War of Numbers

by Sam Adams

  I raised my hand.

  Mr. Graham smiled: “OK, Sam, fire away.”

  I did so. It took almost two hours. First I told them about the difference between Colonel Hawkins’ real and official positions (not using his name, since it might leak back, and get him in trouble.) Next I described the bargain that had led to the exit of the Vietcong self-defense militia—MACV’s note having unfortunately disappeared, but with Bill Hyland there to vouch for its existence. Finally, at a blackboard, I laid out the affair of Long Dat District, showing how MACV had “scaled down” each of its service detachments. Singling out the ordnance unit, which had dropped from fifteen to three, I described the jobs of each of the twelve soldiers who’d been crossed off the list. They might have belonged to an American ordnance detachment. I concluded: “Gentlemen, if we count them in the United States Army, why can’t we count them in the Vietcong’s?”

  There was a moment of uneasy silence. Mr. Graham broke it: “I guess the numbers aren’t settled after all.” Bill Hyland—who was tending the draft at one end of the conference table—said: “Christalmighty, this can go on forever.”28 The meeting dispersed. Two board members came up to where I was erasing the blackboard. One, an old southern gentleman named Ludlow Montagu, said: “It makes my blood boil to see the military cooking the books.” The other was Sherman Kent, the man who had established the Board of National Estimates at CIA in the agency’s early years. He asked: “Sam, have we gone beyond the bounds of reasonable dishonesty?” I replied: “Sir, we passed them in August.”

  During the next week, the agency’s fo’c’stle seethed with rebellion over the terms of the Saigon agreement. George Allen berated Carver twice daily. Major Blascik chomped on his pipe. But the most overt signs of revolt arrived in Carver’s office in the form of comments by various CIA officials on MACV’s latest proposed briefing meant to explain the Saigon numbers to the press. The average comment was angry, but the most bilious of all came from an official in the Office of Economic Research, Paul Walsh. “As seen from this office,” he wrote on 11 October,29 “I must rank it as one of the greatest snow jobs since Potemkin constructed his village.” It was so blatantly misleading, Walsh concluded, that “it gives us all the justification we need to go straight again.” The hint was strong that it was Saigon where the agency had gone astray. I called Walsh’s office to congratulate him, but he was out.

  By Friday the thirteenth, even Carver had begun to waver. Perhaps stung by the ferocity of the attacks on MACV’s briefing, he sent his own comments on it to the Pentagon for relay to Saigon. He wrote that one paragraph of the briefing was so bad that it would be “torn apart by the Saigon press corps.” Another was “a clumsy piece of dissimulation.”30 These were strong words and when I read them Monday afternoon, I allowed myself to think the Saigon agreement was heading for the brink.

  On Tuesday morning a message came from Vietnam that I felt pushed it over the edge. The message relayed the testimony of two midlevel defectors from the Delta. Both had the same story; the communists were reorganizing their army, in part by lumping together the guerrillas and the self-defense militia. I reread the message to make sure I’d gotten it right. I had. Its significance was immense. If the two types of soldiers were to consolidate, we could no longer tell them apart, either through captured documents or any way else. Willy nilly the Vietcong were parading the militia back into the MACV’s Order of Battle.

  I showed the cable to Carver. I hadn’t seen much of him lately, not through distaste, but because I’d begun to repeat myself. I said: “Mr. Carver, see if it strikes you the same way it struck me.” He read it.31

  “Lord,” he said softly. As I had, he read it again. He said: “Send a message to Saigon, and referencing this cable, have the station send a flier to the provinces asking whether the reorganization is countrywide. If it is, we may have to reopen the bidding on the OB. Meanwhile, I’ll tell the director.” He sounded almost cheerful.

  Theresa Wilson gave me some cable blanks, which I commenced to fill out. As I did so, another thought struck me. This was the second big VC reorganization I’d heard of in the last six weeks. The first was the one Tom Becker had pointed out in Saigon; the communists were streamlining their local structure to “expedite operations” into the city. Meanwhile, the Chieu Hoi rate was still falling off. What was going on, anyway? It was interesting to speculate, but my job, as Carver had put it, was to “reopen the bidding” on the VC strength estimate.

