Friday, 21 November 2014 / TRUTH-OUT.ORG

"Kill Anything That Moves" Military Doctrine Began in Vietnam

Monday, 18 February 2013 00:00 By Nick Turse, Metropolitan Books | Book Excerpt

Kill Anything That Moves.(Image: Metropolitan Books)When did the United States adopt, in the contemporary age, military standards of condoning a "kill anything that moves" doctrine of warfare, along with a widespread use of torture?

One need look no further than the Vietnam War, according to Nick Turse, an author and journalist who has documented the dark side of the US imposition of empire through armed intervention. In this assiduously documented book, Turse offers abundant evidence that My Lai was not an exception to military conduct, but rather, a not uncommon occurrence. In addition, the US slaughtered countless civilians in air and ground attacks without ever even seeing who was being killed.

In addition, the reader will also discover that Dick Cheney's backing of torture had ample precedent during the Vietnam War.-MK

Support Truthout's mission. Kill Anything That Moves (hardcover edition) is yours with a minimum donation to Truthout of $35 (which includes shipping and handling) or a monthly donation of $15.

The following excerpt is the introduction to Kill Anything That Moves: The Real American War in Vietnam.

"An Operation, Not an Aberration"

On January 21, 1971, a Vietnam veteran named Charles McDuff
wrote a letter to President Richard Nixon to voice his disgust with the
American war in Southeast Asia. McDuff had witnessed multiple cases
of Vietnamese civilians being abused and killed by American soldiers
and their allies, and he had found the U.S. military justice system to be
woefully ineffective in punishing wrongdoers. "Maybe your advisors
have not clued you in," he told the president, "but the atrocities that
were committed in My Lai are eclipsed by similar American actions
throughout the country." His three-page handwritten missive concluded
with an impassioned plea to Nixon to end American participation
in the war.

The White House forwarded the note to the Department of
Defense for a reply, and within a few weeks Major General Franklin
Davis Jr., the army's director of military personnel policies, wrote
back to McDuff. It was "indeed unfortunate," said Davis, "that some
incidents occur within combat zones." He then shifted the burden of
responsibility for what had happened firmly back onto the veteran.
"I presume," he wrote, "that you promptly reported such actions to
the proper authorities." Other than a paragraph of information on how
to contact the U.S. Army criminal investigators, the reply was only
four sentences long and included a matter-of- fact reassurance: "The
United States Army has never condoned wanton killing or disregard
for human life."

This was, and remains, the American military's official position.
In many ways, it remains the popular understanding in the United
States as a whole. Today, histories of the Vietnam War regularly discuss
war crimes or civilian suffering only in the context of a single
incident: the My Lai massacre cited by McDuff. Even as that one
event has become the subject of numerous books and articles, all the
other atrocities perpetrated by U.S. soldiers have essentially vanished
from popular memory.

The visceral horror of what happened at My Lai is undeniable. On
the evening of March 15, 1968, members of the Americal Division's
Charlie Company, 1st Battalion, 20th Infantry, were briefed by their
commanding officer, Captain Ernest Medina, on a planned operation
the next day in an area they knew as "Pinkville." As unit member
Harry Stanley recalled, Medina "ordered us to 'kill everything in the
village.' " Infantryman Salvatore LaMartina remembered Medina's
words only slightly differently: they were to "kill everything that
breathed." What stuck in artillery forward observer James Flynn's
mind was a question one of the other soldiers asked: "Are we supposed
to kill women and children?" And Medina's reply: "Kill everything
that moves."

The next morning, the troops clambered aboard helicopters and
were airlifted into what they thought would be a "hot LZ"— a landing
zone where they'd be under hostile fire. As it happened, though,
instead of finding Vietnamese adversaries spoiling for a fight, the
Americans entering My Lai encountered only civilians: women, children,
and old men. Many were still cooking their breakfast rice. Nevertheless,
Medina's orders were followed to a T. Soldiers of Charlie
Company killed. They killed everything. They killed everything that
moved.

Advancing in small squads, the men of the unit shot chickens as
they scurried about, pigs as they bolted, and cows and water buffalo
lowing among the thatch-roofed houses. They gunned down old
men sitting in their homes and children as they ran for cover. They
tossed grenades into homes without even bothering to look inside. An
officer grabbed a woman by the hair and shot her point-blank with a
pistol. A woman who came out of her home with a baby in her arms
was shot down on the spot. As the tiny child hit the ground, another
GI opened up on the infant with his M-16 automatic rifle.
Over four hours, members of Charlie Company methodically
slaughtered more than five hundred unarmed victims, killing some
in ones and twos, others in small groups, and collecting many more
in a drainage ditch that would become an infamous killing ground.
They faced no opposition. They even took a quiet break to eat lunch
in the midst of the carnage. Along the way, they also raped women
and young girls, mutilated the dead, systematically burned homes,
and fouled the area's drinking water.

There were scores of witnesses on the ground and still more overhead,
American officers and helicopter crewmen perfectly capable of
seeing the growing piles of civilian bodies. Yet when the military
released the first news of the assault, it was portrayed as a victory over
a formidable enemy force, a legitimate battle in which 128 enemy
troops were killed without the loss of a single American life. In a
routine congratulatory telegram, General William Westmoreland, the
commander of U.S. forces in Vietnam, lauded the "heavy blows"
inflicted on the enemy. His protégé, the commander of the Americal
Division, added a special note praising Charlie Company's "aggressiveness."

Despite communiqués, radio reports, and English-language
accounts released by the Vietnamese revolutionary forces, the My
Lai massacre would remain, to the outside world, an American victory
for more than a year. And the truth might have remained hidden
forever if not for the perseverance of a single Vietnam veteran
named Ron Ridenhour. The twenty-two-year-old Ridenhour had not
been among the hundred American troops at My Lai, though he had
seen civilians murdered elsewhere in Vietnam; instead, he heard
about the slaughter from other soldiers who had been in Pinkville
that day. Unnerved, Ridenhour took the unprecedented step of carefully
gathering testimony from multiple American eyewitnesses. Then,
upon returning to the United States after his yearlong tour of duty,
he committed himself to doing what ever was necessary to expose
the incident to public scrutiny.

Ridenhour's efforts were helped by the painstaking investigative
reporting of Seymour Hersh, who published newspaper articles about
the massacre; by the appearance in Life magazine of grisly full-color
images that army photographer Ron Haeberle captured in My Lai as
the slaughter was unfolding; and by a confessional interview that a
soldier from Charlie Company gave to CBS News. The Pentagon, for
its part, consistently fought to minimize what had happened, claiming
that reports by Vietnamese survivors were wildly exaggerated. At
the same time, the military focused its attention on the lowest ranking
officer who could conceivably shoulder the blame for such a
nightmare: Charlie Company's Lieutenant William Calley.

An army inquiry into the killings eventually determined that
thirty individuals were involved in criminal misconduct during the
massacre or its cover-up. Twenty-eight of them were officers, including
two generals, and the inquiry concluded they had committed a
total of 224 serious offenses. But only Calley was ever convicted of
any wrongdoing. He was sentenced to life in prison for the premeditated
murder of twenty-two civilians, but President Nixon freed him
from prison and allowed him to remain under house arrest. He was
eventually paroled after serving just forty months, most of it in the
comfort of his own quarters.

The public response generally followed the official one. Twenty five years later, Ridenhour would sum it up this way. At the end of it, if you ask people what happened at My Lai, they would say: "Oh yeah, isn't that where Lieutenant Calley went crazy
and killed all those people?" No, that was not what happened. Lieutenant
Calley was one of the people who went crazy and killed a lot of
people at My Lai, but this was an operation, not an aberration.

Looking back, it's clear that the real aberration was the unprecedented
and unparalleled investigation and exposure of My Lai. No
other American atrocity committed during the war— and there were
so many— was ever afforded anything approaching the same attention.
Most, of course, weren't photographed, and many were not
documented in any way. The great majority were never known outside
the offending unit, and most investigations that did result were
closed, quashed, or abandoned. Even on the rare occasions when the
allegations were seriously investigated within the military, the reports
were soon buried in classified files without ever seeing the light of
day. Whistle-blowers within the ranks or recently out of the army
were threatened, intimidated, smeared, or— if they were lucky—
simply marginalized and ignored.

Until the My Lai revelations became front-page news, atrocity
stories were routinely disregarded by American journalists or excised
by stateside editors. The fate of civilians in rural South Vietnam did
not merit much examination; even the articles that did mention the
killing of noncombatants generally did so merely in passing, without
any indication that the acts described might be war crimes. Vietnamese revolutionary sources, for their part, detailed hundreds of massacres and large-scale operations that resulted in thousands of civilian deaths, but those reports were dismissed out of hand as communist propaganda.

And then, in a stunning reversal, almost immediately after the
exposure of the My Lai massacre, war crime allegations became old
hat— so commonplace as to be barely worth mentioning or looking
into. In leaflets, pamphlets, small-press books, and "underground"
newspapers, the growing American antiwar movement repeatedly
pointed out that U.S. troops were committing atrocities on a regular
basis. But what had been previously brushed aside as propaganda
and leftist kookery suddenly started to be disregarded as yawn-worthy
common knowledge, with little but the My Lai massacre in between.

Such impulses only grew stronger in the years of the "culture wars,"
when the Republican Party and an emboldened right wing rose to
power. Until Ronald Reagan's presidency, the Vietnam War was generally
seen as an American defeat, but even before taking office Reagan
began rebranding the conflict as "a noble cause." In the same
spirit, scholars and veterans began, with significant success, to recast
the war in rosier terms.  Even in the early years of the twenty-first
century, as newspapers and magazines published exposés of long hidden
U.S. atrocities, apologist historians continued to ignore much
of the evidence, portraying American war crimes as no more than
isolated incidents.

