From 1968 until 1973, the US military spent about $US1 billion a year on a new computer-powered initiative intended to end the war in Vietnam. It went by many names over the years — including Practice Nine, Muscle Shoals, Illinois City and Dye Marker. But today it’s most commonly known as Operation Igloo White.
Despite being a high-priced technological failure for the US military, Igloo White was the first real-time, computer-driven surveillance operation program, set up during the Vietnam War.
The US military sought to build a virtual fence dividing North and South Vietnam. And in the process they helped to invent the modern electronic battlefield, whose technologies came back to the US in the early 1970s, where they were quickly deployed against drug cartels, smugglers, and anyone else trying to cross the border from Mexico. Igloo White also formed the bedrock of a border surveillance revolution that’s ongoing today. At the US-Mexico border, drones stalk the skies and electronic sensors alert Border Patrol agents to anyone trying to cross into the United States.
What happened to Igloo White is not unlike the tech transfer from battlefield to border that we see today, as the machines used by American military forces in Afghanistan and Iraq find their way to American streets. And as the refugee crisis continues across Europe, the Middle East, and the rest of the world, it’d be a safe bet to say that we can expect more high-tech virtual fences in the future.
To make sense of what’s happening on today’s borders, though, we have to understand their history.
Cleaner War Through Technology
In the twenty-first century, Americans often recall the Vietnam War through the lens of the gritty movies that the conflict left in its wake. Films like Apocalypse Now and Full Metal Jacket paint a picture of a dirty, soggy war, with troops crawling through mud and wading through rice paddies. But there was another side to the war — one that we don’t see in movies about the period. It’s the high-tech side, with pristine white walls, computer terminals, and air conditioning. It was the place where the messy wars of yesteryear collided with the “clean,” remotely-fought wars of the future.
The US military’s Infiltration Surveillance Center in Thailand with an IBM 360/65 computer circa 1970
That remotely-fought war in Vietnam had its base in Thailand, where state-of-the-art computers would interpret data sent from thousands of sensors dropped across Vietnam and Laos. When the sensors picked up rumblings in the battlefield, computer operators would have to decide whether they were hearing were North Vietnamese supply trucks or simply a random Vietnamese civilian going about their day. If it was the former, air strikes would be called in. If the latter, the military held off.
American military planners promised that the new techniques being used with Igloo White meant we weren’t going to get our hands dirty in war anymore. “Boots on the ground” were no longer necessary to win. Except that they were. Because no amount of computer technology can make a war clean. In fact, as Igloo White shows us, tech has the potential to make war messier than ever.
Screenshot from the US military film Bugging the Battlefield
“On the battlefield of the future, enemy forces will be located, tracked, and targeted almost instantaneously through the use of data links, computer assisted intelligence evaluation, and automated fire control,” General William Westmoreland said in a 1969 speech.
Battlefield of the Future
General Westmoreland was the head of all US military operations in Vietnam from 1964 until 1968, when he became Army Chief of Staff. And his predictions may seem obvious with the benefit of hindsight, but they really were visionary when you consider how primitive battlefield surveillance technology was during the late 1960s.
“I see battlefields under 24-hour real or near-real time surveillance of all types,” he continued. “I see battlefields on which we can destroy anything we can locate through instant communications and almost instantaneous application of highly lethal firepower.”
General Westmoreland was certainly prescient, but to paraphrase a quote from the inventor of the hologram, Dennis Gabor, it’s also easier to predict the future when you’re helping to create it. And that’s precisely what the US military establishment was doing in Vietnam — with a little help from a secret group of scientists and academics.
Igloo White was the brainchild of a secretive group called the Jasons, and they were formed as a quasi-independent think tank in 1960 funded by ARPA (now DARPA), the Defence Department’s bleeding edge technology arm.
The Jason Division (named after Jason and the Argonauts by the wife of one member) was comprised of about 45 scientists from America’s top universities. Mostly physicists in the early days, they would convene for 6 weeks every summer in La Jolla, California, just north of San Diego. The Jasons were all certified geniuses. But as a group they were also incredibly arrogant. And that arrogance would have an indelible impact on the future of not only the war in Vietnam, but nearly every American military action that was to come.
