War of Numbers (13 page)

4   BULLETIN 689

MID-MORNING OF THURSDAY 18 August 1966 was quiet. The dozen or so analysts of the newly formed Indo-China Division were in the first stages of pounding out the daily Sitrep. The division chief, Dean Moor, was away. Molly was less busy than usual, the Buddhist Struggle Movement having lost steam with the arrest of General Thi, the I Corps chief. Likewise the military analysts had slim pickings; the VC had set off a routine bomb in Hue, killing twenty-six civilians, and the Australians had just launched a search-and-destroy mission in Phuoc Tuy Province called Operation Smithfield. I read the
Washington Post,
waiting for Lorrie, the division secretary, to deliver my mail. As main recipient of captured documents, I was her last stop—and for good reason. The documents were at best several weeks old, and therefore unsuited to the current Situation Report.

Lorrie made her delivery at about ten o’clock. I put aside the
Post
and transferred the thin sheaf of papers she had dropped in my in-box to the center of my desk. I read them slowly, hoping to solve the puzzle of how long the VC could bear the drain of deserters from their 280,000-man force. At approximately ten-thirty, I came to Bulletin 689.
¹

It was only three pages long, and unlike most MACV bulletins—which normally dealt with several VC documents—Bulletin 689 concerned only one. Entitled the “Recapitulated Report on the People’s
Warfare Movement from Binh Dinh Province,” it had been picked up (according to “capture data” on its right-hand margin) by a trooper from the U.S. First Air Calvary Division in the northern part of the province on 30 May 1966. The document itself was undated, but since it concerned the first quarter of the year, it had to have been written some time after 31 March.

Whoever wrote the report first described the “general situation” in Binh Dinh, observing an “increase in American and satellite strength” there, and noting that the communist guerrillas, in addition to their military tasks, “performed civilian labor, evacuated dead and wounded …” and so forth, I pushed on to some statistics. These listed the number of guerrilla-militia in the province by type. I copied them down on a separate piece of paper, jotting in Vietnamese terms I’d learned from Co Yung earlier in the year: 3,194 village guerrillas (du kich xa); 11,887 hamlet guerrillas (du kich ap); 719 secret guerrillas (du kich mat); and 34,441 self-defense militiamen (tu ve). I added them up: 50,244.

What?
That seemed awfully damn high. I was familiar enough now with the MACV Order of Battle to know that of the 280,000-odd VC it listed, only 100,000 or so were guerrilla-militia. Why should Binh Dinh—one of forty-four provinces—have half of them? The obvious thing to do was to check the OB. I fished my copy (the 31 March edition, which I’d brought back from Vietnam) from my desk drawer, turned to its province listings, and looked up Binh Dinh: 1,446 guerrillas—no breakdown by type—and 3,222 self-defense militia, for a total of 4,668. Dumbfounded, I looked again at the VC total: 50,244.
It was eleven times higher than the OB! Forty-five thousand extra VC in a single province! Good God!!!

“Molly! Molly!” I shouted. “Look at this!”

“What’ve you got?” she asked mildly. I vaulted to her desk, slammed down the OB and the bulletin side by side, and with my finger stabbed at the two sets of numbers.

“Holy cow,” she said, “MACV sure missed the boat on that one. I wonder if it’s true for the rest of the country?”

“Long An!” I cried. Shaking with excitement, I told her about Colonel Anh’s local estimate, 2,000, compared to the order-of-battle figure, still 160. “Christ, I’ve known that for six months,” I moaned. “Why didn’t I check the OB before?”

“Maybe you better write something up,” Molly said. But I was already on my way out the door to the director’s special staff on Vietnam on the sixth floor. The person I wanted to see was George Allen, assigned to the staff after his tour in Vietnam had ended a month or two before. I barged into Allen’s office, and shoved the bulletin and the OB under his nose.

