Authors: Matthew M. Aid
Two hours later, a North Vietnamese Swatow patrol boat (T-142) began shadowing the two American destroyers, periodically reporting
the positions of the
Maddox
and the
Turner Joy
to headquarters by radio, messages that were intercepted by NSA listening posts in South Vietnam and the Philippines. After
completing his assigned patrol orbit off Hon Me, at four twenty-seven p.m. Herrick ordered the
Maddox
to retire to the mouth of the Gulf of Tonkin for the night before resuming its patrol along the coastline in the morning.
38
That night, from ten fifty-two to two past eleven p.m., South Vietnamese PT boats belonging to MACVSOG bombarded North Vietnamese
coastal installations, specifically a radar site at Vinh Son and a coastal defense installation at Mui Ron. These OPLAN 34A
attacks were sure to elicit a military response from the North Vietnamese. On their return to Da Nang, the South Viet nam
-ese boats were pursued for an hour by a North Vietnamese patrol boat.
39
Early the next morning, COMINT began picking up the first North Vietnamese military reactions to the Vinh Son–Mui Ron raids
that had taken place a few hours earlier. Radio intercepts collected by Marine Corps intercept operators at Phu Bai revealed
that the North Vietnamesenavy headquarters in Haiphong had connected the presence of the two American destroyers in the Gulf
of Tonkin with the OPLAN 34A raids on Vinh Son and Mui Ron and that a response was anticipated.
40
The Phantom Battle of August 4, 1964
After a long and sleepless night, at six a.m. on August 4 the
Maddox
and the
Turner Joy
resumed their patrol, making for the North Vietnamese coastline two hundred miles above the DMZ.
On the
Maddox
, Captain Herrick was decidedly unhappy about the position he had been placed in by his superiors, and he decided to take
action to protect his command based on what had happened to the
Maddox
two days previously. Although unaware of the OPLAN 34A attacks that had taken place just a few hours earlier, Herrick was
nevertheless concerned that the day’s patrol track called for him to once again orbit off Hon Me Island, where he knew a force
of North Vietnamese PT boats was based that could easily attack the destroyers with little or no warning. At eight forty a.m.,
Herrick sent the following message to Seventh Fleet headquarters in Japan:
Evaluation of info from various sources indicates that the DRV considers patrol directly involved with 34A operations and
have already indicated readiness to treat us in that category.
DRV are very sensitive about Hon Me. Believe this PT operating base and the cove there presently contains numerous patrol
and PT craft which have been repositioned from northerly bases.
Under these conditions 15 min. reaction time for operating air cover is unacceptable. Cover must be overhead and controlled
by DD’s at all times.
41
Admiral Thomas Moorer, the commander of the Pacific Fleet in Hawaii, read Herrick’s message and fired off an angry cable of
his own to CINCPAC, recommending the continuation of the Desoto patrol and arguing, “Termination of Desoto patrol after two
days of patrol ops subsequent to Maddox incident . . . does not in my view adequately demonstrate United States resolve to
assert our legitimate rights in these international waters.” What had started out as a simple intelligence collection mission
had now become a matter of asserting freedom of navigation on the high seas, as well as not showing any sign of weakness in
the face of North Vietnamese belligerence.
42
Herrick’s sense of apprehension was heightened when at nine thirty a.m. the radar operators on the
Maddox
and the
Turner Joy
picked up a radar contact of a “bogey” (unidentified surface craft) paralleling the course of the two American destroyers,
but then the target disappeared as quickly as it had appeared. Herrick concluded that his task force of destroyers was being
shadowed by at least one Swatow patrol boat.
The destroyers reached Point Delta, off Thanh Hoa, at eleven forty-five. They then shifted course to the south and followed
a course parallel to the North Vietnamese coastline down to a point opposite Hon Me, coming no closer than sixteen miles from
the coast. On the cruise southward, the radar operators on the two ships picked up a few contacts, but otherwise the patrol
was uneventful. After a tension-filled day with little intelligence to show for the effort, a relieved Herrick called off
the patrol at four p.m. and ordered a change of course to the east and the middle of the Gulf of Tonkin, well away from the
coastline, with the intention of resuming the patrol the following morning.
