Professor Jerry Lembcke asserts that it would have been highly unlikely that soldiers or veterans were spat upon because relations between soldiers and the antiwar movement were generally very friendly.
This post raises some serious problems with Lembcke’s use of one 1995 study by Beamish et al. to support this claim. In particular, Lembcke somehow falsely reports a 56% incidence of anti-troop behavior as a 6% incidence of anti-troop behavior, a mistake that he has repeated in several publications.
In a very revealing passage, Lembcke argues:
How do you prove that something did not happen? For this book I adopted two strategies. The first was to make the assumption that two mutually exclusive sets of circumstances cannot coexist in the same time and space. In the case of Vietnam veterans and the anti-war movement, I assumed that those two parties could not have been simultaneously hostile to one another and mutually supportive; anti-war activists could not have been spitting on veterans while at the same time befriending them in off-base coffeehouses. (Jerry Lembcke, The Spitting Image, 1998, pp. 3-4)
This reflects a rather unsophisticated view of human nature. The fact that most people don’t hate African-Americans doesn’t mean that stories of people using racial epithets against them are untrue. To explain spitting, there need be only a non-trivial minority who loathed the military during the Vietnam War (I’ll have more on this in future reports).
Lembcke may also be reflecting his own experience as an activist for Vietnam Veterans Against the War, a group that was always genuinely open to veterans who wanted to give up support for the US government’s position in the war and join its efforts. It is not at all a contradiction that most antiwar activists were welcoming to individual servicemen while a minority of those who opposed the war were quite hostile to them. The flaw in Lembcke’s logic can be illustrated by observing that, during the Cold War, the US welcomed Russian spies who wanted to change their orientation to the Cold War and join the CIA or the US side; yet Russian and US spies who had not had a change of heart were working hard against each other.
To support Lembcke’s view of very little anti-troop behavior by the antiwar movement, he cites a 1995 study by Beamish, Molotch, and Flacks, which counted 495 instances of pro-troop or anti-troop behavior in 380 New York Times and L.A. Times news stories accompanying major antiwar demonstrations.
They counted as an anti-troop stance “any portrayal that implied, even indirectly, a troop-blaming orientation” (Beamish, Molotch, and Flacks 1995).
The researchers used an extremely liberal interpretation of what counted as a report of anti-troop behavior. Many of the reports they counted were actually reports of confrontations between activists and uniformed authorities, such as National Guard reservists or military police, who were policing the demonstrations or peace marches. Even by counting such reports as “anti-troop,” the researchers reported finding only thirty-two instances, or 6 percent of all the stories over the six-year period, that could be construed as antagonistic behavior by the anti-war movement against GIs or veterans. (Lembcke, 1998, pp. 74-75)
It is true that the Beamish study used broad definitions: if antiwar demonstrators were rooting for the troops to lose on the battlefield, such as by waving Viet Cong flags, chanting for Ho Chi Minh to win the war, or holding up a sign saying “Beat Army,” that would apparently be coded as anti-troops. And if after chanting “End the war in Vietnam, we support the Viet Cong,” the demonstrators chanted “Support our boys in Vietnam, bring them home alive” (NY Times, April 17, 1966), Beamish et al. would code the latter chant as pro-troops, if this demonstration were to fall within their sample.
Here is the chart from Beamish et al. reporting their results (click the chart to enlarge it):
Note that in 380 stories from two major papers, Beamish et al. classified 279 instances of anti-troop behavior, which constituted 56% of the total of 495 coded instances. Counting 279 instances of anti-troop behavior in 380 stories sounds like a lot to me. They found 172 instances in which the demonstrators took some sort of action opposed to the troops (such as presumably waving the Viet Cong flag or cheering for a victory for Ho Chi Minh) and 27 instances in which the demonstrators themselves characterized their actions as anti-troops. They also found 75 instances in which others accused the demonstrators of being anti-troops, 74 instances in which the protesters were coded as pro-troops (such as protesters arguing to bring the troops “home alive”), and 142 instances of the demonstrators being against military elites (anti-brass).
Yet Lembcke falsely claims that only 6% of the 495 instances “could be construed as antagonistic behavior by the anti-war movement against GIs or veterans.” If you look at Beamish et al.’s chart (above), you can see what Lembcke did. He took the 6% of cases in which the demonstrators openly declared their opposition to the rank-and-file troops and falsely described this narrow subset of cases as if it were the numbers for the larger, “extremely liberal interpretation of what counted as a report of anti-troop behavior.”
I’ll be generous and call Lembcke’s sleight of hand extraordinarily sloppy. However it occurred, it is a very serious error to report (as Lembcke has repeatedly done) a 56% incidence of anti-troop behavior as a 6% incidence of anti-troop behavior, especially when Lembcke goes out of his way to quote Beamish et al.’s description (of the 56% figure) as representing “any portrayal that implied, even indirectly, a troop-blaming orientation.”
