The MIT article also writes off as anecdotage an episode I witnessed firsthand and described in my 1975 article: the draft physical held at the Boston Navy Yard in May of 1970, a few days after Richard Nixon ordered troops into Cambodia. Several hundred Harvard and MIT students, and several hundred locals from South Boston and Chelsea, were summoned to the same place at the same time. Virtually all the college boys were deferred, because of a weird assortment of "ailments." Virtually all the local boys were approved to go to war.
It is of course possible for careful quantitative studies to reach conclusions that are not obvious from anecdotes. For instance, despite the impression that disproportionately many blacks died in Vietnam, surveys have determined that by the end of the war the proportion of total black casualties was almost exactly the proportion of blacks in the U.S. population. (Blacks did suffer disproportionate casualties in the early years of the war.) The question, therefore, is how carefully each study is set up. Researchers had direct means of figuring out whether dead soldiers were black or white; the MIT scholars had no way of determining just how rich or poor their dead soldiers were. So they simplified reality.
The three-page summary of the MIT study omitted several of the study's potential skews and limitations. For instance, on the researchers' first pass through the data, they had to wrestle with the fact that the census had no median-income figures for towns smaller than 2,500 people. They solved that problem by simply throwing out all the dead from towns smaller than that even though such people represented a fifth of all soldiers killed in Vietnam and were almost certainly not from America's wealthiest class. On subsequent passes the authors reinstated those soldiers and used approximations based on the median incomes of their counties' rural populations -- not a big step toward realism. They assessed and analyzed the data in countless other ways; they offset some of the most obvious distortions of their median-income approach in ways too complicated to go through here.
The most impressive part of the study -- the only impressive part, to my mind -- involved a separate sample of 467 dead. These were drawn from four cities: Chicago, Baltimore, San Antonio, and Portland, Oregon. The researchers tried to find the exact home address of each dead soldier, and from that drew conclusions about his family's economic standing. This study found a relatively small skew in death rates -- although it had limitations of its own. Each family's income was still defined as the median of a surrounding area, though in this case it was a census block containing about a thousand people rather than an entire town.
At the end of all their labors, every correction in place, the authors concluded that there was, in fact, a disparity in death rates. They broke the casualties into "deciles" -- groups each containing 10 percent of the total -- on the basis of family income, subject to all the doubts about how income was defined. They found that the wealthiest decile accounted for 7.8 percent of the casualties, or 22 percent less than its proportionate share. The second-poorest decile suffered 13.1 percent of the casualties, or 31 percent more than its share. (The very poorest had a lower-than-average death rate, for reasons I'll explain.) The poorer, therefore, were 68 percent more likely to die than the richer.
Leaping abruptly from statistical arcana to political and historical opinion, the authors concluded that this disparity rebuts the class-war hypothesis, since it is surprisingly small. "Most people instinctively think as many as three, four times more of the poor died than the rich, which is not true," Arnold Barnett told an interviewer from the Boston Herald. ("Most people"? The stench of anecdote is in the air.) This mistaken impression, the authors say at the end of their study, "demeans the sacrifices of the wealthy by implying that such sacrifices were nonexistent."
BY forming historical judgments, the authors guaranteed that their study would get more attention than a purely statistical exercise. But they also put themselves on terribly shaky ground, since questions about sacrifice and class war are not going to be resolved by bogus "disparity" studies.
Taken even at face value, the study's findings confirm, rather than challenge, part of the conventional assessment of Vietnam. No one I'm aware of contends that it was a "lowest decile," poorest-of-the-poor war. Many of the poorest Americans were disqualified from service, as anyone who observed the process knows, because they couldn't meet medical, educational, or disciplinary standards.
Rather, the class-war concept asserts two things about the American Army that served in Vietnam: it was principally made up of men from working-class and lower-middle-class backgrounds, and the American elite was conspicuously absent. "Conspicuously" does not mean "totally." Service academy graduates came from relatively privileged backgrounds and suffered heavy casualties in Vietnam. Two of the most dangerous specialties in Vietnam were those of pilots (especially helicopter) and infantry lieutenants. Because of the educational requirements for those jobs and the role of ROTC in providing lieutenants, they drew many men from affluent backgrounds, many of whom died. It is possible that death rates in Vietnam were more evenly distributed across income groups than were other measures of hardship which created the impression of a class war. The most obvious of these is that draftees and draft-induced volunteers lost several years of their youth while their contemporaries were starting families and careers.
In political terms the real burden of the war -- a family's sense that it was feeding its sons into a machine over which it had no control -- was shunted away not just from the truly rich but from most of America's upper-middle class. According to the version of history offered by the MIT study, the Bill Clinton of the 1960s was an aberration: while other men, regardless of background, were going off to do their fair share, he was a scheming shirker. But according to the version of history available from every other source, he was typical of people with the same education at the same time. It is anecdotal but significant that sons of Franklin Roosevelt, Joseph P. Kennedy, and Prescott Bush were in combat during the Second World War. It is anecdotal but also significant that the next generation of those families was generally not involved in Vietnam.
If the political conclusions of the MIT study were accurate -- if the burdens of the Vietnam War really were more or less fairly shared -- you would think that someone would have mentioned it by now. At least one reporter watching troops in combat, one novelist re-creating the scene, one politician remembering the rise and fall of domestic support for the war -- one of these people would have said what the MIT group does.
I mentioned the MIT conclusion to David Halberstam. "No!" he thundered. "All you had to do was see them to know that this was America's lower-middle class. Vietnam was a place where the elite went as reporters, not as soldiers. Almost as many people from Harvard won Pulitzer Prizes in Vietnam as died there." I asked James Webb, the novelist, Vietnam veteran, and former Secretary of the Navy. whether he had ever heard anyone involved with the war express a "shared burden" view. He said, "No." In the 1980s Webb called the registrars of Harvard, MIT, and Princeton, and asked each for two figures: the number of men who graduated from the undergraduate school from 1962 to 1972, and the number of those men who died in uniform. A total of 29,701 men graduated from the three schools; a total of twenty died in Vietnam, according to Webb. The classic study of who did and didn't serve in Vietnam, Chance and Circumstance, by Lawrence Baskir and Wllliam Strauss, found that men from disadvantaged backgrounds were more likely than average to be in the military, to serve in Vietnam, to be in combat units.
I asked Donald Rumsfeld, who was Secretary of Defense just after the fall of Saigon, whether the Army in Vietnam had been representative of America. "It was very clear what had happened with the draft," he said. "There was an accommodation between the government and the academic community. Students, teachers, and people who figured out how to work the system were exempted. It is inconceivable that a system designed and operating the way the draft did could have produced a true cross-section of America in the military."
Even William Buckley, who embraced the study's findings in his column, has implicitly rebutted them. I asked him whether his own experience squared with the assertion that the war had been "all-American." "I am lucky enough not to know anyone who died in Vietnam," he replied. "The only one I know who was wounded there is John Kerrey, who was in Skull and Bones." Buckley's own son, Christopher, has himself propounded the class-war thesis, on the basis of the way he and his Yale classmates avoided the draft.
"This is the same bogus scientific worship of 'hard data' that got us in so much trouble during the war," said Wllliam Broyles, the former editor of Newsweek and a creator of China Beach, who served as a Marine during the war. "I always wanted to grab those social-science nitwits and take them into the villages they insisted were safe because the computer said so."
Indeed, the problem with the MIT researchers may simply be that they missed their time. Thirty years ago they would have fit in perfectly. They could have worked with Robert McNamara on his studies proving that we were sure to win the war.