43
At six fifteen, a little more than two hours after Herrick had called it a day, the NSA listening post at Phu Bai sent to
the COMVAN on the
Maddox
a CRITIC message stating, “Poss DRV naval operations planned against the Desoto patrol tonite 04 Aug[ust]. Amplifying data
[ follows].” Twenty-five minutes later, Phu Bai sent a follow-up report, which stated, “Imminent plans of DRV naval action
possibly against Desoto mission,” adding that intercept messages revealed that two hours earlier three North Vietnamese Swatow
patrol boats had been ordered to “make ready for military operations the night of 4 August.”
44
Once again, Lieutenant Moore raced from the COMVAN to the bridge of the
Maddox
to hand-deliver the report to Captains Herrick and Ogier. Both men concluded that the intercept was an authentic order to
attack the destroyers. At seven thirty p.m., Herrick ordered the two destroyers to increase speed from twelve to twenty knots
in the hope of reaching the mouth of the Gulf of Tonkin before the pursuing North Vietnamese could catch up to them. Ten minutes
later, Herrick radioed the captain of the aircraft carrier
Ticonderoga
, steaming nearby, that he had received “info indicating attack by PGM/P-4 imminent. My position 19-107N 107-003E [60 miles
southeast of Hon Me]. Proceeding southeast at best speed.” He described the source of this information as simply “an intelligence
source.”
45
Less than a minute after Herrick’s message to the
Ticonderoga
went out, the radar operators on the
Maddox
picked up an intermittent surface contact (or “skunk”) forty-two miles to the northeast, which was where both destroyers had
anchored the previous evening. Fearing a trap, at seven forty-six p.m. Herrick ordered the
Maddox
and the
Turner Joy
to shift course away from the reported radar contacts. But Herrick was unable to shake his pursuers.
46
Four minutes after the
Maddox
and the
Turner Joy
changed course, at eight fifty a.m. EDT in Washington, Secretary McNamara and the chairman of the JCS, General Wheeler, were
briefed on the contents of the Phu Bai CRITIC message. At nine twelve a.m., McNamara informed President Johnson of the indications
coming from Fort Meade that the North Vietnamese intended to attack the
Maddox
and the
Turner Joy
. Wheeler telephoned Admiral Sharp at CINCPAC headquarters and told him to ensure that the captain of the
Ticon-deroga
, which was stationed off the coast fifteen minutes by air from the two destroyers, was apprised of the situation and to authorize
the carrier commander to take “positive aggressive measures to seek and destroy attacking forces if the attack should occur.”
47
McNamara did not waste any time beginning to plan a retaliatory strike. At nine twenty-five a.m. EDT, only thirteen minutes
after he had spoken to Johnson, McNamara called a meeting in his office attended by his deputy, Cyrus Vance, and representatives
of the JCS to discuss possible retaliatory measures if the North Vietnamese should attack the
Maddox
and the
Turner Joy
.
48
In the Gulf of Tonkin, events moved with astonishing speed. At eight thirty-six p.m. (nine thirty-six a.m. EDT), Captain Herrick
radioed that the radar operators on the
Maddox
and the
Turner Joy
were tracking two unidentified surface contacts and three unidentified aircraft. The unidentified aircraft disappeared from
the radar screens, but the radar operators on the two destroyers reported that the surface contacts were coming ever closer
at speeds of between thirty-five and forty knots. At nine thirty-nine p.m., the
Turner Joy
opened fire on a radar contact believed to have been a North Vietnamese PT boat that had closed to within seven thousand yards.
She was joined almost immediately by the five-inch guns on the
Maddox
. During the three-and-a-half-hour “battle” that ensued, the
Maddox
and the
Turner Joy
fired more than 370 rounds from their three-inch and five-inch guns and dropped four or five depth charges, beating off an
attack of what were believed to be six or more North Vietnamese PT boats and reportedly sinking two of the attackers—and amazingly
without sustaining a single hit from enemy torpedoes or gunfire.
49
The Day of Reckoning: August 5, 1964
The first FLASH-precedence messages about the naval engagement in the Gulf of Tonkin started coming across the teletypes at
the National Military Command Center in the Pentagon at eleven a.m. EDT on August 4, less than twenty minutes into the engagement.