LIMITATIONS OF THE BEAMISH STUDY
The Beamish study is a serious piece of scholarship, but like any empirical study (including my own work) it has some limitations, most of which are no fault of Beamish et al.
First, as Beamish et al. note, the kind of spitting story that most people fixate on (a veteran returning through an airport) is not likely to be captured by looking at big street and campus demonstrations, which after all did not take place at airports. Although Lembcke claims, “No researchers cited reports that veterans were spat on (Beamish, Molotch, and Flacks, 1995),” Beamish et al. do indeed cite Bob Greene’s reports of many such stories (Beamish et al., p. 344, n. 2).
While Beamish et al. seem highly skeptical of these spitting stories, they note that these stories typically involve hostility “not from organized protesters, but from random bystanders.” Since their study covered only organized protests, they make clear that “We have no evidence to test the veracity of Greene’s accounts.” Id. Disturbingly, Lembcke uses Beamish et al. as evidence against stories of spat-upon returning veterans (such as many of Bob Greene’s stories), when the Beamish study explicitly says that they have NO EVIDENCE to test these accounts.
Second, as exhaustive as the 1995 Beamish study may have been for its day, it was done before the New York Times and the LA Times were searchable back to 1965 (the New York Times appeared online for the Vietnam era in 2002). Thus, the authors first determined which demonstrations to examine, and then looked at news reports in the few days before and after the event (a wider window was used for the NYT than for the LAT).
This approach causes them to miss some important behavior they would want to code. For example, Beamish’s article says that only a “smattering” of stories mentioned the presence of Viet Cong flags. I found that odd because Viet Cong flags were present at all of the few antiwar demonstrations I attended in the early 1970s. (I occasionally helped my Trotskyite roommate on some of his tamer activist projects when they were consistent with my McGovern-style liberalism of the day. Even my roommate, who favored a socialist revolution, opposed the presence of such flags.)
Using Proquest, I found references to Viet Cong flags in stories in either or both of the two newspapers Beamish sampled for the great majority of demonstration periods they selected between mid-1965 and mid-1971. When I started to dig deeper into a few demonstrations for which no Vietnam flags were mentioned in the NYT or LAT, I usually found references in other newspapers to such flags—or I found problems with Beamish’s sampling.
For example, a questionable set of demonstrations were those over the People’s Park in Berkeley, which Beamish et al. list for the entire month of May 1969. I skimmed roughly 40-70 articles in the NYT and LAT about those demonstrations during that period, and I didn’t see one that mentioned that they were antiwar or against the war, though again I might have missed one. Yet calling these antiwar demonstrations and (if I read their criteria correctly) including a month’s worth of stories on them might skew the results in one direction or another. I wonder what their results would be without these stories.
The reason that another demonstration (July 10, 1969 in Seattle) might have included no Viet Cong flags is that there were only 50 antiwar demonstrators present, too small to reflect a cross-section of the antiwar community. So beyond the great majority of mid-1965 to mid-1971 demonstration periods where Viet Cong flags were present, perhaps many of the other instances where they aren’t mentioned were either very small demonstrations or were not really concerned much with the war.
The limitations of the Beamish study’s approach are significant in another way: they report no instances of spitting on soldiers, a finding that Lembcke widely trumpets. But at least one of the spitting on troops stories I found started on the front page of the New York Times during their date window: James Reston’s story from the Pentagon on October 23, 1967.
Also, it would also have been better (in the interest of full disclosure) if they had mentioned the story of spitting on police officers (not soldiers) in the January 21, 1969 NYT, though since it involved police, it would not have been coded in their statistical counts. Beamish et al. also strangely limit the Saturday and Sunday, Oct. 21-22, 1967 Pentagon demonstrations, which ended early on the morning of Monday, Oct. 23, 1967, with the arrests of more than 200 people, as having occurred on only one day, Oct. 21, even though the main New York Times account of it is headlined on the carry-over page as a “two-day protest.” Coincidentally, limiting the event to only one of its two days meant that the Tuesday, October 24 LAT story mentioning soldiers being spat upon fell just outside their sampling window.
Further, if the Chicago Tribune had been included, another spitting on troops story would have surfaced, one at the Democratic convention (Sep. 1, 1968).
But there is another problem: not only do Beamish et al. report no instances of spitting (even though there were was at least one instance in the days of their samples), they also claim that there was no evidence of taunting troops. This suggests a possible breakdown by their coders or some problems with their sampling practices. (Did they use students who were influenced by the desired conclusion?) I wonder how they coded taunting, since there are some examples of what I would call taunting of troops in NYT and LAT stories for demonstrations that they examined. And there are more examples of taunting in stories about demonstrations that they did not cover.