But the stunning scale of civilian suffering in Vietnam is far
beyond anything that can be explained as merely the work of some
"bad apples," however numerous. Murder, torture, rape, abuse, forced
displacement, home burnings, specious arrests, imprisonment without
due process — such occurrences were virtually a daily fact of life
throughout the years of the American presence in Vietnam. And as
Ridenhour put it, they were no aberration. Rather, they were the
inevitable outcome of deliberate policies, dictated at the highest levels
of the military.

The first official American combat troops arrived in Vietnam in
1965, but the roots of the conflict go back many decades earlier. In
the nineteenth century, France expanded its colonial empire by taking
control of Vietnam as well as neighboring Cambodia and Laos,
rechristening the entire region as French Indochina. French rubber
production in Vietnam yielded such riches for the colonizers that
the latex oozing from rubber trees became known as "white gold."

The ill-paid Vietnamese workers, laboring on the plantations in
harsh conditions, called it by a different name: "white blood."

By the early twentieth century, anger at the French had developed
into a nationalist movement for independence. Its leaders found
inspiration in communism, specifically the example of Russian Bolshevism
and Lenin's call for national revolutions in the colonial
world. During World War II, when Vietnam was occupied by the
imperial Japanese, the country's main anti-colonial organization—
officially called the League for the Independence of Vietnam, but far
better known as the Viet Minh— launched a guerrilla war against
the Japanese forces and the French administrators running the country.
Under the leadership of the charismatic Ho Chi Minh, the Vietnamese guerrillas aided the American war effort. In return they received arms, training, and support from the U.S. Office of Strategic Services, a forerunner of the Central Intelligence Agency.

In 1945, with the Japanese defeated, Ho proclaimed Vietnam's
Independence, using the words of the U.S. Declaration of Independence
as his template. "All men are created equal," he told a crowd of
half a million Vietnamese in Hanoi. "The Creator has given us certain
inviolable rights: the right to life, the right to be free, and the
right to achieve happiness." As a young man Ho had spent some years
living in the West, reportedly including stretches in Boston and New
York City, and he hoped to obtain American support for his vision of
a free Vietnam. In the aftermath of World War II, however, the
United States was focused on rebuilding and strengthening a devastated
Europe, as the Cold War increasingly gripped the continent.

The Americans saw France as a strong ally against any Soviet designs
on Western Europe and thus had little interest in sanctioning a
communist-led independence movement in a former French colony.
Instead, U.S. ships helped transport French troops to Vietnam, and
the administration of President Harry Truman threw its support
behind a French reconquest of Indochina.

Soon, the United States was dispatching equipment and even
military advisers to Vietnam. By 1953, it was shouldering nearly 80
percent of the bill for an ever more bitter war against the Viet Minh.
The conflict progressed from guerrilla warfare to a conventional military
campaign, and in 1954 a Gallic garrison at the well-fortified
base of Dien Bien Phu was pounded into surrender by Viet Minh
forces under General Vo Nguyen Giap. The French had had enough.
At an international peace conference in Geneva, they agreed to a temporary
separation of Vietnam into two placeholder regions, the north
and the south, which were to be rejoined as one nation following a
reunification election in 1956.

That election never took place. Fearing that Ho Chi Minh, now
the head of the communist Democratic Republic of Vietnam in the
north, was sure to sweep any nationwide vote, the United States picked
up where its French partners had left off. It promptly launched efforts
to thwart reunification by arming its allies in the southern part of
the country. In this way, it fostered the creation of what eventually
became the Republic of Vietnam, led by a Catholic autocrat named
Ngo Dinh Diem.

From the 1950s on, the United States would support an ever more
corrupt and repressive state in South Vietnam while steadily expanding
its presence in Southeast Asia. When President John Kennedy
took office there were around 800 U.S. military personnel in South
Vietnam. That number increased to 3,000 in 1961, and to more than
11,000 the following year. Officially listed as advisers involved in the
training of the South Vietnamese army, the Americans increasingly
took part in combat operations against southern guerrillas— both
communist and noncommunist— who were now waging war to unify
the country.

After Kennedy's assassination, President Lyndon Johnson repeatedly
escalated the war with bombing raids on North Vietnam, and
unleashed an ever more furious onslaught on the South. In 1965 the
fiction of "advisers" was finally dropped, and the American War, as
it is known in Vietnam, began in earnest. In a televised speech, John-
son insisted that the United States was not inserting itself into a faraway
civil war but taking steps to contain a communist menace. The
war, he said, was "guided by North Vietnam . . . Its goal is to conquer
the South, to defeat American power, and to extend the Asiatic
dominion of communism." To counter this, the United States turned
huge swaths of the South Vietnamese countryside— where most of
South Vietnam's population lived— into battered battlegrounds.

At the peak of U.S. operations, in 1969, the war involved more
than 540,000 American troops in Vietnam, plus some 100,000 to
200,000 U.S. troops participating in the effort from outside the
country. They were also aided by numerous CIA operatives, civilian
advisers, mercenaries, civilian contractors, and armed members of
the allied "Free World Forces"— South Korean, Australian, New
Zealand, Thai, Filipino, and other foreign troops. Over the entire
course of the conflict, the United States would deploy more than 3
million soldiers, marines, airmen, and sailors to Southeast Asia.
(Fighting alongside them were hundreds of thousands of South Vietnamese troops: the Army of the Republic of Vietnam would balloon to a force of nearly 1 million before the end of the war, to say nothing of South Vietnam's air force, navy, marine corps, and national police.) Officially, the American military effort lasted until early 1973, when a cease-fire was signed and U.S. combat forces were formally withdrawn from the country, though American aid and other support
would continue to flow into the Republic of Vietnam until Saigon fell
to the revolutionary forces in 1975.

From the U.S. perspective, the enemy was composed of two distinct
groups: members of the North Vietnamese army and indigenous
South Vietnamese fighters loyal to the National Liberation Front, the
revolutionary organization that succeeded the Viet Minh and opposed
the U.S.-allied Saigon government. The NLF's combatants, officially
known as the People's Liberation Armed Forces (PLAF), included guerrillas
in peasant clothing as well as uniformed troops organized into
professionalized units. The U.S. Information Service invented the
moniker "Viet Cong"— that is, Vietnamese Communists—as a derogatory
term that covered anyone fighting on the side of the NLF, though
many of the guerrillas themselves were driven more by nationalism
than by communist ideology. American soldiers, in turn, often shortened
this label to "the Cong" or "VC," or, owing to the military's phonetic
Alpha-Bravo-Charlie alphabet, to "Victor Charlie" or simply
"Charlie."

By 1968 the U.S. forces and their allies in the South were opposed
by an estimated 50,000 North Vietnamese troops plus 60,000
uniformed PLAF soldiers, while the revolutionaries' paramilitary
forces—part-time, local guerrillas— likely reached into the hundreds
of thousands. Americans often made hard-and-fast distinctions
between the well-armed, green- or khaki-uniformed North Vietnamese troops with their fabric-covered, pressed-cardboard pith style helmets; the khaki-clad main force PLAF soldiers, with their floppy cloth "boonie hats"; and the lightly armed, "black pajama"–clad guerrillas (all of whom actually wore a wide variety of types and
colors of clothing depending on the time and place). In reality,
though, they were very hard to disentangle, since North Vietnamese
troops reinforced PLAF units, "local" VC fought in tandem with
"hard-core" professionalized PLAF troops, and part-time farmer fighters assisted uniformed North Vietnamese forces.

The plethora of designations and the often hazy distinctions
between them underscore the fact that the Americans never really
grasped who the enemy was. On one hand, they claimed the VC had
little popular support and held sway over villages only through terror
tactics. On the other, American soldiers who were supposedly
engaged in countering communist aggression to protect the South
Vietnamese readily killed civilians because they assumed that most
villagers either were in league with the enemy or were guerrillas
themselves once the sun went down.

The United States never wanted to admit that the conflict might
be a true "people's war," and that Vietnamese were bound to the revolution because they saw it as a fight for their families, their land, and their country. In the villages of South Vietnam, Vietnamese nationalists had long organized themselves to resist foreign domination, and it was no different when the Americans came. By then, the local population was often inextricably joined to the liberation struggle.
Lacking advanced technology, financial resources, or significant firepower,
America's Vietnamese enemies maximized assets like concealment,
local knowledge, popular support, and something less quantifiable— call it patriotism or nationalism, or perhaps a hope and a dream.

Of course, not every Vietnamese villager believed in the revolution
or saw it as the best expression of nationalist patriotism. Even
villages in revolutionary strongholds were home to some supporters
of the Saigon government. And many more farmers simply wanted
nothing to do with the conflict or abstract notions like nationalism
and communism. They worried mainly about their next rice crop,
their animals, their house and children. But bombs and napalm don't
discriminate. As gunships and howitzers ravaged the landscape, as
soldiers with M-16 rifles and M-79 grenade launchers swept through
the countryside, Vietnamese villagers of every type— supporters of
the revolution, sympathizers of the Saigon regime, and those who
merely wanted to be left alone— all perished in vast numbers.
The war's casualty figures are staggering indeed. From 1955 to
1975, the United States lost more than 58,000 military personnel in
Southeast Asia. Its troops were wounded around 304,000 times, with
153,000 cases serious enough to require hospitalization, and 75,000
veterans left severely disabled. While Americans who served in Vietnam
paid a grave price, an extremely conservative estimate of Vietnamese deaths found them to be "proportionally 100 times greater than those suffered by the United States." The military forces of the U.S.-allied Republic of Vietnam reportedly lost more than 254,000 killed and more than 783,000 wounded. And the casualties of the revolutionary forces were evidently far graver— perhaps 1.7 million,
including 1 million killed in battle, plus some 300,000 personnel still
"missing" according to the official but incomplete Vietnamese government
figures.