The messages reported that the American destroyers were under attack and had evaded numerous enemy torpedoes.
At six past eleven a.m. (six past ten a.m. GOT time), Secretary McNamara called President Johnson to tell him that a sea battle
was then under way in the Gulf of Tonkin. Four minutes later, McNamara convened a meeting in his third-floor conference room
in the E Ring of the Pentagon with the members of the JCS, Secretary of State Rusk, and National Security Advisor McGeorge
Bundy to discuss military retaliation against North Vietnam. At eleven thirty-five a.m., McNamara, Rusk, and Bundy left the
Pentagon to attend a regularly scheduled NSC meeting at the White House, where they intended to recommend an immediate retaliatory
air strike against North Vietnam, which had the blessing of the JCS. At twelve forty p.m., McNamara briefed Johnson and the
NSC on the latest information available concerning what was occurring halfway around the world in the Gulf of Tonkin.
Within an hour of the meeting’s breaking up, Admiral Sharp telephoned McNamara from Hawaii to personally recommend air strikes
against the bases of the North Vietnamese torpedo boats. With this recommendation in hand, the JCS staff began selecting targets
for the retaliatory air strike from a ninety-four-target list that had been secretly compiled earlier in 1964. At a one p.m.
luncheon at the White House, Johnson, McNamara, Rusk, Bundy, and CIA director McCone unanimously agreed that retaliatory air
strikes were required.
50
At twelve twenty-seven a.m. on August 5 in the Gulf of Tonkin, Captain Herrick sent the following cautious message to Sharp:
“Review of action makes many recorded contacts and torpedoes fired appear doubtful. Freak weather effects on radar and overeager
sonarmen may have accounted for many reports. No actual visual sightings by Maddox. Suggest complete evaluation before any
further actions.” At twelve fifty-four, he sent a second message: “Joy also reports no actual visual sightings or wake of
enemy . . . Entire action leaves many doubts except for apparent attempt at ambush at beginning.”
51
At one thirty-five p.m. EDT, August 4 (twelve thirty-five a.m. GOT time, August 5), the JCS informed McNamara that a list
of targets had been compiled for air strikes, which could be executed if approved by the president. At a second NSC meeting
that afternoon, Johnson ordered that the retaliatory air strikes be executed and said that he would seek to obtain as quickly
as possible the support of the U.S. Senate for the strikes. As an NSA historical report notes, “Certainly none of the information
coming out . . . either before or in the hours following the execution order was sufficiently persuasive to support such a
momentous decision.” At three p.m., Secretary McNamara returned to the Pentagon to approve the target list for the air strikes,
leaving the preparation of the execute order to the JCS. He told the JCS that Johnson wanted the air strikes to begin promptly
at seven that evening (six a.m. GOT time, August 5) so as to coincide with a planned prime-time televised address by Johnson
to the nation.
52
As the plans for the retaliatory air strike moved rapidly forward, Captains Herrick and Ogier on the
Maddox
were frantically trying to ascertain what exactly had occurred while battling exhaustion and fending off urgent demands for
information from their superiors. When the two were told that during the engagement they had evaded a total of twenty-six
torpedoes, they immediately knew that something was terribly wrong, since there were only twelve PT boats in the entire North
Vietnamesenavy, each carry ing only two torpedo tubes that could not be reloaded at sea. What this meant was that even if
every single North Vietnamese PT boat had been in the Tonkin Gulf that night (an impossibility to begin with), they could
have fired only twenty-four torpedoes. Their suspicions were reinforced when they learned that all of the torpedoes had been
heard by the
Maddox
’s inexperienced sonar operator, while the more experienced sonar operator on the nearby
Turner Joy
did not hear one torpedo in the water during the entire four-hour battle. Someone on the
Maddox
finally figured out that every torpedo warning issued by the ship’s sonarman had followed a sharp change in course by the
Maddox
. A test proved that the sonar operator on the
Maddox
had mistaken the change in cavitation noises made by the destroyer when it changed course for the noise made by a torpedo.
53