This raises a more serious problem. Since nearly all demonstrations were handled by the police, not the military, there would ordinarily be little occasion for antiwar demonstrators to taunt military people directly except for veteran counter-protesters. (Especially in the earlier years of the war, almost all of the taunting appears to have been initiated by the pro-war people against the anti-war people.) Given the level of taunting of police by antiwar protesters, is there any reason to suppose that many of the antiwar protesters wouldn’t consider the national guardsmen or regular army troops as just another kind of “pig.” Again, the evidence is much more complex than that. Some protesters treated the military differently from the police, some treated them the same.
If one reads the newspaper accounts of the Chicago convention, there is very little mention of protesters taunting police, though enough was mentioned that it should have shown up in the Beamish study. When one reads the Walker Report, which on balances excoriates the police, one can see an incredible amount of abuse heaped on both police and guardsmen by activists over long periods of time. One can only speculate about the reasons for leaving it out of most stories: the relative triviality of repeated spitting given all else that went on, press hostility to the police because the police actually targeted the press for beatings (63 of the 300 press working the street were beaten by police), press sympathy for the activists, or the unprintable nature of the taunts.
One should not discount this last reason for not printing taunts in news stories more generally. For example, here are some of the taunts used against police or guardsmen that the Walker Report appeared to find credible:
Some abuse of guardsmen:
As Guardsmen arrive for the first time at the Hilton: chant of “Pigs, pigs, fascist pigs!” 212
Then chant of “Oink, oink, oink.” 212
Then jeers of “Pigs … sons-a-bitches … f–king pigs … f–king son-of-a-bitching pigs”
Then “Sieg-HEIL! Sieg-HEIL! Sieg-HEIL!” 212
African-American guardsman lt. taunted “mercilessly” as a “tool” of the “bad guys” 213
Calling several Jewish-named guardsmen “Kike” 213
Walking down the line and spitting in guardsmen’s faces 213
Flicking lighted cigarettes at guardsmen 213
Guard subjected to unbelievable abuse without any significant response. 214
To troops: “F–king draft dodgers, too chicken to serve your time in the regular army.” (Walker Report: “Though such taunts went on for about 20 minutes at a time, no response by the Guard was reported.”) 338
Other abuse, mostly of police:
“Pig! …Pig! … Pig!”
“Who’s your wife with now?” 247
“Where’s your wife tonight?” 247
“We’re f–king your wives and daughters while you guys are protecting your city.” 209
Young female pulls up her skirt and says to cop: “You haven’t had a piece in a long time.” 248
To citizens in cars on the street: “How would you like me to f–k your wife?” 276
Chants of “F–k, f–k, f–k.”
“F–k you, pig.”
“Fascist Pigs!” 210
“Pig f–kers.” 154
“Police are pigs.” 164
“Kill the pigs.” 173
“F–k the pigs.” 173
“Sieg heil, schwein!” 174
“Pigs, Oink, Oink!” 179
“Kill the pig, flush him out, bring him in!” [apparently from Lord of the Flies] 144
“F–k you.” 166
“F–k the pigs.” 224
“Dirty pigs.” 224
“You dirty son-of-a-bitch.” 166
Spitting and screaming insults at the police 235
Spitting on a policeman 338
Police bombarded by Pepsi cans “filled with urine”
Plastic bags full of urine and feces dropped from a 15th floor Hilton suite rented out to some of Eugene McCarthy’s staff and supporters
If you think this sounds pretty shocking, you should read some of the stuff I left out. And the Walker Report is justifiably much harder on the police’s behavior. Much of this stuff is not “fit to print” in the New York Times.
My point here is that there was a lot of taunting going on in the Chicago demonstrations that was never mentioned or even hinted at in most of the first-person newspaper accounts of the demonstrations. Most newspapers made no mention of taunting of guardsmen, and many made little or no mention of the taunting of police in their first accounts of the demonstrations.
What this means is that the Beamish study is not a report on what the antiwar demonstrations were like, but rather a report on how they were covered by the press. I also found it odd that the Beamish study included NYT stories from the two days before the demonstration, as well as the five days after a demonstration. This decision would seem to bias the results toward pro-troop behavior, with organizers seldom indicating their aspirations to abuse the troops.
In the days before the Chicago demonstrations, for example, would organizers have said to the press: “If the National Guard is called in, we plan to spit in their faces and taunt them mercilessly to try to get them to commit violent acts against us, which we would then photograph and use against the government”? I doubt that many activists were even thinking along those lines, yet (as the Walker Report makes clear) that is what happened with the taunters standing in front of the Guard at the Hilton. Note that the fact that most demonstrators were probably decent, nonaggressive sorts does not preclude viciously nasty behavior by a significant minority in a crowd.
Breaking down Beamish’s results by the day of the newspaper article (i.e., before or after the demonstration) would make it clear whether their results were biased by the inclusion of pre-demonstration stories. Along similar lines, I would be very curious to see the results broken down by newspaper. I get the impression that the LAT infrequently described negative behavior by antiwar activists at the same demonstrations where the NYT did describe such behavior.
More to come tomorrow . . . .