Horrendous as these numbers may be, they pale in comparison to
the estimated civilian death toll during the war years. At least 65,000
North Vietnamese civilians were killed, mainly from U.S. air raids.
No one will ever know the exact number of South Vietnamese civilians
killed as a result of the American War. While the U.S. military
attempted to quantify almost every other aspect of the conflict— from
the number of helicopter sorties flown to the number of propaganda
leaflets dispersed— it quite deliberately never conducted a comprehensive
study of Vietnamese noncombatant casualties. What ever
civilian casualty statistics the United States did tally were generally
kept secret, and when released piecemeal they were invariably radical
undercounts.

Yet even the available flawed figures are startling, especially given
that the total population of South Vietnam was only about 19 million
people. Using fragmentary data and questionable extrapolations
that, for instance, relied heavily on hospital data yet all but ignored
the immense number of Vietnamese treated by the revolutionary
forces (and also failed to take into account the many civilians killed
by U.S. forces and claimed as enemies), one Department of Defense
statistical analyst came up with a postwar estimate of 1.2 million
civilian casualties, including 195,000 killed. In 1975, a U.S. Senate
subcommittee on refugees and war victims offered an estimate of
1.4 million civilian casualties in South Vietnam, including 415,000
killed. Or take the figures proffered by the political scientist Guenter
Lewy, the progenitor of a revisionist school of Vietnam War history
that invariably shines the best possible light on the U.S. war effort.
Even he posits that there were more than 1.1 million South Vietnamese
civilian casualties, including almost 250,000 killed, as a result of
the conflict.

In recent years, careful surveys, analyses, and official estimates
have consistently pointed toward a significantly higher number of
civilian deaths. The most sophisticated analysis yet of wartime
mortality in Vietnam, a 2008 study by researchers from Harvard
Medical School and the Institute for Health Metrics and Evaluation
at the University of Washington, suggested that a reasonable estimate
might be 3.8 million violent war deaths, combatant and civilian. Given the limitations of the study's methodology, there are good reasons to believe that even this staggering figure may be an underestimate.  Still, the findings lend credence to an official 1995 Vietnamese government estimate of more than 3 million deaths in total— including 2 million civilian deaths— for the years when the Americans were
involved in the conflict.

The sheer number of civilian war wounded, too, has long been a
point of contention. The best numbers currently available, though,
begin to give some sense of the suffering. A brief accounting shows
8,000 to 16,000 South Vietnamese paraplegics; 30,000 to 60,000
South Vietnamese left blind; and some 83,000 to 166,000 South Vietnamese amputees. As far as the total number of the civilian war wounded goes, Guenter Lewy approaches the question by using a ratio derived from South Vietnamese data on military casualties, which shows 2.65 soldiers seriously wounded for every one killed. Such a proportion is distinctly low when applied to the civilian population;
still, even this multiplier, if applied to the Vietnamese government estimate of 2 million civilian dead, yields a figure of 5.3 million civilian wounded, for a total of 7.3 million Vietnamese civilian casualties overall. Notably, official South Vietnamese hospital records indicate that approximately one-third of those wounded were
women and about one-quarter were children under thirteen years of age.

What explains these staggering figures? Because the My Lai massacre
has entered the popular American consciousness as an exceptional,
one-of-a-kind event, the deaths of other civilians during the
Vietnam War tend to be vaguely thought of as a matter of mistakes
or (to use a phrase that would come into common use after the war)
of "collateral damage." But as I came to see, the indiscriminate killing
of South Vietnamese noncombatants— the endless slaughter that
wiped out civilians day after day, month after month, year after year
throughout the Vietnam War— was neither accidental nor unforeseeable.
I stumbled upon the first clues to this hidden history almost by accident,
in June 2001, when I was a graduate student researching posttraumatic
stress disorder among Vietnam veterans. One afternoon, I
was looking through documents at the U.S. National Archives when
a friendly archivist asked me, "Could witnessing war crimes cause
post-traumatic stress?" I had no idea at the time that the archives
might have any records on Vietnam-era war crimes, so the prospect
had never dawned on me. Within an hour or so, though, I held in my
hands the yellowing records of the Vietnam War Crimes Working
Group, a secret Pentagon task force that had been assembled after
the My Lai massacre to ensure that the army would never again be
caught off-guard by a major war crimes scandal.

To call the records a "treasure trove" feels strange, given the nature
of the material. But that's how the collection struck me then, box
after box of criminal investigation reports and day-to-day paperwork
long buried away and almost totally forgotten. There were some
files as thick as a phonebook, with the most detailed and nightmarish
descriptions; other files, paper-thin, hinting at terrible events
that had received no follow-up attention; and just about everything
in between. As I leafed through them that day, I knew one thing
almost instantly: they documented a nightmare war that is essentially
missing from our understanding of the Vietnam conflict.

The War Crimes Working Group files included more than 300 allegations
of massacres, murders, rapes, torture, assaults, mutilations, and other atrocities that were substantiated by army investigators. They detailed the deaths of 137 civilians in mass killings, and 78 smaller-scale attacks in which Vietnamese civilians were killed, wounded, and sexually assaulted. They identified 141 instances in which U.S. troops
used fists, sticks, bats, water torture, and electrical torture on noncombatants. The files also contained 500 allegations that weren't proven at  the time— like the murders of scores, perhaps hundreds, of Vietnamese civilians by the 101st Airborne Division's Tiger Force, which would be confirmed and made public only in 2003.

In hundreds of incident summaries and sworn statements in the
War Crimes Working Group files, veterans laid bare what had
occurred in the backlands of rural Vietnam— the war that Americans
back home didn't see nightly on their televisions or read about
over morning coffee. A sergeant told investigators how he had put a
bullet, point-blank, into the brain of an unarmed boy after gunning
down the youngster's brother; an army ranger matter-of-factly
described slicing the ears off a dead Vietnamese and said that he
planned to continue mutilating corpses. Other files documented
the killing of farmers in their fields and the rape of a child carried
out by an interrogator at an army base. Reading case after case— like
the incident in which a lieutenant "captured two unarmed and unidentified Vietnamese males, estimated ages 2– 3 and 7– 8 years . . . and
killed them for no reason"— I began to get a sense of the ubiquity of
atrocity during the American War.

In the years that followed, with the War Crimes Working Group
documents as an initial guide, I began to track down more information
about little-known or never-revealed Vietnam War crimes. I
located other investigation files at the National Archives, submitted
requests under the Freedom of Information Act, interviewed generals
and top civilian officials, and talked to former military war
crimes investigators. I also spoke with more than one hundred American
veterans across the country, both those who had witnessed
atrocities and others who had personally committed terrible acts.
From them I learned something of what it was like to be twenty years
old, with few life experiences beyond adolescence in a small town or
an inner-city neighborhood, and to be suddenly thrust into villages
of thatch and bamboo homes that seemed ripped straight from the
pages of National Geographic, the paddies around them such a vibrant
green that they almost burned the eye. Veteran after veteran told me
about days of shattering fatigue and the confusion of contradictory
orders, about being placed in situations so alien and unnerving that
even with their automatic rifles and grenades they felt scared walking
through hamlets of unarmed women and children.

Some of the veterans I tried to contact wanted nothing to do with
my questions, almost instantaneously slamming down the phone
receiver. But most were willing to speak to me, and many even seemed
glad to talk to someone who had a sense of the true nature of the war.
In homes from Maryland to California, across kitchen tables and in
marathon four-hour telephone calls, scores of former soldiers and
marines opened up about their experiences. Some had little remorse;
an interrogator who'd tortured prisoners, for instance, told me that
his actions were merely standard operating procedure. Another
veteran, whispering so that his family wouldn't overhear, adamantly
insisted that, though he'd been present at a massacre of civilians, he
hadn't pulled the trigger, no matter what his fellow unit members
said. Then there was the veteran who swore that he knew nothing
about civilians being killed, only to later recount an incident in
which someone in his unit shot an unarmed woman in the back.
And yet another former GI ruefully recounted how, walking through
a Vietnamese village, he had spun around when a local woman chattered
angrily at him (probably complaining about the commotion
that the troops were causing) and driven the butt of his rifle into her
nose. He remembered walking away, laughing, as blood poured from
the woman's face. Decades later, he could no longer imagine how his
nineteen-year-old self had done such a thing, nor could I easily connect
this jovial man to that angry adolescent with a brutal streak.

My conversations with the veterans gave nuance to my under-
standing of the war, bringing human emotion to the sometimes dry
language of military records, and added context to investigation
files that often focused on a single incident. These men also repeatedly
showed me just how incomplete the archives I'd come upon
really were, even though the files detailed hundreds of atrocity allegations.
In one case, for instance, I called a veteran seeking more
information about a sexual assault carried out by members of his
unit, which I found mentioned in one of the files. He offered me
more details about that particular incident but also said that it was
no anomaly. Men from his unit had raped numerous other women as
well, he told me. But neither those assaults nor the random shootings
of farmers by his fellow soldiers had ever been formally investigated.

Among the most poignant of the interviews I conducted was with
Jamie Henry, a former army medic with whom I eventually forged a
friendship. Henry was a whistle-blower in the Ron Ridenhour mold—
the type of man that many want to be but few actually are, a courageous
veteran who spent several years after his return to America
trying to bring to light a series of atrocities committed by his unit.
While many others had kept silent, Henry stepped forward and
reported the crimes he'd seen, taking significant risks for what he
believed was right. He talked to the U.S. Army Criminal Investigation
Command (known as CID), he wrote a detailed article, he spoke
out in public again and again. But the army left him to twist in the
wind, a lone voice repeatedly recounting apparently uncorroborated
tales of shocking violence, while most Americans paid little attention.
Until I sought him out and showed him the documents I'd
found, Henry had no idea that in the early 1970s military investigators
had in fact tracked down and interviewed his fellow unit members,
proving his allegations beyond any doubt — and that the army
had then hidden away this information, never telling him or anyone
else. When he looked over my stacks of photocopies, he was astounded.

