew and even the captains of each boat, until T-333, T-336 and T-339 were lost in 1966.  So the math would indicate, Moise concluded, that torpedo boats couldn’t have been part of the August, 1964 events.  And if the US Navy was really interested in getting North Vietnam’s take on those events, asks Moise, why not quiz the prisoners?  Tran had been Deputy Commander of Squadron 135, and had written the August 2 after action report. Nguyen Van Gian commanded T-339 in 1964 and was still commanding it in 1966.  With nineteen men in custody and their interrogations cross checked for inconsistencies, the chances of an orchestrated cover up would be unlikely.  It’s here, suggested Moise, that the Navy exercised an early version of ‘don’t ask, don’t tell’ with regard to the Tonkin Gulf Incidents.  Word came down from Pacific Headquarters to not ask questions about that subject (195).

It’s not clear how often Cryptologic Quarterly is consulted to answer questions as part of a scholarly inquiry or even the significance of that contribution when something of note is discovered.  But in the case of the Tonkin Gulf Incidents, Robert Hanyok, a senior historian with the Center of Cryptologic History made a significant contribution to the dialogue with his 2005 article Skunks, Bogies, Silent Hounds, and the Flying Fish: The Gulf of Tonkin Mystery, 2-4 August, 1964. In his opening comments Hanyok acknowledges that over time, there were still those who rejected the growing skepticism about the August 4 attack;  Robert McNamara, the National Security Agency and those charged with writing the history of the US Navy figure prominently in that category.  Their convictions rested on the belief that the intelligence (what Hanyok refers to as SIGINT, short for signal intelligence) was both accurate and sufficient. With that information, they’d been willing to endure the charges of critics, citing radar, sonar, eyewitness and archival evidence that the second attack never happened (Hanyok 2).  They had been willing to disregard NSA Deputy Director Louis Tordella’s 1972 testimony that the SIGINT had been misinterpreted.

And they ignored Edwin Moise’s work which utilized small portions of SIGINT released to him under a Freedom of Information Act request.  Yet, argued Hanyok, even those few scraps of information Hanyok had in his possession should have been enough to seriously undermine any validity of the Johnson Administration’s belief the SIGINT reports it was looking at confirmed the August 4 attack (2).  Applying his skills as both a cryptographer and historian, Hanyok assembled the SIGINT discounting the second attack by dismantling the SIGINT the Johnson administration said points in that direction.  His conclusion was that the Johnson administration, including the President himself and the Secretary of Defense were deceived into thinking they had all the information.  In getting himself and the reader to that conclusion Hanyok’s research concluded there was only one word to describe what happened on the night of August 4; nothing.  Through what he referred to as “a compound of analytic errors and an unwillingness to consider contrary evidence”, Hanyok claimed American SIGINT elements in the region, along with the NSA in Washington reached consensus that North Vietnam was attacking the Maddox and Turner Joy that night.  In order to do that they committed further analytic errors and obscured existing information in order to produce ‘evidence’ of the attack (3).  In the end, he said, SIGINT information was presented in such a way as to preclude responsible decision makers from having the complete and objective narrative of events of August 4. For that to have happened, someone with access to critical information either had to withhold it or manipulate it. He stops short of pointing a finger. Although he suggests the only plausible reason for such mishandling was to support the Navy’s claim that the Desoto patrols were separate from any other naval operation in the area (namely the O-Plan 34 A) and had suffered a deliberate and unprovoked North Vietnamese attack.  Such a move would guarantee American retaliation, justify an escalation in US military activities, and a larger role for the US Navy at a time when it felt it was being left out of the action.  Had the intelligence been handled correctly, argued Hanyok, it would have told them that Hanoi’s navy was engaged in nothing that night other than trying to repair the two torpedo boats damaged in the August 2 engagement with the US Navy. Furthermore, it also would have indicated that even if North Vietnam was contemplating an attack, which it wasn’t, it wouldn’t know where to start because the SIGINT confirmed Hanoi didn’t know where the American ships were.  And that’s not because it had lost track of them; it’s because it wasn’t looking for them (3).