Just as many average Americans had grown tired of the escalating war in Vietnam, the majority of Jasons didn’t like the bombings in Southeast Asia. And they were not shy about saying as much to the top military brass.
“The Jasons were, and I don’t mean to be insulting to them, but let’s just say peaceniks,” Steve Lukasik tells me over the phone. Lukasik was the deputy director and later director of DARPA from 1967 until 1974. DARPA funded the Jasons, but for Igloo White the group had a direct line to the Department of Defence.
“You have to understand that the Jasons had a lot of prestige, a lot of self-esteem, a lot of ego, and they truly believed that they had some good ideas,” Lukasik says. One idea in particular was very interesting indeed.
That idea was to interrupt Viet Cong supply routes running from north to south in Vietnam. Trucks brought those supplies along the Ho Chi Minh Trail, the system of roads used by the North Vietnamese. If the US could just stop enough trucks, the Jasons thought, then the Viet Cong wouldn’t be able to fight. The continued bombings of the Johnson administration had accomplished virtually nothing. So why not simply disrupt supply routes, and force Ho Chi Minh to the bargaining table?
Defence Secretary Robert McNamara thought that a physical fence would be the best way to stopping the flow of weapons, people and supplies from north to south. But the Jasons had a much more futuristic idea. They proposed a virtual fence, made out of a distributed sensor network.
And in the summer of 1966, during their annual meeting in La Jolla, the Jasons came up with a plan to build it. By August of that year they produced a report that promised a virtual fence could be the end to the conflict that everybody had dreamed of.
If only it were that simple.
Control board at the Infiltration Surveillance Center in Thailand circa 1970
In his 2005 memoir, Vietnam War pilot John T. Halliday recounted his arrival at the Nakhon Phanom Air Base in Thailand, headquarters for the Jasons’ dream fence. It was a high-tech wonderland that had all the luxuries of home, like air conditioning. And it was a peek into the future.
War Played Like a Pinball Machine
“Step out of the jungle and inside the building, you step back into America — but an America fifteen years from now… maybe 1984,” Halliday’s co-pilot Duke Wiley told him. Wiley continued:
Remember those huge electronic boards from the movie Dr. Strangelovethat showed Russian bombers headed for the U.S.? Well, Task Force Alpha is a lot like that except with real-time displays in full colour, three stories tall… it’s the whole goddamned Ho Chi Minh Trail in full, living colour. [The place is filled with] a whole bunch of civilians who look like IBM guys running around in three-piece suits all wearing glasses… it’s Geek Central.
Nakhon Phanom Air Base in Thailand housed a new state-of-the-art facility called the Infiltration Surveillance Center from which the US military would conduct their electronic war. At about 200,000 square feet, it was said to be the largest building in Southeast Asia at the time. It had to be huge — the Defence Communications Planning Group had packed it with enormous IBM 360 computers and IBM 2260 monitors.
Monitors at the Infiltration Surveillance Center in Thailand circa 1970
The Defence Communications Planning Group is a pretty boring name for a secretive group tasked with building one of the most high-tech war surveillance systems ever devised — and that’s no accident. The name, DCPG for short, was intentionally dull to prevent North Vietnamese forces from getting too suspicious should they hear about it.
Still, the group wasn’t entirely secretive. After the construction of the Nakhon Phanom Air Base ISC command center in 1967, the Air Force was proud to show it off. Senator Barry Goldwater visited in 1970, declaring the high-tech setup to be future of war: “I personally think it has the possibility of being one of the greatest steps forward in warfare since gunpowder.” And Goldwater wasn’t wrong.
The command center in Thailand gave soldiers a kind of distance from the fighting. Today, many pundits say it’s like playing a deadly video game when US forces conduct drone airstrikes from halfway around the world. Back in the early 1970s, the comparison made by airmen was to pinball machines. One Air Force officer explained in a 1971 issue of the Armed Forces Journal: “We wired the Ho Chi Minh Trail like a drugstore pinball machine and we plug it in every night.”