Over time, following leads from the veterans I'd spoken to and
from other sources, I discovered additional long-forgotten court-martial
records, investigation files, and related documents in assorted
archives and sometimes in private homes across the country. Paging
through one of these case files, I found myself virtually inhaling
decades-old dust from half a world away. The year was 1970, and a
small U.S. Army patrol had set up an ambush in the jungle near the
Minh Thanh rubber plantation in Binh Long Province, north of Saigon.
Almost immediately the soldiers heard chopping noises, then
branches snapping and Vietnamese voices coming toward them.
Next, a man broke through the brush— he was in uniform, they
would later say, as was the entire group of Vietnamese following
behind him. In an instant, the Americans sprang the ambush, setting
off two Claymore mines— each sending seven hundred small
steel pellets flying more than 150 feet in a lethal sixty-degree arc —
and firing an M-60 machine gun. All but one of the Vietnamese in
the clearing were killed instantly. The unit's radioman immediately
got on his field telephone and called in ten "enemy KIA"— killed in
action.

Later, however, something didn't ring right at headquarters. Despite the claim of ten enemy dead, the Americans had no weapons to show for it. With the My Lai trials garnering headlines back in the United States, the commanding general of the 25th Infantry Division did something unusual: he asked the division's Office of the
Inspector General, whose job it was to probe instances of alleged
misconduct, to investigate. The next day, a lieutenant colonel and his
team arrived at the site of the ambush, where they found the corpses
of five men, three women, and two children scattered on the forest
floor. None was wearing enemy uniforms, and civilian identification
cards were found on the bodies. The closest thing to a weapon was a
piece of paper with "a small drawing of a rifle and of an airplane."
The soldiers who sprang the ambush claimed it was evidence that the
dead were enemy fighters, but the lieutenant colonel noted that it
looked like "something a child would do." Similarly, "the makings of
booby traps" found on the bodies, and cited by the soldiers as evidence
of hostile intent, turned out to be a harmless agricultural tool.

As the American investigators photographed the corpses, it was apparent
that the Vietnamese had been civilians carrying bags of bamboo
shoots and a couple of handfuls of limes— regular people simply trying
to eke out an existence in a war-ravaged landscape.

The lime gatherers' deaths were typical of the kind of operation
that repeatedly wiped out civilians during the Vietnam War. Most of
the time, the noncombatants who died were not herded into a ditch
and gunned down as at My Lai. Instead, the full range of the American
arsenal— from M-16s and Claymore mines to grenades, bombs,
mortars, rockets, napalm, and artillery shells— was unleashed on
forested areas, villages, and homes where perfectly ordinary Vietnamese just happened to live and work.

As the inspector general's report concluded in this particular
incident, the "Vietnamese victims were innocent civilians loyal to
the Republic of Vietnam." Yet, as so often happened, no disciplinary
action of any type was taken against any member of the unit. In fact,
their battalion commander stated that the team performed "exactly
as he expected them to." The battalion's operations officer explained
that the civilians had been in an "off-limits" or free-fire zone, one of
many swaths of the country where everyone was assumed to be the
enemy. Therefore, the soldiers had behaved in accordance with
the U.S. military's directives on the use of lethal force.

It made no difference that the lime gatherers happened to live there,
as their ancestors undoubtedly had for decades, if not centuries,
before them. It made no difference that, as the local province chief of
the U.S.-allied South Vietnamese government told the army, "the
civilians in the area were poor, uneducated and went wherever they
could get food." The inspector general's report pointed out that there
was no written documentation regarding the establishment of a free-fire zone in the area, noting with bureaucratic understatement that "doubt exists" that the program to warn Vietnamese civilians about off-limits areas was "either effective or thorough." But that, too, made no difference. As the final investigation report put it, the platoon
had operated "within its orders which had been given and/or sanctioned by competent authority . . . The rules of engagement were not violated."

Seeking to connect such formal military records with the actual
experience of the ordinary Vietnamese people who had lived through
these events, I made several trips to Vietnam, making my way to
remote rural villages with an interpreter at my side. The jigsaw-puzzle
pieces were not always easy to align. In the files of the War Crimes
Working Group, for example, I located an exceptionally detailed
investigation of a massacre of nearly twenty women and children by
a U.S. Army unit in a tiny hamlet in Quang Nam Province on February
8, 1968. It was clear that the ranking officer there had ordered his
men to "kill anything that moves," and that some of the soldiers had
obeyed. What was less than clear was exactly where "there" was.
With only a general location to go by— fifteen miles west of an old
port town known as Hoi An— we embarked on a shoe-leather search.
Inquiries with locals led us to An Truong, a small hamlet with a
monument to a 1968 massacre. But this particular mass killing took
place on January 9, 1968, rather than in February, and was carried
out by South Korean forces allied to the Americans rather than
by U.S. soldiers themselves. It was not the place we had been looking
for.

After we explained the situation, one of the residents led us to
another village not very far away. It, too, had a memorial— this one
commemorating thirty-three locals who died in three separate massacres
between 1967 and 1970. However, none of these massacres
had taken place on February 8, 1968, either. After interviewing villagers
about these atrocities, we asked if they knew of any other mass
killings in the area. Yes, they said: not the next hamlet down the
road but a little bit beyond it. So on we went. Daylight was rapidly
fading when we arrived in that hamlet and found a monument that
spelled out the basics of the grim story in spare terms: U.S. troops
had killed dozens of Vietnamese there in 1968. Conversations with
the farmers made it clear, though, that these Americans were marines,
not army soldiers, and the massacre had taken place in August. Such
is the nature of investigating war crimes in Vietnam. I'd thought that
I was looking for a needle in a haystack; what I found was a veritable
haystack of needles.

In the United States, meanwhile, the situation in the archives was
often frustratingly the opposite. At one point, a Vietnam veteran
passed on to me a few pages of documents from an investigation into
the killing of civilians by U.S. marines in a small village in the extreme
north of South Vietnam. Those pages provided just enough information
for me to file a Freedom of Information Act request for court-martial
transcripts related to American crimes there. The military's
response to my request was an all too common one: the documents
were inexplicably missing. But the government file was not entirely
empty. Hundreds of pages of trial transcripts, sworn testimony, supporting
documents, and the like had vanished into thin air, but the
military could offer me something in consolation: a copy of the
protective jacket that was once wrapped around the documents. I
declined.

Indeed, an astonishing number of marine court-martial records
of the era have apparently been destroyed or gone missing. Most air
force and navy criminal investigation files that may have existed
seem to have met the same fate. Even before this, the formal investigation
records were an incomplete sample at best; as one veteran of
the secret Pentagon task force told me, knowledge of most cases
never left the battlefield. Still, the War Crimes Working Group files
alone demonstrated that atrocities were committed by members of
every infantry, cavalry, and airborne division, and every separate
brigade that deployed without the rest of its division— that is, every
major army unit in Vietnam.

The scattered, fragmentary nature of the case files makes them
essentially useless for gauging the precise number of war crimes committed
by U.S. personnel in Vietnam.  But the hundreds of reports
that I gathered and the hundreds of witnesses that I interviewed in
the United States and Southeast Asia made it clear that killings of
civilians — whether cold-blooded slaughter like the massacre at My
Lai or the routinely indifferent, wanton bloodshed like the lime gatherers'
ambush in Binh Long — were widespread, routine, and directly
attributable to U.S. command policies.

And such massacres by soldiers and marines, my research showed,
were themselves just a tiny part of the story. For every mass killing
by ground troops that left piles of civilian corpses in a forest clearing
or a drainage ditch, there were exponentially more victims killed by
the everyday exercise of the American way of war from the air.
Throughout South Vietnam, women and children were asphyxiated
or crushed to death when their bunkers collapsed on them, burying
them alive after direct hits from jets' 500-pound bombs or 1,900-
pound shells launched from off-shore ships. Countless others, crazed
with fear, bolted for safety when helicopters swooped toward their
villages, only to have a door gunner cut them in half with bursts from
an M-60 machine gun — and many others, who froze in place, suffered
the same fate. There's only so much killing a squad, a platoon, or a
company can do. Face-to-face atrocities were responsible for just a
fraction of the millions of civilian casualties in South Vietnam.
Matter-of-fact mass killing that dwarfed the slaughter at My Lai normally
involved heavier firepower and command policies that allowed
it to be unleashed with impunity.

This was the real war, the one that barely appears at all in the tens
of thousands of volumes written about Vietnam. This was the war
that Ron Ridenhour spoke about— the one in which My Lai was an
operation, not an aberration. This was the war in which the American
military and successive administrations in Washington produced
not a few random massacres or even discrete strings of atrocities, but
something on the order of thousands of days of relentless misery— a
veritable system of suffering. That system, that machinery of suffering
and what it meant for the Vietnamese people, is what this book
is meant to explain.

Published by Metropolitan Books, an imprint of Henry Holt and Company. Copyright © 2013 by Nick Turse.  Not to be reprinted without permission.

This piece was reprinted by Truthout with permission or license. It may not be reproduced in any form without permission or license from the source.

Nick Turse

Nick Turse is a historian, essayist, investigative journalist, the associate editor of TomDispatch.com, and currently a fellow at Harvard University’s Radcliffe Institute. His latest book is The Case for Withdrawal from Afghanistan.