“Christalmighty!” he exclaimed, adjusting his glasses as if in disbelief. “I knew the OB was screwed up, but I didn’t realize it was this bad.” It ought to be written up, he said, reminding me that the CIA was about to publish a big study on the war for Secretary of Defense McNamara. I should be sure and get the Binh Dinh document mentioned in the study.

For the rest of the day I galloped around CIA headquarters like Paul Revere, hallooing about Vietcong guerrillas. The reaction was uniform: astonishment, and the realization that this was a terribly important finding that cried for further research. My last visit was to Dean Moor, back from wherever he’d been that morning. He was relatively calm, having been told about the document by Molly. Yes, I should write it up, he said, “but there’s no use doing it right away. We couldn’t get it out before tomorrow afternoon at the latest—and tomorrow’s a Friday. Nobody in Washington ever reads anything on weekends. Have it to me by close-of-business Monday.”

At home that evening, I could scarcely eat or sleep, I was so busy thinking about Bulletin 689. Its ramifications were enormous. For obvious starters, it suggested that Vietnam was a much bigger war than we thought it was. Furthermore, there were all kinds of analytical side effects. At this time, all our other intelligence estimates about the VC were tied to the number in MACV’s Order of Battle: how much rice the Vietcong ate, how much ammunition they shot off, and so forth. If the OB collapsed, our whole statistical system would go along with it.

My first act on Friday was to check in with Bobby Layton. He gave me a draft of “Will to Persist” so I could jigger around with its morale section. I did so, suggesting that the drain of deserters might not hurt the VC as much as I had originally believed because they had a much bigger army to draw from.
²

Also I asked Carver if I could highlight the implications of Bulletin 689 in the first part of “Will to Persist’s” summary, since McNamara was more likely to read that than anything else. Carver said no, the study was about to go to print, and besides my findings were only tentative; however, I could stick some caveats about Vietcong guerrillas in the body of the text.
³
This seemed fair enough on such short notice. Layton footnoted the right passages, and I went downstairs to the Indo-China Division.

Molly was waiting there for me with a wad of papers she had squirreled away from previous years. The first was an unpublished memo written by George Allen in October 1963, when George had worked for Ed Hauck. George’s memo complained that the guerrilla-militia estimate was too low even back then. The second was a Vietcong document dated 30 November 1965, which claimed they controlled six million people in South Vietnam (about
double
our own estimate for the same date!).
⁴
This helped explain where the communists recruited the additional soldiers, and was an extraordinary piece of news in its own right. A third was another VC report, this one from COSVN headquarters, which suggested the communists had tried “to increase the guerrilla-militia strength … to 250,000–300,000 men” by the end of 1965.
⁵
This last document gave me a rough idea of the kind of total we were dealing with. It was good to have a lid. In some of my wilder fancies the night before, I had gone much higher.

For the rest of Friday and over the weekend, I tore through my document file. Among the first ones I read applied to Phu Yen, the province on Binh Dinh’s southern border.
⁶
The MACV number for Phu Yen was almost as bad as its number for Binh Dinh.
^*
There were several other
provinces with estimates askew. I even came across one, Phuoc Long, where the order of battle was slightly too high.
^†

The initial draft took shape on early Monday afternoon. As usual, I wrote the text first—carefully referencing the VC documents that had emerged from the files—and then the summary. Its wording was clear, but cautious. It said that enemy reports “strongly suggested” that MACV’s guerrilla-militia estimate of some 100,000 was “too low,” and that it should “probably be at least doubled.” Observing that the higher numbers “would help explain why the southern communists have been able to field an increasingly large regular army despite heavy casualties and a high desertion rate,” it suggested dumping the matter in MACV’s lap.
⁷
(I had in mind its Order of Battle section chief, Colonel Hawkins.) Lorrie typed up the draft, and I gave it to Dean Moor, saying: “That phrase ‘at least doubled’ means it could be a hell of a lot more than double.”

“Hold on a sec,” he told me. I stood by as he finished editing Tuesday’s Sitrep. Ten minutes later, he took up my draft, and (looking bored, I thought) read it. He made a couple of minor changes, and spun it in to his out-box). He said, “The front office’ll get it first thing tomorrow morning.”