But there were plenty of hiccups, owing to the system’s relatively primitive capabilities. “The only communication link between the sensors on the ground and Thailand was 7 by 24 aircraft,” Lukasik tells me. Which is to say that unless planes were in the sky, the sensors couldn’t communicate information back to the command center.
Sensors now sitting in a museum via the US Air Force
The Jasons had imagined that the sensors could be off-the-shelf, meaning that they wouldn’t have to be produced solely for Igloo White. And indeed the first batch were largely sono-buoys already used by the Navy to detect enemy submarines. The sonar components were simply replaced with microphones and other sensors as needed.
The Sensor Network
The first sensors were incredibly expensive at about $US2,000 a pop (over $US14,000 in inflation-adjusted dollars) and didn’t work very well. The batteries lasted just two weeks. And the engineers quickly learned that off-the-shelf sensors weren’t resilient enough to be thrown out of planes, which is how the majority were delivered to their positions on the ground. An embarrassingly high number break on landing.
Above, a soldier tosses a sensor out of a plane for Igloo White. The pointed end is supposed to bury itself in the ground, leaving the fake foliage on the other end sticking out.
Dozens of different types of sensors were developed and deployed in Vietnam. Most were disguised, as you can see in the photo above, to be camouflaged and look like vegetation on top, while the bottom of the sensor would be buried in the ground.
There were acoustic sensors which would listen in on the battlefield, seismic sensors which were used to detect the rumblings of vehicles, metal detecting sensors for determining whether guns were in the area, and RF sensors for picking up radio signals. There were even chemical sensors, most notably the “people sniffer” which was designed to detect human sweat and urine.
Each of these sensors posed its own problems. When a metal detecting sensor is meant to be collecting data on soldiers marching with guns, who’s to say that it’s not actually reading the metal from a civilian with a shovel? And what of the rumbling of trucks down the Ho Chi Minh Trail? Was every single vehicle deemed an enemy vehicle? How do you tell the difference between friend and foe? That, of course, was a bigger question for the US military in Vietnam that had to be asked again and again of more than just the sensors.
And then there was the intentional “spoofing” of the sensors by the enemy. Most infamously, North Vietnamese fighters learned to divert the American forces by leaving buckets of urine near sensors in non-essential areas. When the sensors detected a large group’s worth of urine, US airstrikes would begin heavily bombing no one in particular.
Illustration of an acoustic sensor circa 1970
Igloo White sensors were developed by some of America’s most prominent electronics companies, including Texas Instruments, Magnavox, General Electric, Western Electric (under their Sandia Corporation research arm) and Hazeltine Corporation of Little Neck, New York. This was cutting-edge electronics work in the 1960s — totally unexplored territory.
All of the companies came up against a major technical hurdle that will sound familiar: battery life. Nickel Cadmium batteries were used at first (the kind you’ll probably find in your wireless mouse or TV remote), but they’d only last for about two weeks. Other sensors were developed that could be turned on and off remotely as needed, hypothetically conserving power. But the real solution came when companies developed lithium batteries, which extended the life of the sensors for up to two months. But still, the sensors were undependable even when they did “work,” often giving Air Force commanders a false impression of what was happening on the ground.
It wasn’t just sensors that were being dropped. To help their acoustic sensors hear movement in the jungle, the military started dropping “acoustic mines.” These made a loud noise when somebody stepped on them, though they were harmless. But as long as the US military was dropping acoustic mines, they may as well drop deadly mines too, right? That’s why another kind of special mine was tossed out of planes along with the sensors and acoustic mines.
“They were devilish little mines,” Lukasik says. “They were mines you stepped on […] and they had this whole range of what we would now call cluster bombs. Except they weren’t cluster bombs, they were cluster mines.”
Those cluster mines “properly get a very bad name,” Lukasik says, and they really weren’t a “little” thing. Their impact is still being felt today in Vietnam, where thousands of unexploded landmines dot the landscape. It’s estimated that over 42,000 Vietnamese have been killed by leftover mines since American forces left in 1975. An additional 62,000 have been seriously maimed or injured.