How could this have happened, asked Hanyok?  First, large portions (ninety percent, he claims) of the SIGINT from August 4 either never made it into the post-attack summary reports or were purged from the final report written in October 1964. The missing information told just where the North Vietnamese Navy was that night and what it was doing.  What actually went into the reports were the aforementioned analytic errors, plus unexplained translation changes and the conjunction of two unrelated messages into one translation. It was this cryptologic Frankenstein which would become the foundation of the Johnson administration’s proof of the August 4 attack.  The summaries weren’t manufactured out of whole cloth, however.  Legitimate intercepts were taken out of context and inserted into the summaries to give them a fig leaf of credibility. Whoever was responsible knew what he was doing and covered their tracks, Hanyok argues.  The sources of these fragments were never referenced in the summaries, and significant research had to be done to identify those sources before the wheat could be separated from the chaff.  Finally, in what could be considered a missing smoking gun, Hanyok points to the unexplained disappearance of a decrypted North Vietnamese after-action report from August 4.  The original went missing so administration officials had to rely on a translated text.  The problem, he says is that the Navy and NSA English translations are inconsistent.  Without the original it’s hard to figure out why there are differences in the translations and more importantly why the NSA took two separate messages were turned them into one (4). But regardless of the whereabouts of the original, the doctored report claimed Hanyok was all the administration needed to retaliate.  It quoted an unidentified North Vietnamese official who said that “two US planes had been shot down, two NVA ships had been ‘sacrificed’ and that “the enemy ship could also have been damaged” (23).

Apparently no one had stopped to reflect on the events of August 2: that two American planes were reported leaving the scene of battle; one was smoking as a result of a mechanical problem; the other was providing escort; that during the engagement all three of North Vietnam’s torpedo boats had been damaged, two seriously; and that one of them had managed to get off a few rounds at one of the US destroyers, hitting her.  How did this scrap of information, at that time two days old suddenly find itself the news of the hour and the snippet which made all the other pieces fall into place to prompt US retaliation?  Not even Hanyok is completely comfortable with mounting an accusation, but he has a couple of ideas.  He starts with the Marine listening post in Phu Bai, South Vietnam, first reporting on August 4 of possible North Vietnamese Naval operations planned against the Desoto patrol, and shortly after that elevating possible to imminent, and adding that North Vietnamese boats T-142, T-146 and T-333 had been ordered to make ready for military operations (20). The Marines concluded the ‘military action’ was an attack on the Desoto patrol, and made no attempt to investigate further.  Too bad for US policy makers, concludes Hanyok, because if Phu Bai had been ordered to do a more thorough job, none of the events after August 4 might have happened.  The communiqués from Phu Bai never mentioned a target or any objective of the military operation or even the nature of the operation.  Another problem with the message intercepted by the Marines was that it contained references which Hanyok says were misinterpreted; T-146 and T-333 had been ordered to execute what translates into English as a ‘long march or movement’.  What it really meant argues Hanyok, is that T-333 and a sister ship, T-336, both of whom had been involved in the August 2 action, were going to be towed for repairs.  A tugboat had been dispatched along with the T-142 to handle this mission.  Additionally all North Vietnamese boats were under strict orders to avoid contact with US ships.  The North Vietnamese message, claims Hanyok, implied that Hanoi thought the destroyers were close enough to its coast to place its ships in danger, when in fact the Maddox and Turner Joy were far out at sea.  The second error committed by the Marines at Phu Bai was its failure to consider just what the intercept meant when ship T-142 radioed that the tugboat it was escorting was indeed towing Torpedo Boats T-333 and T-336. The most confounding part of the communiqué was the conclusion made by the analysts at Phu Bai; “With torpedo boat T-336 added to its string, it appears that T-333will not participate in any military operations.” It would be hard for T-333 to do that, given the fact it was tied to a tugboat and bound for safe harbor to affect repairs.  So, the boats originally reported being ready to attack the Desoto patrol were incapable of even moving on their own.  In fact, this attempted salvage of the two damaged torpedo boats would occupy the efforts of Hanoi’s sailors for much of the night of August 4.  The Vietnamese would try various methods of getting the two damaged torpedo boats to a port for repairs.  Late in the evening of August 4 T-142 was ordered to escort the tugboat to its home base, and was then sent to a location near Haiphong.  It was then issued new orders; she was now to tow torpedo boat T-336. All of this chatter says Hanyok was being monitored not only by the Marine Station at Phu Bai, but also at a Navy station in the Philippines.  The traffic included reports of fuel transfers between the two damaged torpedo boats and the efforts of boats T-142 and T-146 to complete the towing exercise, a mission which still wasn’t completed by the morning of August 5.  So in reality, says Hanyok, none of the boats named in the original Marine warning participated in anything but salvage efforts (25). The question Hanyok poses at the end of this inquiry is this:  if the original suspect vessels, the T-142 and T-146, and T-333 and T-336 were not participating in the anticipated attack on the Maddox and Turner Joy, just who was?  Hanyok says there weren’t any further intercepted messages giving the mission to other boats.

