In depth | Tra Binh Dong — Why didn't the breached company base fall?

2026-09-17 · 1967-02-14 – 1967-02-15

Overview

Part 1: Tra Binh Dong (February 1967) set out three conditions for holding an isolated base: early warning, a reserve, and linked fires. This piece breaks the same night into minutes, using three sources part 1 lacked: an excerpt from the Korean brigade staff history Sillok Cheongnyong Budae [3], a Leatherneck article built on the testimony of the two US Marine ANGLICO radio operators at the base [2], and the Quang Ngai provincial page on the battle site [8]. Together they show that the company fought for over two hours with about a third of its base in enemy hands [2], and that its counterattack went in only after artillery had cut the penetration off [3].

Strategic and operational context

The ROK 2nd Marine Brigade (Blue Dragon) landed at Cam Ranh Bay in autumn 1965 and moved to Tuy Hoa that December [7]. In August 1966 it moved north into the III Marine Amphibious Force (III MAF) area south of Chu Lai [10]. The US Marine Corps history puts three Korean battalions in Quang Ngai Province at the start of 1967 and calls the relationship with III MAF "cooperation and coordination", not command [6]. The Koreans had no aircraft of their own; Sub-Unit One, 1st ANGLICO, put two radio operators with every rifle company as their link to air support, medevac and resupply [2].

From January 1967 the PAVN 2nd Division struck the brigade; on 10 January the 3rd Battalion command group was ambushed [2], and a Korean encyclopedia records a battalion operations officer falling that day [5]. The enemy used the 8–12 February Tet truce to resupply, and in southern I Corps chose isolated units as its main targets [6]. The 11th Company was one.

The enemy's aim is told three ways: to reach the US base at Chu Lai [2]; to finish the company within an hour and then hit the brigade artillery and Chu Lai airfield [3]; or to remove a strongpoint blocking a supply route north-west of Quang Ngai town [8]. In each, the company base was the first gate.

Terrain and weather in detail (OAKOC)

Detailed map of the company base: an oval trench line about 300 by 200 metres inside a 40-metre wire belt; 3rd Platoon and listening posts to the north-west, 1st Platoon north-east, 2nd Platoon south, the OP and 4.2-inch mortars in the centre. Arrows show two enemy battalions from the west into 3rd Platoon and one from the north into 1st Platoon, the 04:20 breach and the enemy-held western third, the 1st Platoon leader's sortie, the 06:40–06:52 double envelopment, the 08:00 landing zone, paddies to the south and a village to the south-east — Figure 1. Detailed map of the company base (approximate, estimated). Base size from Durand [2]; positions and times reconcile the Korean account [3] — own work

Observation and fields of fire (O). The base sat on a flat rise about 30 m high, with a rocky hill 1 km west and Route 1 1 km east [3]. Open paddies lay south, a village of about a hundred houses 200 yards south-east. North and north-west, low hills, chest-high scrub and jungle out to 300 yards came close to the wire [3]. The concealed approach was north-west from the start.

Avenues of approach (A). The enemy assembled in jungle behind the rocky hill, then after dark dug foxholes and laid telephone wire about 230 m outside the wire [3]. Two battalions came out of the north-western jungle at 3rd Platoon, one at 1st Platoon [2][3]. Nobody attacked across the paddies.

Key terrain (K). The OP at the centre held the command post and fire control, with the attached 4.2-inch mortars 20 yards south-west [3]. A radio operator recalled that enemy fire went straight for the command post and mortar pits [2]. A weapons site behind a rock about 100 m north of the wire overlooked this ground [2][3].

Obstacles (O). Seven rows of wire filled a 40 m belt, with 140 anti-personnel mines and 63 claymores [3]. Bangalores still cut it in front of 3rd Platoon's 1st Squad [2][3]. The wire did not stop the attack; it funnelled it into one gap.

Cover, concealment and weather (C). ERA5 reanalysis for the night shows 95–100 percent cloud, 80–83 percent humidity, light rain and 16–18 km/h winds [14], matching the Korean record of fog from 20:10, drizzle from 03:00 and wind that made listening very hard [3]. New moon fell on 9 February [13], so a thin crescent set around 22:30 (my estimate); the 04:10 assault came in total darkness. Reanalysis sunrise is 06:10 in UTC+7 [14]; if Saigon then used UTC+8, that is 07:10 local, fitting enemy troops seen leaving "as it grew light" at 07:24 [3] — an inference.