Related Stories

Drone Memo Puts Brennan on Spot
By Ray McGovern, Consortium News | News Analysis

Hide Comments

blog comments powered by Disqus
GET DAILY TRUTHOUT UPDATES

FOLLOW togtorsstottofb


"Kill Anything That Moves" Military Doctrine Began in Vietnam

Monday, 18 February 2013 00:00 By Nick Turse, Metropolitan Books | Book Excerpt

Kill Anything That Moves.(Image: Metropolitan Books)When did the United States adopt, in the contemporary age, military standards of condoning a "kill anything that moves" doctrine of warfare, along with a widespread use of torture?

One need look no further than the Vietnam War, according to Nick Turse, an author and journalist who has documented the dark side of the US imposition of empire through armed intervention. In this assiduously documented book, Turse offers abundant evidence that My Lai was not an exception to military conduct, but rather, a not uncommon occurrence. In addition, the US slaughtered countless civilians in air and ground attacks without ever even seeing who was being killed.

In addition, the reader will also discover that Dick Cheney's backing of torture had ample precedent during the Vietnam War.-MK

Support Truthout's mission. Kill Anything That Moves (hardcover edition) is yours with a minimum donation to Truthout of $35 (which includes shipping and handling) or a monthly donation of $15.

The following excerpt is the introduction to Kill Anything That Moves: The Real American War in Vietnam.

"An Operation, Not an Aberration"

On January 21, 1971, a Vietnam veteran named Charles McDuff
wrote a letter to President Richard Nixon to voice his disgust with the
American war in Southeast Asia. McDuff had witnessed multiple cases
of Vietnamese civilians being abused and killed by American soldiers
and their allies, and he had found the U.S. military justice system to be
woefully ineffective in punishing wrongdoers. "Maybe your advisors
have not clued you in," he told the president, "but the atrocities that
were committed in My Lai are eclipsed by similar American actions
throughout the country." His three-page handwritten missive concluded
with an impassioned plea to Nixon to end American participation
in the war.

The White House forwarded the note to the Department of
Defense for a reply, and within a few weeks Major General Franklin
Davis Jr., the army's director of military personnel policies, wrote
back to McDuff. It was "indeed unfortunate," said Davis, "that some
incidents occur within combat zones." He then shifted the burden of
responsibility for what had happened firmly back onto the veteran.
"I presume," he wrote, "that you promptly reported such actions to
the proper authorities." Other than a paragraph of information on how
to contact the U.S. Army criminal investigators, the reply was only
four sentences long and included a matter-of- fact reassurance: "The
United States Army has never condoned wanton killing or disregard
for human life."

This was, and remains, the American military's official position.
In many ways, it remains the popular understanding in the United
States as a whole. Today, histories of the Vietnam War regularly discuss
war crimes or civilian suffering only in the context of a single
incident: the My Lai massacre cited by McDuff. Even as that one
event has become the subject of numerous books and articles, all the
other atrocities perpetrated by U.S. soldiers have essentially vanished
from popular memory.

The visceral horror of what happened at My Lai is undeniable. On
the evening of March 15, 1968, members of the Americal Division's
Charlie Company, 1st Battalion, 20th Infantry, were briefed by their
commanding officer, Captain Ernest Medina, on a planned operation
the next day in an area they knew as "Pinkville." As unit member
Harry Stanley recalled, Medina "ordered us to 'kill everything in the
village.' " Infantryman Salvatore LaMartina remembered Medina's
words only slightly differently: they were to "kill everything that
breathed." What stuck in artillery forward observer James Flynn's
mind was a question one of the other soldiers asked: "Are we supposed
to kill women and children?" And Medina's reply: "Kill everything
that moves."

The next morning, the troops clambered aboard helicopters and
were airlifted into what they thought would be a "hot LZ"— a landing
zone where they'd be under hostile fire. As it happened, though,
instead of finding Vietnamese adversaries spoiling for a fight, the
Americans entering My Lai encountered only civilians: women, children,
and old men. Many were still cooking their breakfast rice. Nevertheless,
Medina's orders were followed to a T. Soldiers of Charlie
Company killed. They killed everything. They killed everything that
moved.

Advancing in small squads, the men of the unit shot chickens as
they scurried about, pigs as they bolted, and cows and water buffalo
lowing among the thatch-roofed houses. They gunned down old
men sitting in their homes and children as they ran for cover. They
tossed grenades into homes without even bothering to look inside. An
officer grabbed a woman by the hair and shot her point-blank with a
pistol. A woman who came out of her home with a baby in her arms
was shot down on the spot. As the tiny child hit the ground, another
GI opened up on the infant with his M-16 automatic rifle.
Over four hours, members of Charlie Company methodically
slaughtered more than five hundred unarmed victims, killing some
in ones and twos, others in small groups, and collecting many more
in a drainage ditch that would become an infamous killing ground.
They faced no opposition. They even took a quiet break to eat lunch
in the midst of the carnage. Along the way, they also raped women
and young girls, mutilated the dead, systematically burned homes,
and fouled the area's drinking water.

There were scores of witnesses on the ground and still more overhead,
American officers and helicopter crewmen perfectly capable of
seeing the growing piles of civilian bodies. Yet when the military
released the first news of the assault, it was portrayed as a victory over
a formidable enemy force, a legitimate battle in which 128 enemy
troops were killed without the loss of a single American life. In a
routine congratulatory telegram, General William Westmoreland, the
commander of U.S. forces in Vietnam, lauded the "heavy blows"
inflicted on the enemy. His protégé, the commander of the Americal
Division, added a special note praising Charlie Company's "aggressiveness."

Despite communiqués, radio reports, and English-language
accounts released by the Vietnamese revolutionary forces, the My
Lai massacre would remain, to the outside world, an American victory
for more than a year. And the truth might have remained hidden
forever if not for the perseverance of a single Vietnam veteran
named Ron Ridenhour. The twenty-two-year-old Ridenhour had not
been among the hundred American troops at My Lai, though he had
seen civilians murdered elsewhere in Vietnam; instead, he heard
about the slaughter from other soldiers who had been in Pinkville
that day. Unnerved, Ridenhour took the unprecedented step of carefully
gathering testimony from multiple American eyewitnesses. Then,
upon returning to the United States after his yearlong tour of duty,
he committed himself to doing what ever was necessary to expose
the incident to public scrutiny.

Ridenhour's efforts were helped by the painstaking investigative
reporting of Seymour Hersh, who published newspaper articles about
the massacre; by the appearance in Life magazine of grisly full-color
images that army photographer Ron Haeberle captured in My Lai as
the slaughter was unfolding; and by a confessional interview that a
soldier from Charlie Company gave to CBS News. The Pentagon, for
its part, consistently fought to minimize what had happened, claiming
that reports by Vietnamese survivors were wildly exaggerated. At
the same time, the military focused its attention on the lowest ranking
officer who could conceivably shoulder the blame for such a
nightmare: Charlie Company's Lieutenant William Calley.

An army inquiry into the killings eventually determined that
thirty individuals were involved in criminal misconduct during the
massacre or its cover-up. Twenty-eight of them were officers, including
two generals, and the inquiry concluded they had committed a
total of 224 serious offenses. But only Calley was ever convicted of
any wrongdoing. He was sentenced to life in prison for the premeditated
murder of twenty-two civilians, but President Nixon freed him
from prison and allowed him to remain under house arrest. He was
eventually paroled after serving just forty months, most of it in the
comfort of his own quarters.

The public response generally followed the official one. Twenty five years later, Ridenhour would sum it up this way. At the end of it, if you ask people what happened at My Lai, they would say: "Oh yeah, isn't that where Lieutenant Calley went crazy
and killed all those people?" No, that was not what happened. Lieutenant
Calley was one of the people who went crazy and killed a lot of
people at My Lai, but this was an operation, not an aberration.

Looking back, it's clear that the real aberration was the unprecedented
and unparalleled investigation and exposure of My Lai. No
other American atrocity committed during the war— and there were
so many— was ever afforded anything approaching the same attention.
Most, of course, weren't photographed, and many were not
documented in any way. The great majority were never known outside
the offending unit, and most investigations that did result were
closed, quashed, or abandoned. Even on the rare occasions when the
allegations were seriously investigated within the military, the reports
were soon buried in classified files without ever seeing the light of
day. Whistle-blowers within the ranks or recently out of the army
were threatened, intimidated, smeared, or— if they were lucky—
simply marginalized and ignored.

Until the My Lai revelations became front-page news, atrocity
stories were routinely disregarded by American journalists or excised
by stateside editors. The fate of civilians in rural South Vietnam did
not merit much examination; even the articles that did mention the
killing of noncombatants generally did so merely in passing, without
any indication that the acts described might be war crimes. Vietnamese revolutionary sources, for their part, detailed hundreds of massacres and large-scale operations that resulted in thousands of civilian deaths, but those reports were dismissed out of hand as communist propaganda.

And then, in a stunning reversal, almost immediately after the
exposure of the My Lai massacre, war crime allegations became old
hat— so commonplace as to be barely worth mentioning or looking
into. In leaflets, pamphlets, small-press books, and "underground"
newspapers, the growing American antiwar movement repeatedly
pointed out that U.S. troops were committing atrocities on a regular
basis. But what had been previously brushed aside as propaganda
and leftist kookery suddenly started to be disregarded as yawn-worthy
common knowledge, with little but the My Lai massacre in between.

Such impulses only grew stronger in the years of the "culture wars,"
when the Republican Party and an emboldened right wing rose to
power. Until Ronald Reagan's presidency, the Vietnam War was generally
seen as an American defeat, but even before taking office Reagan
began rebranding the conflict as "a noble cause." In the same
spirit, scholars and veterans began, with significant success, to recast
the war in rosier terms.  Even in the early years of the twenty-first
century, as newspapers and magazines published exposés of long hidden
U.S. atrocities, apologist historians continued to ignore much
of the evidence, portraying American war crimes as no more than
isolated incidents.

