The Statistics Trap in Vietnam

A Dec. 23 headline in Outlook -- "The Military Misled Us in Vietnam" -- is a sad reminder that, long after that tragic war, we remain caught in a statistical trap we created for ourselves. What a pity that our press, which prides itself on its liberation from other illusions, has not fought free from this one.

"The profoundest truth of war is that the issue of battles is usually decided in the minds of the opposing commanders. . . . The delusion to the contrary has been fostered by the typical military history, filled with . . . statistical computations of the numbers engaged." So wrote the British historian Liddell Hart about the First World War; his words apply with even greater force to the Vietnamese conflict.

Among the artifacts that we Americans carried to Vietnam in our cultural baggage, one of the most highly prized was "statistics." This wonderful Western science would at last bring order to the jumbled mystery of the East. Not only would it help us know our enemy; better still, it would unravel the complexities of our South Vietnamese ally.

The results, unfortunately, were disappointing in both respects. I remember one day sitting as a member of the Mission Council, the body that ran the war under the direction of the American ambassador. On our agenda was a decision as to what figure we should use for the population of South Vietnam. We had only the vagu notion of how many people lived in that benighted country, and there was certainly no way of taking a census in the midst of war. Yet to make our statistics work, we needed this fundamental figure.

After a brief discussion, we settled, if I remember correctly, on 14 million -- as good a guess as any other, and roughly the number we assumed for North Vietnam. Actually, we didn't even know how many people there were in Saigon itself; the local police chief told me there might be in his jurisdiction as many as a million refugees who did not figure in any official record.

As this example shows, the ebb and flow of war did not lend itself to statistical techniques, which are properly applied only to discrete, identifiable objects or quantities. They are not applicable to objects in flux, such as drops of water -- nor to the surging, scintillating sea of cryptic Vietnamese. How would an American pollster fare if 50 percent of our voting population were Republicans during most of the daylight hours, but Democrats after dark and all day on alternate Tuesdays?

The problem of determining enemy strength, central to the Westmoreland law suit against CBS News' "60 Minutes," was to those of us on the spot just one segment of a vast numerical and classificatory muddle. Even the word "Vietnamese," applied to both sides in the civil war, was an almost meaningless abstraction in the South, so great was the ethnic, linguistic and religious heterogeneity of our prot,eg,e, the Republic of Vietnam. As an independent nation, it was less than two decades old and had not had time to enlist the loyalty of most of its disparate population. Thus one of the imponderables was always: Who is enemy? Who is friend?

In redefining and narrowing his count of the enemy, Westmoreland was not simply exercizing a prerogative but, more important, fulfilling a duty, when he specified which enemy forces were crucial to North Vietnam's conquest of the South, and which were not. By dismissing certain categories, he upgraded the reliability of his order of battle, while also clearly implying that the war was no longer to be thought of as a mainly guerrilla conflict. It had not in fact been so for some time. It was not therefore Westmoreland's revised order of battle, but rather those persons who overvalued the people's revolution in the South as against the invasion from the North who, if you will pardon the expression, "misled the president."

At this point, someone is sure to invoke the concern expressed by Gen. Wheeler, Ambassador Komer and others regarding public reaction to the divergent CIA and MACV figures. Such reaction was not surprising nor necessarily sinister. Wars are won by strong, contentious men, totally devoted to their own side's cause.

That is part of what Lidell Hart is talking about when he says that battles are won or lost in the minds of the opposing commanders. It would have been unreasonable for the leadership in Washington and Saigon to be happy about the release of some rather dubious figures, likely to be universally misunderstood, and certainly used to the detriment of the war effort.

My own feeling at the time, as CIA chief of station for Vietnam, was that this statistical controversy was simply an annoying irrelevancy; acceptance ofthe figures proposed by my organization would not have affected a single decision in the war. Though I had to release the CIA cables from Saigon, I otherwise maintained my distance from the whole matter.

Reverting again to Liddell Hart, an enemy is not defeated until he thinks he is. Even if we Americans had by our criteria "won" the Vietnamese war, it is unlikely the North Vietnamese would have credited us with victory. Buoyed by our evident fear of attacking their national sanctuary except by air, they would have treated their reverses more as a temporary setback than a defeat. With inexhaustible patience, they would then have geared down once again to a prolonged guerrilla struggle.

Our side can't work that way. Our war- time leaders are not bolstered by a quasi- mystical ideology that assures them of the righteousness of their cause, nor by authoritarian control of their people. On the contrary, they find themselves buffeted by all sorts of criticism, some of it fully justified, some of it -- as in the case now at issue -- simply puerile. It is arguable that the latter type is more effective, being largely immune to reason. Eventually, therefore, our leadership may lose confidence in itself, and even as stubborn a man as Lyndon Johnson, defeated in his own mind, will step aside.