Did It All “Work”?
If Igloo White demonstrated anything, it was the limits of remote warfare. Despite being “automated,” there was a lot of guesswork involved in the system. As one Air Force report from 1968 explained:
By noting how long each of the sensors was activated, he could estimate the number of vehicles in the group moving through a string. This gave him enough confidence to identify this activation pattern as a potential target, and he estimated that target to be five trucks moving northward at 17 kilometers per hour. He then transmitted this target sequence to the ISC operations room for recording and relay to the ABCCC for possible visual investigation by a FAC and possible strike action, if confirmed.
Nobody in the US military wanted to admit that their $US1 billion per year system wasn’t getting results. The Air Force claimed that 75,000 trucks had been destroyed as a result of their sensor network. The only problem? By the CIA’s estimates, there were only about 6,000 trucks in all of North Vietnam.
The lag time and uncertainty of the Igloo White system in the early 1970s was unsustainable. At ARPA, Lukasik concluded that “you don’t put in a number of cheap sensors [on the ground] and then maintain them.” Instead, you develop a more compact platform that doesn’t rely on so many relay points.
The Lockheed Batcat EC-121R, which would relay the sensor’s data to the command center
Lukasik thought the weakness of the Igloo White system was that the sensors were on the ground, and drones were merely relay devices between them and the computers in Thailand.
“So we said, take a really good sensor platform and fly it around,” Lukasik said. “It’s a hell of a lot easier to buy between 50 and 500 remotely piloted vehicles and that takes out this horrible 24/7 airborne link.”
These new drones had optical sensors, radar sensors, infrared sensors, among plenty of others. Suddenly these drones didn’t have to receive a message from an unreliable sensor on the ground and relay it to the IRC anymore.
“We called them in those days RPVs,” recalled Lukasik. “They’re now called autonomous unmanned vehicles. But they’re not unmanned. Every one of them, there’s a guy sitting in Nevada or some other place, flying the thing.”
Even here in the 21st century we haven’t gone full Skynet quite yet. Humans are still in the loop. There’s no way of knowing whether their drone strikes on military targets were any more “effective” since the measure of success in Vietnam is a muddy and frightful thing to calculate. But some people find comfort in knowing, especially here in the year 2015, that at least there were humans behind those triggers.
Professor Paul Edwards of the University of Michigan has written extensively about Operation Igloo White, most famously in his classic 1997 history The Closed World: Computers and the Politics of Discourse in Cold War America. I called him at his office in Michigan to talk about Igloo White. And unsurprisingly, he put Lukasik’s plan for a mobile platform into a similar historical context.
What they were trying to do then is what we do now, except a lot of those sensors are mobile because they were on drones. But the basic principle of remote control war where you get the soldiers as far away as battle as possible and do it with planes if you can or missiles, that’s been a steady theme in US military history from World War II right on up.
And if your goal is to protect your soldiers, it’s a good idea. The problem, then as now, is that it’s hard to discriminate civilians from military targets when you’re very far away from them. And that continues to be an issue. We keep on killing civilians with our drones in Pakistan and places like that because we’re not there on the ground to really identify the people we’re going after.
IBM 2250 display console used to track sensors in the late 1960s
Bringing the War Home
Some people were frankly excited by the prospect of these futuristic battlefield conditions, as long as they didn’t include land mines on US soil. The February 14, 1971 edition of the Sunday newspaper insert The Family Weekly described the tactics to come. As long as our boys in Vietnam were getting high-tech tools to kill people half-way around the world, that was good enough for them. But there were others who noticed that what goes around comes around.
One of the few voices of the 1970s to notice that all this technology was being brought back to the United States was Robert Barkan.
In the June 15, 1972 issue of New Scientist, Barkan wrote:
Smugglers on the US/Mexican border are treading softly these days, now that the US Border Patrol (an arm of the Justice Department) has adopted the same anti-infiltration barrier used by the military to detect troop and truck movements of the Ho Chi Minh Trail.
