Free War of Numbers by Sam Adams Page A

the weekly number of enemy defectors and POWs, or refugees from communist territory. Under the third card, marked “Everything Else,” was all the paper that didn’t belong under the first two. The third pile was so big I threw it away.
    I began with the captured documents. They were not originals, of course, but translations from the Vietnamese (in English, like everything else arriving in the cubicle, which explained why Molly’s dictionary had become a footrest). It took only an hour to read them, the vast majoritybeing routine reports like VC company rosters, ordnance chits, and ID cards, which had little to do with morale. In fact, the only document I came across that clearly bore on the subject was a letter from a North Vietnamese private to his parents in Haiphong. He hated the army, his sergeant picked on him, and someone had stolen his flashlight, he wrote, sounding very much like an American GI. I went on to the POW and defector reports.
    It took me a short time to realize their bias. Clearly anyone who’s fallen into the hands of his enemies is either feeling bad, or that’s what he’ll tell whoever asks him. The misfortunes varied. Some VC said that they had malaria, others that they were scared, and still others that they had grown sick of communism. The single exception was a VC lieutenant who would only say to his interrogator that he (the interrogator) was a “dirty running-dog imperialist lackey.” Hard-core VC. I went on to the statistics.
    The first account sheet I looked at was a weekly Chieu Hoi report, Chieu Hoi being Vietnamese for “Open Arms,” the name of Saigon’s defector program. The report listed the number of VC soldiers to have shown up at government defector centers for a week in late August. The number was 211.
    WHAT? Two hundred and eleven? Once again my mind raced back to the Japanese experience during World War II. I racked my brain, and couldn’t recall a single instance, not one, of a Japanese soldier becoming a turncoat. Well, this had at last become interesting. Right away, I looked at the other Chieu Hoi reports for August. Two hundred and eleven wasn’t unique; it was an average! That meant that VC soldiers were defecting at a rate of more than 10,000 a year. The Pentagon’s factbook put the size of the VC army at just under 200,000. * In other words, Vietcong soldiers were turning traitor at an annual rate of 5 percent, or one man in twenty. It was absolutely phenomenal!
    In a state of high excitement, I ran down to the archives in the agency basement, and pulled every Chieu Hoi report for the last twoyears. I compared them week by week, month by month, and year by year. Then, to see whether the defector figures simply kept pace with other types of VC losses, I compared them to the number of VC reported killed. An extraordinary story emerged. Not only was the defector rate growing fast, it was growing much faster than the number of Vietcong dead. For example, between August 1964 and August 1965, VC dead had about doubled. But for the same two months, VC defectors had almost quintupled. In other words, the more VC we killed, the more there were, proportionately, who wanted to quit. Therefore, if the war expanded (and the ratios stayed the same), the communists would eventually find themselves with a huge defection rate.
    Double-checking the figures, I wrote a memo to this effect. It was carefully worded, with a chart for the numbers. The secretary typed it up, and with some trepidation—knowing his bias—I handed it to Ed Hauck. It was just as I feared.
    “Statistics,” he smiled. “McNamara dotes on statistics, but I’ve never been able to make head nor tail of them.”
    “OK, I know how you feel,” I replied. “But look at the trend. Let’s grant the numbers are inaccurate, probably even padded. But the increase is too big to be explained away by that. Five times higher than before. Five times. Why?”
    It was a good question, and Ed Hauck knew it. He reread