Chicago Tribune July 21, 2002
U.S. forces have yet to field systems to prevent 'friendly fire'
By STEPHEN J. HEDGES
WASHINGTON - In the night sky over Afghanistan several months ago, an F-16 fighter pilot from the Illinois Air National Guard reported weapons fire below and moments later radioed that he was attacking in self-defense. The 500-pound, laser-guided bomb was deadly accurate, but it didn't hit the enemy. Instead, it landed on the Princess Patricia's Canadian Light Infantry, killing four soldiers.
The accident in April near Kandahar may be the most egregious episode of "friendly fire" in a conflict that is experiencing a growing number of such incidents. So far, 10 incidents are under review. Analysts say the accidents are striking not just for their number, but because they resemble the 35 U.S. friendly fire deaths in the Persian Gulf war. Back then, the military vowed to find ways to prevent the fatal accidents.
But 11 years and more than 100 million dollars later, U.S. forces have yet to field systems that will adequately allow ground and air forces to identify each other. In fact, the troops in Afghanistan are equipped with only modest improvements of the systems deployed during the gulf war.
"What they came up with is a series of different things that can be used," said John Pike, a defense analyst who runs GlobalSecurity.org. "But they never came up with a universal system that would always be used, because under most circumstances, you don't need it."
The Afghan accidents included errant bombs and assaults that have killed Americans and allied Afghan fighters and civilians as well.
Just last week a team of U.S. military specialists arrived in Afghanistan to investigate an American aerial assault at the village of Kakarak.
The Pentagon says U.S. forces were tracking legitimate al-Qaida terrorist targets on July 1 when an Air Force AC-130 gunship opened fire. The Afghans say a group of people was celebrating a wedding, and the American assault killed 48 of them.
Human-rights organizations estimate that more than 1,000 Afghan civilians have died from off-course bombs, or because they were mistaken for Taliban and al-Qaida fighters.
Service by service, officers and defense contractors have spent millions of dollars during the past decade looking for ways to diminish incidents of firing on the wrong targets. But workable solutions are still not in sight.
The Army recently canceled an ambitious "Identify Friend or Foe" program after spending $100 million on its development. The Battlefield Combat Identification System was ended because it would have cost another $920 million to place it in Army vehicles.
The program was the subject of a March 2001 Pentagon Inspector General's audit, which found that, among other things, the Army lacked, "an up-to-date and comprehensive test and evaluation master plan" for the system, and that there were no plans to "test a production prototype of the system in cold, fog, snow or rain."
The Air Force has long had a "friend or foe" transponders for its aircraft, and its next generation system, Link 16, is designed to augment that capability. But the Air National Guard, which flies alongside Air Force planes in Afghanistan, has its own program, the Situational Awareness Data Link, which relies on frequency space borrowed from an already clogged Army communications system.
The Navy, like the Air Force, has a reliable air-to-air system, but it is not designed to coordinate with ground force systems.
And though the Department of Defense in recent years has led a number of multi-service operations in the Persian Gulf, Bosnia, Kosovo and now Afghanistan, none of its services' combat identification equipment is designed to work as an integrated system, according to a June 2001 General Accounting Office Audit.
Nor are the U.S. systems capable of working with those of America's allies, the audit found. In fact, only recently has the North Atlantic Treaty Organization, or NATO, developed a standard for "combat identification" equipment that it wants member nations to adopt.
"We tend to approach the problem from a very limited scope," said Christian Carrier, director of Science and Technology/Air, for Defense Research and Development Canada, a government agency. "For example, if there is an incident involving tanks, well, somebody will develop the solution for the problems of friendly fire involving tanks. Then there will be another incident involving infantry soldiers, and then someone will develop a system for infantry soldiers."
Friendly fire has always been an element of war. Those who study it invariably note that Confederate Gen. Stonewall Jackson was mortally wounded by his own men at Chancellorsville, Va. For years, officers have accepted the belief that friendly fire would cause 2 percent of casualties in any war. Recent reviews of World War II and Korean War casualty records suggest the correct figure might be as high as 12 percent to 15 percent.
Attitudes about friendly fire changed, however, with the high number of incidents during the gulf war.
When the U.S.-led forces launched their ground assault against dug-in Iraqi forces, their sweeping phalanx of armored vehicles moved faster than even the commanders had planned. Bradley fighting vehicles carrying U.S. soldiers sped into sectors where American tank crews expected to find Iraqi tanks, and became unwitting targets.
The relatively low numbers of Americans killed by hostile fire in both the gulf war and the current conflict--18 have fallen to enemy fire in Afghanistan_may serve to highlight the friendly fire deaths.
But military analysts say the numbers are also climbing because American weapons have become more complex and more accurate, inviting human mistakes and eliminating the margin for error.
"The problem is the weapons go exactly now where they're told to go, and we're seeing an increase in human error," said retired Col. Dan Smith of the Center for Defense Information in Washington. "A munition may be smart, but it's not intelligent. There's no way to correct for faulty input."
A tragic example of just such a mistake occurred in December, when three U.S. soldiers and an estimated 25 Afghans died after their position was struck by a satellite-guided bomb.
The bomb, dropped from a B-52, was supposed to hit target coordinates sent to it by a forward ground controller. But after sending the coordinates, the controller reportedly stopped to change the batteries in his handheld satellite navigation unit. When he restarted it, the unit defaulted to the coordinates of the controller's own position, inadvertently redirecting the bomb back to him.
In recent years, Pentagon efforts to find friendly fire solutions have followed two paths.
The first intended to give forces on the field their own combat identification capability. The Army, stung by the worst of the gulf war's friendly fire, was banking on its recently canceled Battlefield Combat Identification System. By pointing at a target with a special gunsight, a soldier or tank driver could send a wave beam signal and wait for a reply. If it one came back, the target was friendly. If it didn't, the sight was pointed at the enemy.
Though it did its job, the system, developed by TRW Inc., would have cost up to $50,000 a unit.
The second path for friendly fire solutions centers on what the military calls "situational awareness." With the advance in information technology, the military is hoping to use global positioning system satellite receivers, digital radios, laptops, coded signals to provide a near real-time picture of unfolding battles or operations.
"A lot of work is being done, and money's being spent, to develop this common operating picture," said Carrier, "and it's mainly being developed from the point of view of mission effectiveness. A secondary benefit of that common operating picture is also it will reduce friendly fire incidents."
And over the past several years, the Joint Combat Identification Evaluation Team, based at Eglin Air Force base in Florida, has held series of promising combat exercises using new battlefield technologies. They include everything from simple signal identifiers to the thermal imaging of specific targets, such as Russian-made T-72 tanks.
"There's no one golden BB or bullet out there to solve the issue," said the Lt. Col. Mark Jenner, program director of the evaluation effort. "Combat situations are very, very dynamic."
The evaluation team held its latest exercise last April in the Gulf of Mexico. It was no small task, employing 4,700 personnel, a British frigate and Navy cruiser, Navy, Air Force and British planes, Army and Marine troops and vehicles, as well as a number of former Eastern bloc weapons. Every element of the simulated battle was wired so the evaluation staff could collect data.
But even in that closely controlled environment, friendly fire_the very problem the office is trying to solve_became a factor.
"We did have some instances," said Joe Gordon, a evaluation team technical adviser. "I'm not familiar with the details. Through a number of years (of exercises), we've been dealing with instances of friendly fire. Bad things still happen."
Copyright 2002 Chicago Tribune