If that’s the case, then just what was going on in the Gulf of Tonkin? At this point, regardless of the emphasis added by the Phu Bai station, all the SIGINT would accurately state was that there was no signals intelligence reflecting a planned or ongoing attack against the Desoto mission (26).  The NSA issued a summary report of the August 4 and 5 events which Hanyok claims was an attempt to throw up smoke. It claimed the ships attacking the Desoto patrol were Swatows, coastal patrol boats with no torpedoes.  But the real issue according to Hanyok was time, distance and speed, though not in that particular order.  In order for the Swatows to make the attack, and to appear on either Maddox or Turner Joy’s radar when they did, the attacking vessels would have had to cover 180 nautical miles traveling at a speed of nearly seventy miles per hour.  Impossible, claims Hanyok, since that would mean the boats were traveling 58% faster than their known top speed (27). So the Swatow’s lack of speed and armament ruled them out of the equation. What about torpedo boats? Same situation, asserts Hanyok.  The boats, which did carry torpedoes, were 140 nautical miles from where the two destroyers were.  So in order for them to show up on radar and mount and attack, they would have to have been traveling at about 70 miles per hour, nearly 40% higher than its known top speed.  In both cases, if Hanyok’s, math is to be believed, impossible.  The only other possibility which Hanyok raises, and in all honesty makes sense, is that somehow Phu Bai interpreted the movement of OPlan 34A vessels as potential attackers.  They were, according to Hanyok, moving along North Vietnam’s coastline at about the time the ships of the Desoto mission were shooting at those radar returns.  If that’s correct, one could only imagine the embarrassment at all levels of the Navy and the National Security Agency.  And it might explain in part why no reference to that OPlan mission was made in the US Navy’s history in Vietnam, why official Washington never acknowledged it or why Defense Secretary McNamara never admitted to its existence during his February 1968 testimony before the Senate Foreign Relations Committee.  To mention it would have undercut the administration’s contention that the US was doing nothing to provoke an attack. To admit that the Marine listening post at Phu Bai, and the Navy’s own listening post in the Philippines mistook them for the North Vietnamese and spread a false alarm would have been too much to bear.

The events of the Tonkin Gulf were played out over forty-eight hours.  The events of the war it helped inspire played out over eleven years.  The effort to make sense of the former to give meaning to the latter has played out over forty years.  And, like the war which seemed to some to lack clarity of purpose, the search for that same clarity with regard to the Tonkin Gulf may yet be an unresolved quest.

Part 1:

Part 2:

Part 3:


Austin, Anthony.  (1971). The President’s War. New York: Lippincott, 1971.

Ellsberg, D.  (2002). Secrets: A Memoir of Vietnam and the Pentagon Papers.  New York: Penguin Books.

Galloway, John.  (1970). The Gulf of Tonkin Resolution. Rutherford, New Jersey: Farleigh Dickinson University Press.

Goulden, Joseph.  (1969). Truth is the First Casualty: The Gulf of Tonkin Affair-Illusion and Reality. Chicago: Rand McNally.

Hanyok, Robert. (2005). “Skunks, Bogies, Silent Hounds and the Flying Fish; The Gulf of Tonkin Mystery, 2-4 August, 1964” Cryptologic Quarterly.

Karnow, Stanley.  (1983). Vietnam: A Television History. New York: Viking Press.

Kurland, Gerald.  (1975). The Gulf of Tonkin Incidents. Charlotteville, New York: SamHar Press.

Maitland, Terrence, Weiss, Stephen.  (1982). Raising the Stakes: The Vietnam Experience, Boston: Boston Publishing.

Marolda, Edward J., and Fiztgerald, Oscar P.  (1986). The United States Navy and the Vietnam Conflict, Vol. 2: From Military Assistance to Combat, 1959-1965. Washington, D.C.: Naval Historical Center.

Moise, Edwin.  (1996). Tonkin Gulf and the Escalation of the Vietnam War. Chapel Hill: North Carolina University Press.

Moyar,   M.  (2006). Triumph Forsaken: the Vietnam War, 1954-1965. Cambridge: Cambridge University Press.

Moss, George Donelson.  (2010). Vietnam: An American Ordeal, Sixth Edition. Upper Saddle River, New Jersey; Prentice Hall.

Porter, G.  (2005). Perils of Dominance: Imbalance of Power and the Road to War in Vietnam.  Berkeley: University of California Press.

Prados, J.  (1995). The Hidden History of the Vietnam War. Chicago: Ivan Dee.

_______ . (2005). “Tonkin Gulf Intelligence ‘Skewed’ According to Official History and Intercepts”. n.d.  National Security Archive Electronic Briefing Book

Guest Author

War History Online welcomes many guest authors who share their knowledge of the history on our pages. We work with various museums, historical societies and media outlets around the world. If you are interested in working with us or have a great story, please get in touch.