Organisation and command relationships

Side	Unit / element	Role / equipment	Source
ROK	2nd Marine Brigade (brigadier general)	Three battalions; cooperation with III MAF	[6]
ROK	3rd Battalion (lieutenant colonel)	Controlling HQ; requested reaction force and air	[3]
ROK	11th Company (captain), 10 officers and 284 men with attachments	Three rifle platoons, weapons platoon; three 60 mm, two 81 mm mortars, four 3.5-inch launchers	[3]
ROK	Attached: a rifle platoon of 1st Bn; four 4.2-inch mortars; a 106 mm recoilless rifle section; signal detachment	Extra infantry and heavy weapons	[3]
ROK	Brigade 105 mm and 155 mm guns, forward observer	Counter-mortar and blocking fires	[2][3]
US	Two ANGLICO radio operators	Air control; with the FO's radio, the only link once landlines were cut	[2]
US	AC-47, O-1 spotter, four A-4s, gunships	Early illumination (disputed); strikes after dawn	[2][3]
Enemy	PAVN 2nd Division	Overall control	[2][8]
Enemy	Two to three battalions of the 1st and 21st Regiments, plus a local battalion	Bangalores, flamethrowers, recoilless rifles, mortars	[2][3][9]
Enemy	Strength: ~2,400 (defector) / ~1,500 / "reinforced regiment"	Estimates only	[3][10]

The base carried heavy weapons a company would not normally have [3], and once enemy fire cut every landline, two PRC-25 radios — the observer's and an ANGLICO operator's — were the company's whole network [2].

Day-by-day account

Times follow the Korean account [3], checked against Leatherneck [2]; both give 04:10 for the assault and 08:00 for relief.

Animated map in six phases: enemy probes from the north-west at 19:00 and 23:20 on 14 February; mortar fire and a three-battalion assault at 04:10 on the 15th; the 04:20 breach and the enemy holding the western third; the 1st Platoon leader's team silencing a weapons site to the north; artillery isolating the gap at 06:30 and the counterattack at 06:40; the 07:24 lure, the 08:00 relief landing and the enemy withdrawing north-west. The still frame is the final situation — Figure 2. Movement by phase (18-second loop; with reduced motion only the final situation is shown). Approximate — own work, after the Korean account [3] and Durand [2]

14 February, 19:00–04:10 — probes and preparation

Around 19:00 an enemy patrol made light contact and pulled back, apparently a feint [3]. At 23:20 the listening post in front of 3rd Platoon's 1st Squad reported an enemy platoon at the wire; the commander held fire to five metres, then volleyed under flares, and the enemy withdrew [2][3]. He then added listening posts, redistributed ammunition, reviewed fire concentrations, ordered flares every 30 minutes and drew up a counterattack plan [3]. At 03:46 the brigade went on alert [3].

15 February, 04:10–05:10 — the breach

At 04:10 listening posts reported movement and wire-cutting, and enemy mortars and recoilless rifles opened fire [2][3]. The commander pulled the posts in and ordered fire held until the enemy was in range [3]. One battalion hit 1st Platoon, two hit 3rd Platoon [2]. At 04:20 bangalores blew the wire in front of 1st Squad, and by 04:22 the bulk of two battalions was pushing through [3]. The commander gave each supporting weapon one task: artillery on the enemy mortars and beyond the gap, rockets on the front, 60 mm on planned concentrations, the 106 mm on follow-on troops [3]. At 04:40 1st Squad's position fell; of about ten men, all but one became casualties [3][4]. Private First Class Cho Jeong-nam died at his post [4].

05:10–06:30 — a third of the base

At 05:10 the enemy came through the same gap again, reaching trenches and mortar pits; at 06:00 a flamethrower team entered the 4.2-inch position [3]. Two hours into the fight the enemy held nearly a third of the base [2]. The 1st Platoon leader led a small team outside the wire against an enemy weapons site, then helped stop a flamethrower team inside [2][3]. The commander moved a squad from intact 2nd Platoon to 3rd Platoon's right [3].

06:30–08:00 — counterattack, last push, relief

At 06:30 the forward observer hit what looked like the enemy command post and the guns cut off troops heading for the gap [3]. At 06:40 the commander attacked: a squad each from 1st and 2nd Platoons moved along the trenches to close behind the enemy while headquarters men and 3rd Platoon's survivors, led by the weapons platoon leader, drove into the centre [2][3]. By 06:52 the gap was sealed [3]. At 07:24 the commander pulled 3rd Platoon back to the OP to draw the remaining enemy in, then hit them with massed fire and mortars; as the sky cleared, an ANGLICO operator brought in four A-4s and gunships [2][3]. Fighting ended at 07:30, and at 08:00 sixteen helicopters landed a relief company [3]. The government newsreel of 4 March called it "four hours against two North Vietnamese battalions" [11].