  I tried. Fourteen Three continued to meet, and at each session I brought in new evidence on the numbers. They wouldn’t budge, but Mr. Graham agreed to some changes in wording. The overall estimate became “at least.” A sentence was added that it “could be considerably higher.” Then there was the exchange between Graham and me over the phrase “we believe their military force is in the range of …”

  ADAMS: “Sir, how about taking out ‘we believe’ and putting in ‘we estimate?’ ”

  GRAHAM: “You mean on the grounds that we don’t really believe it?”

  ADAMS: “Yessir.”

  He did so. But he wouldn’t change the numbers. Once again, the Estimates staffer, Bill Hyland, tried to explain: “You’re tilting at windmills. Helms has us locked into the military’s figures. We can’t change them without his permission.”

  Fourteen Three met for the last time in the board conference room on Friday, 20 October 1967. With the completion of the last draft, resistance to the Saigon agreement ran out of steam. Gloomily I read the comments of the once-incipient rebel, Paul Walsh, on Monday afternoon. On behalf of his office, he wrote: “We share with many others numerous reservations about the estimate. We feel that the OB figures generally understate the strength of the enemy forces but recognize the apparent obligation for the estimate to be consistent with the figures agreed to at Saigon.”32 It was disgusting. Less than two weeks before, Walsh had recommended that the agency “go straight again,” and had called MACV’s proposed press briefing a “Potemkin’s village.”

  The village was almost up. At the agency’s insistence, the Pentagon had reluctantly inserted mention of the militia’s exit from the OB into MACV’s yet-to-be-given briefing, but this candor disturbed Saigon, including Ambassador Bunker. He cabled the White House (“Eyes Only Rostow”) on 28 October that telling the press about the militia’s departure “still bothers General Westmoreland, Bob Komer, and myself. Given the overriding need to demonstrate progress in grinding down the enemy, it is essential that we do not drag too many red herrings across the trail.” To admit dropping the militia from the OB was “simply to invite trouble … Far better in our view is to deal with the matter orally if it arises (in the hopes of) forestalling many confusing and undesirable questions.” He concluded: “Sorry to badger you on this, but the credibility gap is such that we don’t want to end up conveying the opposite of what we intend.”33 I tried to envisage the white-haired old gentleman whose last words to me in Saigon were to give his best to my father. I couldn’t, and stuck his cable in a file marked “self-defense” along with the messages Komer, Abrams, and Westmoreland had sent about the militia in August.*

  My hopes flickered briefly on the morning of 3 November. Helms still hadn’t signed Fourteen Three and two big pieces of evidence arrived that I felt might stop him. The first was a cable from the Saigon Station, answering my query about the communists’ lumping together the guerrillas and self-defense militia. The station couldn’t yet tell whether such a consolidation had taken place, but it had news from the provinces that was equally important. The VC home guard was everywhere in a state of unprecedented flux, with guerrillas joining regular infantry units in some areas, training as sappers in others—a vast roiling about of the entire guerrilla-militia. Although the VC’s purpose was unclear, the station recommended a sharp increase in the guerrilla numbers in Fourteen Three. I took the cable to Carver.34 “A little late,” he said, “but I’ll send it on to the director.”

  The other evidence was a captured documen
t. It came from —of all places—the DDI. Its bearer was a young analyst named Douglas Parry from a newly formed office whose main purpose, at last, was to study the Vietcong. Although MACV had published the document in late September, I’d missed it. A staff officer at COSVN headquarters had written the thing in “early 1967,” I guessed about April.35 “Take a look at page ten,” said Parry. I did so. There it said that VC guerrillas in South Vietnam numbered 150,000. Assuming April was correct for the date, this figure was only six months old. Yet Fourteen Three’s “currently agreed” number was 80,000. Could the guerrillas have dwindled that much in just half a year? It was almost inconceivable.* I thanked Parry, the first DDI-er to have given me a document since Molly had in August 1966. Then I showed it to Carver. “I’ll send it on.” he said.