But the stunning scale of civilian suffering in Vietnam is far
beyond anything that can be explained as merely the work of some
"bad apples," however numerous. Murder, torture, rape, abuse, forced
displacement, home burnings, specious arrests, imprisonment without
due process — such occurrences were virtually a daily fact of life
throughout the years of the American presence in Vietnam. And as
Ridenhour put it, they were no aberration. Rather, they were the
inevitable outcome of deliberate policies, dictated at the highest levels
of the military.

The first official American combat troops arrived in Vietnam in
1965, but the roots of the conflict go back many decades earlier. In
the nineteenth century, France expanded its colonial empire by taking
control of Vietnam as well as neighboring Cambodia and Laos,
rechristening the entire region as French Indochina. French rubber
production in Vietnam yielded such riches for the colonizers that
the latex oozing from rubber trees became known as "white gold."

The ill-paid Vietnamese workers, laboring on the plantations in
harsh conditions, called it by a different name: "white blood."

By the early twentieth century, anger at the French had developed
into a nationalist movement for independence. Its leaders found
inspiration in communism, specifically the example of Russian Bolshevism
and Lenin's call for national revolutions in the colonial
world. During World War II, when Vietnam was occupied by the
imperial Japanese, the country's main anti-colonial organization—
officially called the League for the Independence of Vietnam, but far
better known as the Viet Minh— launched a guerrilla war against
the Japanese forces and the French administrators running the country.
Under the leadership of the charismatic Ho Chi Minh, the Vietnamese guerrillas aided the American war effort. In return they received arms, training, and support from the U.S. Office of Strategic Services, a forerunner of the Central Intelligence Agency.

In 1945, with the Japanese defeated, Ho proclaimed Vietnam's
Independence, using the words of the U.S. Declaration of Independence
as his template. "All men are created equal," he told a crowd of
half a million Vietnamese in Hanoi. "The Creator has given us certain
inviolable rights: the right to life, the right to be free, and the
right to achieve happiness." As a young man Ho had spent some years
living in the West, reportedly including stretches in Boston and New
York City, and he hoped to obtain American support for his vision of
a free Vietnam. In the aftermath of World War II, however, the
United States was focused on rebuilding and strengthening a devastated
Europe, as the Cold War increasingly gripped the continent.

The Americans saw France as a strong ally against any Soviet designs
on Western Europe and thus had little interest in sanctioning a
communist-led independence movement in a former French colony.
Instead, U.S. ships helped transport French troops to Vietnam, and
the administration of President Harry Truman threw its support
behind a French reconquest of Indochina.

Soon, the United States was dispatching equipment and even
military advisers to Vietnam. By 1953, it was shouldering nearly 80
percent of the bill for an ever more bitter war against the Viet Minh.
The conflict progressed from guerrilla warfare to a conventional military
campaign, and in 1954 a Gallic garrison at the well-fortified
base of Dien Bien Phu was pounded into surrender by Viet Minh
forces under General Vo Nguyen Giap. The French had had enough.
At an international peace conference in Geneva, they agreed to a temporary
separation of Vietnam into two placeholder regions, the north
and the south, which were to be rejoined as one nation following a
reunification election in 1956.

That election never took place. Fearing that Ho Chi Minh, now
the head of the communist Democratic Republic of Vietnam in the
north, was sure to sweep any nationwide vote, the United States picked
up where its French partners had left off. It promptly launched efforts
to thwart reunification by arming its allies in the southern part of
the country. In this way, it fostered the creation of what eventually
became the Republic of Vietnam, led by a Catholic autocrat named
Ngo Dinh Diem.

From the 1950s on, the United States would support an ever more
corrupt and repressive state in South Vietnam while steadily expanding
its presence in Southeast Asia. When President John Kennedy
took office there were around 800 U.S. military personnel in South
Vietnam. That number increased to 3,000 in 1961, and to more than
11,000 the following year. Officially listed as advisers involved in the
training of the South Vietnamese army, the Americans increasingly
took part in combat operations against southern guerrillas— both
communist and noncommunist— who were now waging war to unify
the country.

After Kennedy's assassination, President Lyndon Johnson repeatedly
escalated the war with bombing raids on North Vietnam, and
unleashed an ever more furious onslaught on the South. In 1965 the
fiction of "advisers" was finally dropped, and the American War, as
it is known in Vietnam, began in earnest. In a televised speech, John-
son insisted that the United States was not inserting itself into a faraway
civil war but taking steps to contain a communist menace. The
war, he said, was "guided by North Vietnam . . . Its goal is to conquer
the South, to defeat American power, and to extend the Asiatic
dominion of communism." To counter this, the United States turned
huge swaths of the South Vietnamese countryside— where most of
South Vietnam's population lived— into battered battlegrounds.

At the peak of U.S. operations, in 1969, the war involved more
than 540,000 American troops in Vietnam, plus some 100,000 to
200,000 U.S. troops participating in the effort from outside the
country. They were also aided by numerous CIA operatives, civilian
advisers, mercenaries, civilian contractors, and armed members of
the allied "Free World Forces"— South Korean, Australian, New
Zealand, Thai, Filipino, and other foreign troops. Over the entire
course of the conflict, the United States would deploy more than 3
million soldiers, marines, airmen, and sailors to Southeast Asia.
(Fighting alongside them were hundreds of thousands of South Vietnamese troops: the Army of the Republic of Vietnam would balloon to a force of nearly 1 million before the end of the war, to say nothing of South Vietnam's air force, navy, marine corps, and national police.) Officially, the American military effort lasted until early 1973, when a cease-fire was signed and U.S. combat forces were formally withdrawn from the country, though American aid and other support
would continue to flow into the Republic of Vietnam until Saigon fell
to the revolutionary forces in 1975.

From the U.S. perspective, the enemy was composed of two distinct
groups: members of the North Vietnamese army and indigenous
South Vietnamese fighters loyal to the National Liberation Front, the
revolutionary organization that succeeded the Viet Minh and opposed
the U.S.-allied Saigon government. The NLF's combatants, officially
known as the People's Liberation Armed Forces (PLAF), included guerrillas
in peasant clothing as well as uniformed troops organized into
professionalized units. The U.S. Information Service invented the
moniker "Viet Cong"— that is, Vietnamese Communists—as a derogatory
term that covered anyone fighting on the side of the NLF, though
many of the guerrillas themselves were driven more by nationalism
than by communist ideology. American soldiers, in turn, often shortened
this label to "the Cong" or "VC," or, owing to the military's phonetic
Alpha-Bravo-Charlie alphabet, to "Victor Charlie" or simply
"Charlie."

By 1968 the U.S. forces and their allies in the South were opposed
by an estimated 50,000 North Vietnamese troops plus 60,000
uniformed PLAF soldiers, while the revolutionaries' paramilitary
forces—part-time, local guerrillas— likely reached into the hundreds
of thousands. Americans often made hard-and-fast distinctions
between the well-armed, green- or khaki-uniformed North Vietnamese troops with their fabric-covered, pressed-cardboard pith style helmets; the khaki-clad main force PLAF soldiers, with their floppy cloth "boonie hats"; and the lightly armed, "black pajama"–clad guerrillas (all of whom actually wore a wide variety of types and
colors of clothing depending on the time and place). In reality,
though, they were very hard to disentangle, since North Vietnamese
troops reinforced PLAF units, "local" VC fought in tandem with
"hard-core" professionalized PLAF troops, and part-time farmer fighters assisted uniformed North Vietnamese forces.

The plethora of designations and the often hazy distinctions
between them underscore the fact that the Americans never really
grasped who the enemy was. On one hand, they claimed the VC had
little popular support and held sway over villages only through terror
tactics. On the other, American soldiers who were supposedly
engaged in countering communist aggression to protect the South
Vietnamese readily killed civilians because they assumed that most
villagers either were in league with the enemy or were guerrillas
themselves once the sun went down.

The United States never wanted to admit that the conflict might
be a true "people's war," and that Vietnamese were bound to the revolution because they saw it as a fight for their families, their land, and their country. In the villages of South Vietnam, Vietnamese nationalists had long organized themselves to resist foreign domination, and it was no different when the Americans came. By then, the local population was often inextricably joined to the liberation struggle.
Lacking advanced technology, financial resources, or significant firepower,
America's Vietnamese enemies maximized assets like concealment,
local knowledge, popular support, and something less quantifiable— call it patriotism or nationalism, or perhaps a hope and a dream.

Of course, not every Vietnamese villager believed in the revolution
or saw it as the best expression of nationalist patriotism. Even
villages in revolutionary strongholds were home to some supporters
of the Saigon government. And many more farmers simply wanted
nothing to do with the conflict or abstract notions like nationalism
and communism. They worried mainly about their next rice crop,
their animals, their house and children. But bombs and napalm don't
discriminate. As gunships and howitzers ravaged the landscape, as
soldiers with M-16 rifles and M-79 grenade launchers swept through
the countryside, Vietnamese villagers of every type— supporters of
the revolution, sympathizers of the Saigon regime, and those who
merely wanted to be left alone— all perished in vast numbers.
The war's casualty figures are staggering indeed. From 1955 to
1975, the United States lost more than 58,000 military personnel in
Southeast Asia. Its troops were wounded around 304,000 times, with
153,000 cases serious enough to require hospitalization, and 75,000
veterans left severely disabled. While Americans who served in Vietnam
paid a grave price, an extremely conservative estimate of Vietnamese deaths found them to be "proportionally 100 times greater than those suffered by the United States." The military forces of the U.S.-allied Republic of Vietnam reportedly lost more than 254,000 killed and more than 783,000 wounded. And the casualties of the revolutionary forces were evidently far graver— perhaps 1.7 million,
including 1 million killed in battle, plus some 300,000 personnel still
"missing" according to the official but incomplete Vietnamese government
figures.