Decision points in depth

Phase titles follow the training scenario; x1–x3 are what-if branches (Figure 3).

Decision timeline. Bands show darkness, estimated moonset, fog from 20:10 and drizzle from 03:00. The axis marks the 19:00 probe, 23:20 listening-post report, 03:46 brigade alert, 04:10 assault, 04:20 breach, 04:40 1st Squad overrun, 05:10 second push, 06:00 flamethrowers, 06:30 artillery strike, 06:40 counterattack, 06:52 gap closed, 07:24 withdrawal seen, 07:30 end and 08:00 relief. Boxes below give the actual decision at phases p1 to p7. The scale widens from 03:30 — Figure 3. Timeline and decision points. Times from the Korean account [3]; weather bands estimated from ERA5 [14] and moon phase [13] — own work

p1 An unconfirmed warning (14 Feb, dusk)

Known: a defector reported a planned attack without size or date [3]; battalion HQ had been ambushed a month before [2]. Options: listening posts and wire; registered fires; spoiling patrols; routine watch. Actual: nothing is recorded for dusk; after the 23:20 probe the company did (a) and (b) together [3]. Otherwise: patrols would have met an enemy already dug in 230 m out [3].

p2 The LP makes contact (23:20 / 04:10)

Known: an enemy platoon was at the wire; wind and rain made listening unreliable [3][14]. Options: keep the LP out; pull it in; use it to guide guns; hold fire to the wire. Actual: at 23:20 the company held fire and volleyed at five metres — the scenario's (d) [2][3]; at 04:10 the posts were pulled in [3]. Otherwise: the scenario's "LP holds" follows the encyclopedia [4]; the staff account has 1st Squad falling later in its trench [3] (see source criticism). Either way, the warned company met the first rush at the wire.

p3 Attack from every side (04:10)

Known: fire on the whole base, contact at two platoons, main effort unclear [2]. Options: fight in place and hold the reserve; perimeter-wide final protective fires; commit the reserve early; contract. Actual: platoons fought in place, supporting weapons got one task each, the reserve stayed [3]. Otherwise: an early commitment to 3rd Platoon would have guessed right but left nothing for the battalion at 1st Platoon [2].

p4 3rd Platoon's line is breached (04:20–06:52)

Known: bangalores and flamethrowers inside 3rd Platoon's sector, more troops coming, two radios left [2][3]. Options: contract; counterattack at once; artillery on the breach; hold until daylight. Actual: a squad shored up the gap; at 06:30 the guns cut off follow-on troops; at 06:40 the counterattack went in [3] — (c) isolating the penetration before (b) recovering it. Otherwise: a counterattack at 04:20 would have met two battalions in the gap; contracting meant leaving the trenches in the dark.

p5 Enemy mortars (about 05:00–06:00)

Known: direct fire from behind a rock in front of 1st Platoon kept hitting the OP [3]; grids were uncertain. Options: counter-battery fire; a sortie; ride it out; company mortars. Actual: the 1st Platoon leader proposed a sortie, the commander approved, and with three men under covering fire he crossed some 90 m and silenced the position with grenades from 20 m [3]; Leatherneck says 100 m [2]. The guns were hunting other mortars at the same time, so (a) and (b) ran together. Otherwise: a site overlooking the OP would have made the 06:40 counterattack hard to assemble.

p6 The last push (07:24)

Known: gap sealed, enemy still near, dawn coming. Options: fight from positions; shell the assembly area; draw in; go out through the wire. Actual: 3rd Platoon fell back to lure the enemy onto massed fire, then aircraft struck [2][3] — (a) and (c) mixed. Otherwise: going out would have tangled the company with the enemy, beyond help from guns or aircraft.

p7 Morning (08:00)

Known: enemy withdrawing, heavy casualties, wire cut. Options: reorganise; pursue; fire on withdrawal routes. Actual: the heliborne relief company pursued while 11th Company rebuilt [2][3]. Otherwise: a tired platoon sent out would have thinned the base; the manual counts landing space for relief as a planning factor [12].

What-ifs. x1–x3 are not in the record, but the core of x2 — gathering even command post staff for another counterattack — is what happened at 06:40 [3].

Doctrine re-read

Local security (FM 3-90, paras 1-92, 1-93, 1-121). The manual's active measures include observation posts, alert levels and, in very poor visibility, listening posts [12]. The post-23:20 measures fit [3].