  It was useless. Late that afternoon Theresa handed me a memo from Helms. It wasn’t to me personally, but to everyone who held copies of the final draft of Fourteen Three. His memo was called an “Introductory Note,” explaining how American intelligence had managed to underestimate the size of the communist army.36 First, Helms blamed the South Vietnamese, because their information was “unreliable.” Then he blamed “a social environment where basic data is incomplete and often untrustworthy.” Finally, he condemned “complex methodological approaches which cannot rise above uncertain data inputs.” Baloney! The reason we’d underestimated the communist army was that no one had looked at a damn thing until August 1966! With growing anger, I read the remaining paragraphs. The last one said that the VC had a “deliberate” policy of “sacrificing” the lower levels of their army in order to maintain the strength of the regulars (which was true), but that the lower level to sacrifice the most was the guerrillas, “now estimated to total some 80,000.”36

  Eighty thousand? So Helms was sticking to the Saigon agreement after all! I quickly reviewed what the “all” consisted of. First, George Allen and me, the only ones at agency headquarters who’d worked on the subject; second, Colonel Hawkins, whose warning I’d passed to George Carver; third, the agency’s Saigon Station, its advice for a “sharp increase” in guerrillas only hours old; and fourth—and most important—the latest VC document, which gave a number almost double eighty thousand. Having blithely disregarded the “all,” just who was Helms regarding?

  I said to myself: The damn liars at MACV headquarters, that’s who!

  At that moment, I happened to be holding a yellow Eberhart Faber no. 2 pencil. I snapped it in half. That was the turning point of my career at the CIA, although I didn’t realize it at the time.

  I quickly simmered down, and for several minutes looked at the two halves of the yellow pencil, wondering what to do. Gradually the thought formed: Helms still hadn’t signed the estimate, and by God, I’d make one last try to stop him, this time putting all my thoughts in writing. That was important, in writing. The maxim had come to mind (I believe it was my old Africa boss, Dana Ball’s), “Bureaucrats cringe from written complaints as Dracula does from the cross.” Ok, that was funny, but I’d have to be careful.

  I reread the Introductory Note. Its one piece of useful information was that the chiefs of the big intelligence agencies were scheduled to gather at 10:30 Thursday morning, 9 November, to review the final draft of Fourteen Three. That gave me six days to write my complaint. I tossed the pencil halves into a wastepaper basket, took out a ballpoint pen, and wrote down the title: “Comments on the Current Drafts of the Introductory Note and Text of National Intelligence Estimate Fourteen Three.” Not very pithy, but precise. I chewed on the end of the ball point, and began:

  “Having studied the Vietcong manpower problem since early 1966, and discussed various drafts of Fourteen Three for what seems almost as long, I wish to make the following comments as a matter of permanent record. They are my views and not necessarily those of my office.” (This last took George Carver off the hook.)

  I kept at the comments off and on for five days, using the same documents and arguments I’d used since early summer. By midafternoon of Wednesday, 8 November, I was beating an old hobbyhorse of mine, how the low manpower numbers had skewed all the other estimates, such as logistics, when I lost power. I read the comments over. There was something missing. It hit me what was wrong. Until that point they simply rehashed evidence. There were no conclusions. I chewed some more on the ballpoint pen, and under the heading “General Comments on the Text,” wrote as follows:

  I see no reason to dwell at length on why I think the current draft of Fourteen Three is an inadequate piece of analytical work. I will make four points, briefly.

  First, the current draft is ill-formed and incoherent. Discussions of logistics, losses, and manpower are largely unrelated.

  Second, the draft is less than candid. It conceals rather than edifies, using such devices as the phrase “at least” to obscure the possible existence of tens of thousands of Vietcong soldiers. Too often, it attempts to blame the evidence as inadequate, when the fault is not in information but in analysis.