Horrendous as these numbers may be, they pale in comparison to
the estimated civilian death toll during the war years. At least 65,000
North Vietnamese civilians were killed, mainly from U.S. air raids.
No one will ever know the exact number of South Vietnamese civilians
killed as a result of the American War. While the U.S. military
attempted to quantify almost every other aspect of the conflict— from
the number of helicopter sorties flown to the number of propaganda
leaflets dispersed— it quite deliberately never conducted a comprehensive
study of Vietnamese noncombatant casualties. What ever
civilian casualty statistics the United States did tally were generally
kept secret, and when released piecemeal they were invariably radical
undercounts.

Yet even the available flawed figures are startling, especially given
that the total population of South Vietnam was only about 19 million
people. Using fragmentary data and questionable extrapolations
that, for instance, relied heavily on hospital data yet all but ignored
the immense number of Vietnamese treated by the revolutionary
forces (and also failed to take into account the many civilians killed
by U.S. forces and claimed as enemies), one Department of Defense
statistical analyst came up with a postwar estimate of 1.2 million
civilian casualties, including 195,000 killed. In 1975, a U.S. Senate
subcommittee on refugees and war victims offered an estimate of
1.4 million civilian casualties in South Vietnam, including 415,000
killed. Or take the figures proffered by the political scientist Guenter
Lewy, the progenitor of a revisionist school of Vietnam War history
that invariably shines the best possible light on the U.S. war effort.
Even he posits that there were more than 1.1 million South Vietnamese
civilian casualties, including almost 250,000 killed, as a result of
the conflict.

In recent years, careful surveys, analyses, and official estimates
have consistently pointed toward a significantly higher number of
civilian deaths. The most sophisticated analysis yet of wartime
mortality in Vietnam, a 2008 study by researchers from Harvard
Medical School and the Institute for Health Metrics and Evaluation
at the University of Washington, suggested that a reasonable estimate
might be 3.8 million violent war deaths, combatant and civilian. Given the limitations of the study's methodology, there are good reasons to believe that even this staggering figure may be an underestimate.  Still, the findings lend credence to an official 1995 Vietnamese government estimate of more than 3 million deaths in total— including 2 million civilian deaths— for the years when the Americans were
involved in the conflict.

The sheer number of civilian war wounded, too, has long been a
point of contention. The best numbers currently available, though,
begin to give some sense of the suffering. A brief accounting shows
8,000 to 16,000 South Vietnamese paraplegics; 30,000 to 60,000
South Vietnamese left blind; and some 83,000 to 166,000 South Vietnamese amputees. As far as the total number of the civilian war wounded goes, Guenter Lewy approaches the question by using a ratio derived from South Vietnamese data on military casualties, which shows 2.65 soldiers seriously wounded for every one killed. Such a proportion is distinctly low when applied to the civilian population;
still, even this multiplier, if applied to the Vietnamese government estimate of 2 million civilian dead, yields a figure of 5.3 million civilian wounded, for a total of 7.3 million Vietnamese civilian casualties overall. Notably, official South Vietnamese hospital records indicate that approximately one-third of those wounded were
women and about one-quarter were children under thirteen years of age.

What explains these staggering figures? Because the My Lai massacre
has entered the popular American consciousness as an exceptional,
one-of-a-kind event, the deaths of other civilians during the
Vietnam War tend to be vaguely thought of as a matter of mistakes
or (to use a phrase that would come into common use after the war)
of "collateral damage." But as I came to see, the indiscriminate killing
of South Vietnamese noncombatants— the endless slaughter that
wiped out civilians day after day, month after month, year after year
throughout the Vietnam War— was neither accidental nor unforeseeable.
I stumbled upon the first clues to this hidden history almost by accident,
in June 2001, when I was a graduate student researching posttraumatic
stress disorder among Vietnam veterans. One afternoon, I
was looking through documents at the U.S. National Archives when
a friendly archivist asked me, "Could witnessing war crimes cause
post-traumatic stress?" I had no idea at the time that the archives
might have any records on Vietnam-era war crimes, so the prospect
had never dawned on me. Within an hour or so, though, I held in my
hands the yellowing records of the Vietnam War Crimes Working
Group, a secret Pentagon task force that had been assembled after
the My Lai massacre to ensure that the army would never again be
caught off-guard by a major war crimes scandal.

To call the records a "treasure trove" feels strange, given the nature
of the material. But that's how the collection struck me then, box
after box of criminal investigation reports and day-to-day paperwork
long buried away and almost totally forgotten. There were some
files as thick as a phonebook, with the most detailed and nightmarish
descriptions; other files, paper-thin, hinting at terrible events
that had received no follow-up attention; and just about everything
in between. As I leafed through them that day, I knew one thing
almost instantly: they documented a nightmare war that is essentially
missing from our understanding of the Vietnam conflict.

The War Crimes Working Group files included more than 300 allegations
of massacres, murders, rapes, torture, assaults, mutilations, and other atrocities that were substantiated by army investigators. They detailed the deaths of 137 civilians in mass killings, and 78 smaller-scale attacks in which Vietnamese civilians were killed, wounded, and sexually assaulted. They identified 141 instances in which U.S. troops
used fists, sticks, bats, water torture, and electrical torture on noncombatants. The files also contained 500 allegations that weren't proven at  the time— like the murders of scores, perhaps hundreds, of Vietnamese civilians by the 101st Airborne Division's Tiger Force, which would be confirmed and made public only in 2003.

In hundreds of incident summaries and sworn statements in the
War Crimes Working Group files, veterans laid bare what had
occurred in the backlands of rural Vietnam— the war that Americans
back home didn't see nightly on their televisions or read about
over morning coffee. A sergeant told investigators how he had put a
bullet, point-blank, into the brain of an unarmed boy after gunning
down the youngster's brother; an army ranger matter-of-factly
described slicing the ears off a dead Vietnamese and said that he
planned to continue mutilating corpses. Other files documented
the killing of farmers in their fields and the rape of a child carried
out by an interrogator at an army base. Reading case after case— like
the incident in which a lieutenant "captured two unarmed and unidentified Vietnamese males, estimated ages 2– 3 and 7– 8 years . . . and
killed them for no reason"— I began to get a sense of the ubiquity of
atrocity during the American War.

In the years that followed, with the War Crimes Working Group
documents as an initial guide, I began to track down more information
about little-known or never-revealed Vietnam War crimes. I
located other investigation files at the National Archives, submitted
requests under the Freedom of Information Act, interviewed generals
and top civilian officials, and talked to former military war
crimes investigators. I also spoke with more than one hundred American
veterans across the country, both those who had witnessed
atrocities and others who had personally committed terrible acts.
From them I learned something of what it was like to be twenty years
old, with few life experiences beyond adolescence in a small town or
an inner-city neighborhood, and to be suddenly thrust into villages
of thatch and bamboo homes that seemed ripped straight from the
pages of National Geographic, the paddies around them such a vibrant
green that they almost burned the eye. Veteran after veteran told me
about days of shattering fatigue and the confusion of contradictory
orders, about being placed in situations so alien and unnerving that
even with their automatic rifles and grenades they felt scared walking
through hamlets of unarmed women and children.

Some of the veterans I tried to contact wanted nothing to do with
my questions, almost instantaneously slamming down the phone
receiver. But most were willing to speak to me, and many even seemed
glad to talk to someone who had a sense of the true nature of the war.
In homes from Maryland to California, across kitchen tables and in
marathon four-hour telephone calls, scores of former soldiers and
marines opened up about their experiences. Some had little remorse;
an interrogator who'd tortured prisoners, for instance, told me that
his actions were merely standard operating procedure. Another
veteran, whispering so that his family wouldn't overhear, adamantly
insisted that, though he'd been present at a massacre of civilians, he
hadn't pulled the trigger, no matter what his fellow unit members
said. Then there was the veteran who swore that he knew nothing
about civilians being killed, only to later recount an incident in
which someone in his unit shot an unarmed woman in the back.
And yet another former GI ruefully recounted how, walking through
a Vietnamese village, he had spun around when a local woman chattered
angrily at him (probably complaining about the commotion
that the troops were causing) and driven the butt of his rifle into her
nose. He remembered walking away, laughing, as blood poured from
the woman's face. Decades later, he could no longer imagine how his
nineteen-year-old self had done such a thing, nor could I easily connect
this jovial man to that angry adolescent with a brutal streak.

My conversations with the veterans gave nuance to my under-
standing of the war, bringing human emotion to the sometimes dry
language of military records, and added context to investigation
files that often focused on a single incident. These men also repeatedly
showed me just how incomplete the archives I'd come upon
really were, even though the files detailed hundreds of atrocity allegations.
In one case, for instance, I called a veteran seeking more
information about a sexual assault carried out by members of his
unit, which I found mentioned in one of the files. He offered me
more details about that particular incident but also said that it was
no anomaly. Men from his unit had raped numerous other women as
well, he told me. But neither those assaults nor the random shootings
of farmers by his fellow soldiers had ever been formally investigated.

Among the most poignant of the interviews I conducted was with
Jamie Henry, a former army medic with whom I eventually forged a
friendship. Henry was a whistle-blower in the Ron Ridenhour mold—
the type of man that many want to be but few actually are, a courageous
veteran who spent several years after his return to America
trying to bring to light a series of atrocities committed by his unit.
While many others had kept silent, Henry stepped forward and
reported the crimes he'd seen, taking significant risks for what he
believed was right. He talked to the U.S. Army Criminal Investigation
Command (known as CID), he wrote a detailed article, he spoke
out in public again and again. But the army left him to twist in the
wind, a lone voice repeatedly recounting apparently uncorroborated
tales of shocking violence, while most Americans paid little attention.
Until I sought him out and showed him the documents I'd
found, Henry had no idea that in the early 1970s military investigators
had in fact tracked down and interviewed his fellow unit members,
proving his allegations beyond any doubt — and that the army
had then hidden away this information, never telling him or anyone
else. When he looked over my stacks of photocopies, he was astounded.