Remote stretches of the border have been seeded with sensors similar to the Acousids and Minisids that detect sounds and vibrations from footsteps and vehicles in Vietnam.
Barkan explained that it was all coming back to the North America — the drones, the sensors, the computers. The only thing that wasn’t coming were the bombs. Or at least we hoped.
Beech QU-22B, National Museum of the United States Air Force
In the same New Scientist article, Barkan described the Mexican border of 1972:
The US Air Force’s QU-22B remote controlled pilotless aircraft — made surplus in Vietnam by the introduction of more sophisticated drones — have been returned to the US where they fly over the border to monitor the sensors and relay data to central control points.
Sylvania Electronics, one of the companies making the sensors for use in Vietnam, assured Congress in 1970 that a virtual fence would work well along the US-Mexico border because by its very nature nobody would notice it: “The political implications of using surveillance equipment along a friendly foreign border have been considered by selecting equipment that can be deployed without attracting attention and easily concealed.”
It wasn’t just the US-Mexico border, either. President Nixon’s administration experimented with sensors at the White House and at his vacation homes.
As Barkan explained in New Scientist:
War-tested sensors have been placed under the White House lawn and in the yards of President Nixon’s other homes in San Clemente, California, and Key Biscayne, Florida. This caused a problem in Florida, where salt spray from the nearby ocean activated the sensors (the community reacted against a proposal to alter the shoreline so the spray wouldn’t hit the lawn!).
But stepping back, we have to understand Igloo White as a failed experiment. Most firsts are.
“It didn’t work,” Lukasik says of the virtual wall in Vietnam. “Or I should say, it didn’t work as advertised. How do I know that? Because we loaded everybody aboard helicopters in ’75 and got the hell out.”
It may not have worked in Vietnam. But it would radically shift the tenor on the US-Mexico border for years to come.
We’re Building a Wall, It’s Just Virtual
The existence of the Jasons, the secretive group of scientists who conceived Igloo White, would finally become known to Americans through the Pentagon Papers. And there was subsequently a debate among academics about what kind of contributions scientists should be making to the military establishment in a war like Vietnam.
Suddenly the physicists who had helped win World War II were no longer seen by their peers as heroes. In a conflict like the Vietnam War, technologists solving problems for the military were regarded with suspicion. Especially as those technologies were brought back from Vietnam to the US-Mexico border.
2005 photo showing a Southern California border monitoring station via AP Photo/Lenny Ignelzi
Whether they understood it in the 1960s or not, the Jasons would come to learn by the 1980s that the technologies they had held to devise for Vietnam were coming back to the US border.
“We were working for Customs and Immigration, I guess,” legendary physicist and Jason, Freeman Dyson told Ann Finkbeiner for the 2006 book The Jasons: The Secret History of Science’s Postwar Elite. Dyson had been invited to a trip to the Mexico border in the 1980s to see how the virtual fence idea might work there at the behest of the military. “We were trying to stop the drug traffic. Simply, we wanted to educate ourselves about what was going on.”
As Todd Miller explains in his book Border Patrol Nation, in 2012 the US-Mexico border contained, “377 remote video surveillance systems, 195 local video surveillance systems, 305 large-scale nonintrusive inspections systems, 75 Z Backscatter vans, 261 Recon FLIRs, more than 12,000 sensors, and 41 mobile surveillance system trucks.”
And the tech has only expanded and gotten more sophisticated. Shit, they’re even hiding cameras in cacti now.
The technologies developed in wartime inevitably have an indelible impact on the countries that manufacture them. World War II gave us computers and even drones. And the Vietnam War expanded the use of those technologies both in Southeast Asia and North America.
Slowly but surely the field of battle would become precisely what General Westmoreland had predicted: 24-hour monitored battlefields dominated by “instant communications and almost instantaneous application of highly lethal firepower.”
What General Westmoreland and others didn’t tell us was that the battlefield always comes home.
Top illustration by Tara Jacoby
Sources: Defence Today