Perimeter defence (paras 9-59 to 9-62, 9-68 to 9-73). The manual requires security outside the perimeter, a central reserve and reinforcement of the threatened sector once it is known; it warns that shallow perimeters are vulnerable to penetration and prescribes early warning and obstacles (9-68). Para 9-71 allows the reserve to be "a provisional force organized from available personnel" [12] — exactly the 06:40 force [3]. Paras 9-69 and 9-73 call for integrated outside fires, rehearsed counterattack plans and supply economy [12]; a rehearsal is not recorded [3].

Reserve (paras 9-10, 9-11). Lower echelons use reserves for local counterattacks and to block or contain penetrations [12]. The company did both in sequence: block, contain with artillery, recover [3].

The attacker's logic. The Vietnamese page describes the method as "strike fast, destroy cleanly" before reinforcements can arrive [8]; the Korean account has the enemy planning one hour for the company [3]. When the base still held two hours after 04:10, daylight brought back observed artillery and aircraft [2][3]. Time was on the defender's side.

Three flagpoles fly the flags of the United States, South Vietnam and South Korea above a circular plinth with a unit emblem; a wet square, trees and distant hills under an overcast sky — Photo 2. US, South Vietnamese and South Korean flags at III MAF headquarters, Da Nang, about 1967; the Korean brigade worked with III MAF by coordination, not under its command — official USMC photograph (Edward Graeff Collection, Archives Branch, USMC History Division), CC BY 2.0, via Wikimedia Commons [16]

Source criticism

The listening post and 1st Squad. The encyclopedia says all but one of the ten-odd men on LP duty became casualties [4]. The staff account has the posts pulled in at 04:10 and 1st Squad's trench falling at 04:40 [3]. They probably describe two moments of one squad; the staff account is the more detailed [3].

The AC-47. Summaries used in part 1 said fog kept the gunship out, and the Korean account says the flareship circled above the fog [3]. An ANGLICO operator recalled that the gunship did drop flares and fire until the company commander, fearing the light exposed his positions, sent it away [2]. Late but first-hand, this fits a brief pass under near-total cloud [14].

Enemy strength. About 2,400 comes from a defector [3]; a veterans' newsletter also gives 1,500 [10]; the newsreel said two battalions [11]. Unit numbers vary across sources [2][3][8][9], but all describe three battalions plus local forces — roughly a regiment.

Losses. Korean sources give 15 dead and 33 wounded [2][3]; a Vietnamese blog cites 11 dead [9]. Korean records count 243 enemy dead confirmed and 60 estimated [3], which the newsreel's 303 seems to combine [11]. The Vietnamese provincial page says the strongpoint was "completely destroyed" [8], yet the company held and was relieved that morning [2][3]. Both sides inflated; enemy dead are used here only to gauge the attack's size.

Date and place. HistoryNet dates the attack to early 14 February [1], against 04:10 on the 15th in [2][3] — probably counting the night probes. Korean and US accounts place the base south-west of Chu Lai near Route 1 [2][3]; the provincial page puts it in Tinh Tho commune, 14 km west of Quang Ngai city [8].

The records themselves. The Korean excerpt, retyped by a veterans' group, spells two commanders' names two ways [3]; its individual feats need care. Vietnamese sources are commemorative [8][9], Korean ones frame a "myth" [1], and neither cites the other.

Lessons

Use the hours after the first probe. The 23:20 contact gave three hours of preparation.
Contain a penetration, then recover it. The counterattack worked once the guns had closed the gap's mouth.
A reserve can be improvised from headquarters staff and broken-platoon survivors, but its leader should be fixed in advance.
When communications rest on two radios, their operators are key terrain.
Light and air support cut both ways. The commander must decide when to call them and when to send them away.

Unresolved questions

Times and losses in the original Korean official histories.
The exact site, base shape and the attached platoon's position.
Whether, and how long, the AC-47 fired.
Enemy units and the Vietnamese official account.
Distance and timing of the 1st Platoon leader's sortie.