Over time, following leads from the veterans I'd spoken to and
from other sources, I discovered additional long-forgotten court-martial
records, investigation files, and related documents in assorted
archives and sometimes in private homes across the country. Paging
through one of these case files, I found myself virtually inhaling
decades-old dust from half a world away. The year was 1970, and a
small U.S. Army patrol had set up an ambush in the jungle near the
Minh Thanh rubber plantation in Binh Long Province, north of Saigon.
Almost immediately the soldiers heard chopping noises, then
branches snapping and Vietnamese voices coming toward them.
Next, a man broke through the brush— he was in uniform, they
would later say, as was the entire group of Vietnamese following
behind him. In an instant, the Americans sprang the ambush, setting
off two Claymore mines— each sending seven hundred small
steel pellets flying more than 150 feet in a lethal sixty-degree arc —
and firing an M-60 machine gun. All but one of the Vietnamese in
the clearing were killed instantly. The unit's radioman immediately
got on his field telephone and called in ten "enemy KIA"— killed in
action.

Later, however, something didn't ring right at headquarters. Despite the claim of ten enemy dead, the Americans had no weapons to show for it. With the My Lai trials garnering headlines back in the United States, the commanding general of the 25th Infantry Division did something unusual: he asked the division's Office of the
Inspector General, whose job it was to probe instances of alleged
misconduct, to investigate. The next day, a lieutenant colonel and his
team arrived at the site of the ambush, where they found the corpses
of five men, three women, and two children scattered on the forest
floor. None was wearing enemy uniforms, and civilian identification
cards were found on the bodies. The closest thing to a weapon was a
piece of paper with "a small drawing of a rifle and of an airplane."
The soldiers who sprang the ambush claimed it was evidence that the
dead were enemy fighters, but the lieutenant colonel noted that it
looked like "something a child would do." Similarly, "the makings of
booby traps" found on the bodies, and cited by the soldiers as evidence
of hostile intent, turned out to be a harmless agricultural tool.

As the American investigators photographed the corpses, it was apparent
that the Vietnamese had been civilians carrying bags of bamboo
shoots and a couple of handfuls of limes— regular people simply trying
to eke out an existence in a war-ravaged landscape.

The lime gatherers' deaths were typical of the kind of operation
that repeatedly wiped out civilians during the Vietnam War. Most of
the time, the noncombatants who died were not herded into a ditch
and gunned down as at My Lai. Instead, the full range of the American
arsenal— from M-16s and Claymore mines to grenades, bombs,
mortars, rockets, napalm, and artillery shells— was unleashed on
forested areas, villages, and homes where perfectly ordinary Vietnamese just happened to live and work.

As the inspector general's report concluded in this particular
incident, the "Vietnamese victims were innocent civilians loyal to
the Republic of Vietnam." Yet, as so often happened, no disciplinary
action of any type was taken against any member of the unit. In fact,
their battalion commander stated that the team performed "exactly
as he expected them to." The battalion's operations officer explained
that the civilians had been in an "off-limits" or free-fire zone, one of
many swaths of the country where everyone was assumed to be the
enemy. Therefore, the soldiers had behaved in accordance with
the U.S. military's directives on the use of lethal force.

It made no difference that the lime gatherers happened to live there,
as their ancestors undoubtedly had for decades, if not centuries,
before them. It made no difference that, as the local province chief of
the U.S.-allied South Vietnamese government told the army, "the
civilians in the area were poor, uneducated and went wherever they
could get food." The inspector general's report pointed out that there
was no written documentation regarding the establishment of a free-fire zone in the area, noting with bureaucratic understatement that "doubt exists" that the program to warn Vietnamese civilians about off-limits areas was "either effective or thorough." But that, too, made no difference. As the final investigation report put it, the platoon
had operated "within its orders which had been given and/or sanctioned by competent authority . . . The rules of engagement were not violated."

Seeking to connect such formal military records with the actual
experience of the ordinary Vietnamese people who had lived through
these events, I made several trips to Vietnam, making my way to
remote rural villages with an interpreter at my side. The jigsaw-puzzle
pieces were not always easy to align. In the files of the War Crimes
Working Group, for example, I located an exceptionally detailed
investigation of a massacre of nearly twenty women and children by
a U.S. Army unit in a tiny hamlet in Quang Nam Province on February
8, 1968. It was clear that the ranking officer there had ordered his
men to "kill anything that moves," and that some of the soldiers had
obeyed. What was less than clear was exactly where "there" was.
With only a general location to go by— fifteen miles west of an old
port town known as Hoi An— we embarked on a shoe-leather search.
Inquiries with locals led us to An Truong, a small hamlet with a
monument to a 1968 massacre. But this particular mass killing took
place on January 9, 1968, rather than in February, and was carried
out by South Korean forces allied to the Americans rather than
by U.S. soldiers themselves. It was not the place we had been looking
for.

After we explained the situation, one of the residents led us to
another village not very far away. It, too, had a memorial— this one
commemorating thirty-three locals who died in three separate massacres
between 1967 and 1970. However, none of these massacres
had taken place on February 8, 1968, either. After interviewing villagers
about these atrocities, we asked if they knew of any other mass
killings in the area. Yes, they said: not the next hamlet down the
road but a little bit beyond it. So on we went. Daylight was rapidly
fading when we arrived in that hamlet and found a monument that
spelled out the basics of the grim story in spare terms: U.S. troops
had killed dozens of Vietnamese there in 1968. Conversations with
the farmers made it clear, though, that these Americans were marines,
not army soldiers, and the massacre had taken place in August. Such
is the nature of investigating war crimes in Vietnam. I'd thought that
I was looking for a needle in a haystack; what I found was a veritable
haystack of needles.

In the United States, meanwhile, the situation in the archives was
often frustratingly the opposite. At one point, a Vietnam veteran
passed on to me a few pages of documents from an investigation into
the killing of civilians by U.S. marines in a small village in the extreme
north of South Vietnam. Those pages provided just enough information
for me to file a Freedom of Information Act request for court-martial
transcripts related to American crimes there. The military's
response to my request was an all too common one: the documents
were inexplicably missing. But the government file was not entirely
empty. Hundreds of pages of trial transcripts, sworn testimony, supporting
documents, and the like had vanished into thin air, but the
military could offer me something in consolation: a copy of the
protective jacket that was once wrapped around the documents. I
declined.

Indeed, an astonishing number of marine court-martial records
of the era have apparently been destroyed or gone missing. Most air
force and navy criminal investigation files that may have existed
seem to have met the same fate. Even before this, the formal investigation
records were an incomplete sample at best; as one veteran of
the secret Pentagon task force told me, knowledge of most cases
never left the battlefield. Still, the War Crimes Working Group files
alone demonstrated that atrocities were committed by members of
every infantry, cavalry, and airborne division, and every separate
brigade that deployed without the rest of its division— that is, every
major army unit in Vietnam.

The scattered, fragmentary nature of the case files makes them
essentially useless for gauging the precise number of war crimes committed
by U.S. personnel in Vietnam.  But the hundreds of reports
that I gathered and the hundreds of witnesses that I interviewed in
the United States and Southeast Asia made it clear that killings of
civilians — whether cold-blooded slaughter like the massacre at My
Lai or the routinely indifferent, wanton bloodshed like the lime gatherers'
ambush in Binh Long — were widespread, routine, and directly
attributable to U.S. command policies.

And such massacres by soldiers and marines, my research showed,
were themselves just a tiny part of the story. For every mass killing
by ground troops that left piles of civilian corpses in a forest clearing
or a drainage ditch, there were exponentially more victims killed by
the everyday exercise of the American way of war from the air.
Throughout South Vietnam, women and children were asphyxiated
or crushed to death when their bunkers collapsed on them, burying
them alive after direct hits from jets' 500-pound bombs or 1,900-
pound shells launched from off-shore ships. Countless others, crazed
with fear, bolted for safety when helicopters swooped toward their
villages, only to have a door gunner cut them in half with bursts from
an M-60 machine gun — and many others, who froze in place, suffered
the same fate. There's only so much killing a squad, a platoon, or a
company can do. Face-to-face atrocities were responsible for just a
fraction of the millions of civilian casualties in South Vietnam.
Matter-of-fact mass killing that dwarfed the slaughter at My Lai normally
involved heavier firepower and command policies that allowed
it to be unleashed with impunity.

This was the real war, the one that barely appears at all in the tens
of thousands of volumes written about Vietnam. This was the war
that Ron Ridenhour spoke about— the one in which My Lai was an
operation, not an aberration. This was the war in which the American
military and successive administrations in Washington produced
not a few random massacres or even discrete strings of atrocities, but
something on the order of thousands of days of relentless misery— a
veritable system of suffering. That system, that machinery of suffering
and what it meant for the Vietnamese people, is what this book
is meant to explain.

Published by Metropolitan Books, an imprint of Henry Holt and Company. Copyright © 2013 by Nick Turse.  Not to be reprinted without permission.

This piece was reprinted by Truthout with permission or license. It may not be reproduced in any form without permission or license from the source.

Nick Turse

Nick Turse is a historian, essayist, investigative journalist, the associate editor of TomDispatch.com, and currently a fellow at Harvard University’s Radcliffe Institute. His latest book is The Case for Withdrawal from Afghanistan.

Related Stories

Drone Memo Puts Brennan on Spot
By Ray McGovern, Consortium News | News Analysis

Hide Comments

blog comments powered by Disqus