  Third, the draft is timid. Its history is one of attacks by soldiers and politicians, and retreats by intelligence officials. Rather than admit the extent of past underestimates of enemy strength, its authors hide behind disclaimers and refuse to add up numbers, while protesting that it is inadvisable to make sums of apples and oranges.

  Finally, it is unwise. Although it intimates that there are “considerable” numbers beside the “at least 223,000 to 248,000” (listed in paragraph 37), it does not come to grips with the probability that the number of Vietcong, as currently defined, is something over half a million. Thus it makes canyons of gaps, and encourages self delusion.”37

  The “gaps” referred to were credibility gaps. Now, I realize by current standards, these conclusions look pretty dry, but for me back in those days, they were high rhetoric. Therefore, I decided to try the comments out on some other people before handing them to George Carver. Theresa Wilson typed it all up—saying “I hope you know what you’re doing” when she was done—then I showed them to Don Blascik. “You’ve hoisted the Jolly Roger,” he said. Next was George Allen.

  “Chicken thieves,” he said. “You’re calling the CIA leadership a pack of chicken thieves.”

  “Is it too strongly worded?” I asked nervously.

  “No,” he replied. “It’s about right. Much weaker, and they wouldn’t get the point. Much stronger, and they’d hang you for insubordination. We’ve got to stop those numbers from getting in concrete.” He handed the comments back, saying “When you give this to Carver, tell him it’s with my blessing.”

  Carver read them the following morning with what I took to be horrified disbelief. When he was through, I said: “Sir, I have two requests. The first is that you show these to Mr. Helms before his ten-thirty meeting today. The second is for permission to give copies to the other people around the building, particularly the Board of National Estimates.”

  Looking me square in the eye, he said: “Permission granted.” Under the circumstances, it was an admirable reaction, especially so since he’d had a hand in what I was complaining about. Carver disappeared—I presumed to the director’s office—and I went to a Xerox machine. I distributed twenty-five copies by nine-thirty. Among the recipients were Mr. Graham and several other board members: R. Jack Smith, the head of the DDI, and his deputy, Edward Proctor, plus a number of lesser entities, such as Bill Hyland, Paul Walsh, Dean Moor, and Molly. Then I hunkered down to await the arrival of goodness knows what, maybe a posse.

  No posse arrived, however, either for the rest of Thursday, or on Friday either. In fact nothing happened at all. On late Friday afternoon, curiosity overcame me, and I went to see my old friend in the DDI front office, the ex-Laotian analyst, Jack Ives. He showed me a note Edward Proctor had sent to R. Jack Smith concerning my comments. The note said: “This is the work of an angry young man. We ought not to allow it in our files without writing an answer.” However, no answer was forthcoming.

  Helms signed Fourteen-Three
on Monday morning, 13 November 1967.38 Since the first meeting 144 days earlier, the estimate had gone through twenty-two separate drafts, the hardest-fought in agency history. When I saw Helm’s signature—“concurred in” by the entire “United States Intelligence Board” (it said just below it)—I went to Carver to say I wanted to quit the director’s office. Carver looked embarrassed when I told him why, but he said there was nothing he could do. I thanked him for his help on the order of battle earlier in the year, and asked to transfer to the new DDI office whose job was to follow the Vietcong. He said OK, he’d see what he could do. I returned to my desk.

  Well, maybe Helms and the U.S. Intelligence Board thought the numbers problem was laid to rest, but there was one group who didn’t: the Vietcong. They were trying to build up their regular army. Several more missing units had turned up lately, and I increasingly suspected that the order of battle might be as low for this type of soldier as it was for the others. There was a problem, however; checking regulars was a prodigious task. Whereas the OB carried guerrillas, for instance, in a lump sum by province, it listed regulars by individual unit. Thus I’d have to snake them out battalion by battalion, platoon by platoon. There seemed no way of accomplishing this except by plodding once more through the VC documents. This I commenced to do.