War Stories

Generals’ Apathy

The Pentagon’s appalling record on “friendly fire.’

Soldier killed by friendly fire

The Pentagon revealed earlier this week that it is investigating whether the first U.S. soldier killed in Operation Anaconda might have been the victim not of an Al-Qaida mortar, as was earlier reported, but of a U.S. AC-130 gunship. This is a reminder that the U.S. military has made appallingly little progress against one of its most persistent problems: killing its own. The Anaconda death is one of four Afghanistan “friendly fire” cases currently under investigation. And it’s equally appalling that neither the press nor Congress has over the years found this a particularly interesting topic. All this inaction says a lot about what, in between flag waves, the powers that be really think about the value of our soldiers’ lives.

The dirty little secret of the Gulf War is that “friendly fire” accounted for 24 percent of the U.S. dead. A rate this high is not an inevitable consequence of the “fog of war.” According to a study done at the Army’s Command and General Staff College, “friendly fire” casualties accounted for less than 2 percent of all those occurring in battle during World War II, Korea, and Vietnam. Nor was the Gulf War statistic skewed by one or two anomalous incidents—the Pentagon found that during the 100-hour Desert Storm ground war, there were 27 separate cases of fratricide. No, what has happened is not inevitable—it’s the result of expanding our offensive capability in such areas as night-fighting and beyond-visual-range targeting without commensurately expanding our capability to identify targets. This is like a car company building bigger engines without building better brakes.

You’d think that the U.S. military would be on the case, given that it has the most to lose from failure here. You’d think. But the Gulf War fiascos didn’t prevent U.S. fighter planes from shooting down two U.S. helicopters over Iraq a few years later, killing 21 military and civilian personnel, including 15 Americans.

What happens instead is that the military-industrial complex talks about solving the problem and is skilled at making optimism seem warranted. So, for instance, a 1992 Air Force magazine article attributes to Gen. Wesley Clark (former NATO commander and now a studio general for CNN) the goal of having target identification systems in every Army vehicle. A 1995 Journal of Electronic Defense article states that “the Army has passed through the engineering and manufacturing development … stage of a near-term, ground-to-ground combat ID solution to provide armored units with a much more robust capability” and concludes that “the friendly fire statistics should show considerable improvement as the military heads deeper into the post-Cold War era.” A 1997 Defense Daily story emphasizes one officer’s enthusiasm for an identification system being tested and says “the technology permits his tank crews to locate the movements of friendly forces on the battlefield by merely checking the icons on their laptops.” Only near the end is a senior Army analyst quoted saying there is only “a very small sample” of the system preventing fratricide incidents in exercises.

Similarly, although Philip Coyle, the head of Pentagon testing during the Clinton administration, tells me that the Army has a couple of  $1 billion-plus target identification programs, and that its top general says the “friendly fire” problem is the Army’s “No. 1 priority,” Coyle admits that he doesn’t know how much of this technology has gotten into the actual hands of actual soldiers. When he pulls out a document from his former office to check out the career of one such Army program (involving laptop computers that would display all friendly and enemy forces), he notes that it started in 1994, underwent its first limited testing in 1998—when problems became evident—and then experienced further difficulties during each of its subsequent field tests. Its December 2001 test was postponed. As of now, says Coyle, the gear still hasn’t undergone realistic operational testing. At this point, I don’t want to hear how the Army is doing on its No. 2 priority.

A similar lack of urgency lurks beneath one of the most notorious of the “friendly fire” episodes on the Pentagon’s investigations list: the one last December in which a B-52 airstrike killed three Green Berets and numerous anti-Taliban Afghans—and nearly killed Hamid Karzai, the man who subsequently became Afghanistan’s prime minister. It has been widely reported that the disaster occurred because a ground-based U.S. targeter mistakenly radioed his own coordinates to the bomber. The Los Angeles Times reported that this happened because the ground controller failed to re-enter the proper target coordinates after changing the batteries on his global positioning system device, which was necessary because “after a battery change, the unit automatically displays its own location.” Coverage of the incident has tended to view this episode as “fog of war” stuff, but nobody in the media (or in the Pentagon or the Congress) has asked the key question: How could this obviously dangerous battery-changing feature of the Army’s GPS have possibly survived the military’s development, testing, and procurement process? Surely the manufacturer of the device intentionally built in this feature—but did it tell the Army? And if it did, why didn’t the Army have it removed?

And if computer-based approaches to the “friendly fire” problem are proving too recondite, where are the Pentagon’s quick fixes? Here’s an idea that seems pretty good to me: How about putting receivers on each Army vehicle that can detect when it’s being painted by an airborne laser? Once the United States establishes air superiority, as it has in Afghanistan, being lased from above can only mean one thing—and the targeted vehicle could then immediately get on the horn with a “Don’t shoot me, I’m here” message (which needn’t tell the enemy anything because that broadcast could go out over a scrambled friendly radio frequency and/or be in code). Laser detection technology already exists and is already in use on some military aircraft and vehicles—it’s even marketed to those wishing to detect police speed traps. Why shouldn’t GI’s have as much protection as speeders?

Given all this, here are a couple of questions the defense press might want to pose the next time Donald Rumsfeld gives a briefing: How much has the Pentagon spent on the “friendly-fire” problem since the Gulf War, compared to, say, on the B-2 stealth bomber, the Joint Strike Fighter, the F-22 fighter, or nuclear submarines? And what does it have to show for it? If a U.S. aircraft is about to attack a U.S. infantry unit in Afghanistan right this minute, what can our soldiers there really do about it?