Sources

James F. Durand, "Korea's Myth-Making Marines", Vietnam, October 2008 (posted on HistoryNet 2018-06-07). https://www.historynet.com/koreas-myth-making-marines/ (accessed 2026-10-14)
James F. Durand, "ANGLICO Marines at Tra Binh Dong", Leatherneck vol. 100 no. 2, Marine Corps Association, February 2017 (Internet Archive copy of 2018-06-20). https://web.archive.org/web/20180620154916/https://www.mca-marines.org/leatherneck/2017/02/anglico-marines-tra-binh-dong (accessed 2026-10-14)
Lee Kwang-hee (Blue Dragon Brigade operations staff), excerpt from Sillok Cheongnyong Budae, "The Blue Dragon heroes of Tra Binh Dong", transcribed on the veterans' site vietvet.co.kr, undated (in Korean). http://www.vietvet.co.kr/sugy/trabin/trabin.htm (accessed 2026-10-14)
Kim Dae-jung, "Cho Jeong-nam", Encyclopedia of Korean Culture, Academy of Korean Studies, 1995 (rev. 2023; cites the Marine Corps combat history) (in Korean). https://encykorea.aks.ac.kr/Article/E0052575 (accessed 2026-10-14)
Kim Dae-jung, "Cho Gyeong-sik", Encyclopedia of Korean Culture, Academy of Korean Studies, 1995 (rev. 2023) (in Korean). https://encykorea.aks.ac.kr/Article/E0051560 (accessed 2026-10-14)
Gary L. Telfer, Lane Rogers and V. Keith Fleming Jr., U.S. Marines in Vietnam: Fighting the North Vietnamese, 1967, History and Museums Division, HQ USMC, 1984 (PDF held on Wikimedia Commons; chapters 1 and 9). https://commons.wikimedia.org/wiki/File:U.S._Marines_in_Vietnam_-_fighting_the_North_Vietnamese_1967_(IA_usmarinesinvietn00gary).pdf (accessed 2026-10-14)
Stanley R. Larsen and James L. Collins Jr., Allied Participation in Vietnam (Vietnam Studies, CMH Pub 90-5), Department of the Army, 1975, chapter 6 (Göttingen university library mirror). https://webdoc.sub.gwdg.de/ebook/p/2005/CMH_2/www.army.mil/cmh-pg/books/vietnam/allied/ch06.htm (accessed 2026-10-14)
Quang Ngai Department of Culture, Sports and Tourism, "Di tích Đồi tranh Quang Thạnh" (Quang Thanh grass hill relic site), 2014-02-17 (Internet Archive copy of 2017-02-02) (in Vietnamese). https://web.archive.org/web/20170202034650/http://www.quangngai.gov.vn/vi/sovhttdl/pages/qnp-ditichdoitranhquangthanh-qnpnd-587-qnpnc-20-qnpsite-1.html (accessed 2026-10-14)
Chiến Trường Việt Nam (blog), "Trận đồi Quang Thạnh – Trà Bình Đông", 2019-02-20 (compilation of Vietnamese-side accounts; low reliability) (in Vietnamese). http://chientruongvietnam.com/2019/02/20/tran-doi-quang-thanh-tra-binh-dong-battle-of-tra-binh-dong/ (accessed 2026-10-14)
EC-47 Association, "1967 Timeline – February: Blue Dragons & Tally Ho" (newsletter PDF, undated). http://www.ec47.com/storage/UserFileFolder/1967_Timeline_-_February_-_Blue_Dragons_&_Tally_Ho_Article.pdf (accessed 2026-10-14)
KTV Archive, Daehan News no. 612, "News from Vietnam" (produced 1967-03-04; posted on YouTube 2016) (in Korean). https://www.youtube.com/watch?v=swbP1eX9YVQ (accessed 2026-10-14)
Department of the Army, FM 3-90 Tactics, 1 May 2023 (ch. 1 paras 1-92, 1-93, 1-121; ch. 9 paras 9-10, 9-11, 9-59 to 9-62, 9-68 to 9-73). https://armypubs.army.mil/epubs/DR_pubs/DR_a/ARN38160-FM_3-90-000-WEB-1.pdf (accessed 2026-10-14)
NASA Goddard Space Flight Center, Phases of the Moon: 1901 to 2000 (February 1967). https://eclipse.gsfc.nasa.gov/phase/phases1901.html (accessed 2026-10-14)
Open-Meteo, Historical Weather API (ERA5 reanalysis), grid cell 15.19° N 108.76° E, hourly temperature, humidity, precipitation, cloud cover and wind for 14–15 February 1967. https://archive-api.open-meteo.com/v1/archive?latitude=15.19&longitude=108.76&start_date=1967-02-14&end_date=1967-02-15&daily=temperature_2m_max,temperature_2m_min,precipitation_sum,sunrise,sunset&hourly=temperature_2m,relative_humidity_2m,precipitation,cloud_cover,wind_speed_10m&timezone=Asia%2FBangkok (accessed 2026-10-14)
Wikimedia Commons, File:Operation Dragon Fire October 1967.jpg (US Marine Corps, public domain). https://commons.wikimedia.org/wiki/File:Operation_Dragon_Fire_October_1967.jpg (accessed 2026-10-14)
Wikimedia Commons, File:Flags at III MAF, Vietnam, circa 1967.jpg (USMC History Division, CC BY 2.0). https://commons.wikimedia.org/wiki/File:Flags_at_III_MAF,_Vietnam,_circa_1967.jpg (accessed 2026-10-14)