The next day I arrived at work in a fever of suspense. Surely the discovery of “at least” one hundred thousand extra Vietcong would cause a major convulsion. I imagined all kinds of sudden and dramatic telephone calls. “Mr. Adams, come brief the director.” “The president’s got to be told about this and you’d better be able to defend your paper.” I wasn’t certain what would happen, but I was sure it would be significant because I knew this was the biggest intelligence find of the war so far.
At least a hundred thousand extra Vietcong.
Surely President Johnson would have to send a lot more soldiers to fight them. I envisaged him calling
the director on the carpet, asking him why this information hadn’t been uncovered before.

Nothing happened. No phone calls from anybody. At the end of the day, I approached Dean Moor about the memo. “It’s kicking around the front office,” he told me. My agitation grew as the scene repeated itself on Wednesday and Thursday. On Friday morning, Lorrie dropped a memo on my desk. It wasn’t a carbon but the original, with no comments on it at all—no request for amplification, no questions about evidence, just a routine buckslip attached showing the entire DDI hierarchy had read it.

I boiled into Dean Moor’s office. “What the hell’s going on here?” I asked. “Don’t those people upstairs know what this means?”

He said: “If I were you, I’d make another stab at it. But not today. Remember about weekends. Have it to me close-of-business Monday.”

Once again I spent the weekend at CIA headquarters. On the theory that the numbers were covered well enough for the time being, I concentrated on describing exactly who these “guerrillas” and “militiamen” were. After looking at my notes from the Long An Chieu Hoi center, I wrote that guerrillas served in the villages or hamlets, that most carried rifles, and that they often acted as replacements for “regular Vietcong formations.” (This last thought stemmed from my recollection of the fast comeback of the VC 506th Battalion after the battle of the Right Testicle.) As for the militia, I wrote, they “dug trenches, burrowed tunnels” and “were expected to defend their hamlets as best they could when the enemy arrived.” “This is sometimes hard to do,” I explained, “because they are at the bottom of the logistical ladder,” and therefore poorly armed. I left intact the memo’s opening that the guerrilla-militia strength “should be at least doubled.”

New draft in hand, I reported back to Dean Moor, as requested, on late Monday afternoon. This time he took the draft right away, and edited it with a pencil. The first casualty was “at least doubled.” He replaced it with the phrase that the guerrilla-militia strength “should probably now be placed at around 200,000 …”
⁸

“Damnitall, Dean,” I said, “how do you know it’s around ‘200,000’? I already told you ‘at least’ could mean a hell of a lot more than that. If you ask me, it
is
a hell of a lot more than that.”

He said: “They didn’t accept the first draft. Let’s try it this way.”

“But that’s crazy,” I complained. “This isn’t a numbers game. There’s a war on. These people are shooting at us.”

“I want this memo to get out,” he replied; “I think it’s important.” Lorrie typed it with “around 200,000,” and the memo winged back to the front office.

By now, I was getting mad. The Sitrep normally made a fuss each time MACV found a new regiment, and here I’d discovered the numerical equivalent of
at least
fourteen VC divisions, and the only mention of it to leave the building so far was in the rear pages of “Will to Persist.” My spleen flew mostly at the DDI front office, but also now toward Dean Moor. I might mention here that he was none too popular with the other analysts, most of whom clearly preferred their ex-bosses, Ed Hauck and R. Sams Smith. In fact one such analyst wanted to quit the agency,
^*
and a couple of others were casting about for ways to leave the division.

For the rest of the week, I worked on the numbers. No new documents had come to light, so I constructed a simple equation. It went like this. If there were so many guerrillas and militiamen in the provinces on which we had documents (such as Binh Dinh, Phu Yen, and Phuoc Long), then their strength in the provinces for which we lacked documents must be
x.
I got some astronomical
x
’s before deciding to shoot for the minimum. That is, I concluded that the Vietcong had reached their minimum goal for 1965 of 250,000 guerrilla-militia.