CDI Headlines Hot Spots Research Topics CDI Publications Television Search
CDI Mission CDI Staff CDI Expertise Paid CDI Internships Support CDI
CDI Home
CDI Terrorism Project Home    

Terrorism Project Home
Eye on Iraq
Operation Enduring Freedom
The Terrorist Network
Homeland Security
U.S. Weapon Systems
Foreign Policy/Security Strategy
News and Opinion
Arms Trade
Killing your own:  the Problem of friendly fire during the Afghan campaign
June 12, 2002 Printer-Friendly Version

“Friendly fire” must surely rank as the most unfortunate contradiction in terms in the military lexicon.  The U.S. Department of Defense defines friendly fire as ‘a casualty circumstance applicable to persons killed in action or wounded in action mistakenly or accidentally by friendly forces actively engaged with the enemy, who are directing fire at a hostile force or what is thought to be a hostile force.’1  Soldiers from Princess Patricia’s Canadian Light Infantry (PPCLI) are the most recent coalition troops to die from friendly fire — a phenomenon also known, with equal irony, as fratricide — during the ongoing U.S.-led campaign in Afghanistan.2  The incident involving the PPCLI occurred on April 17 near the Afghan city of Kandahar when a U.S. Air National Guard F-16 dropped a 500-pound bomb on the Canadians who were conducting a night live-fire exercise at the time.  Four soldiers were killed and eight others injured in the incident, which was the subject of a joint investigation by Canadian and American authorities.  It was not the only such occurrence to warrant an investigation by the American military, with U.S. Central Command (Centcom) releasing a list on March 29 of ten incidents that “warranted review.”3 The list included the following eight instances of potential friendly fire or fratricide:

Oct 16 & 26, 2001: Coalition air strikes were launched against International Committee of the Red Cross (ICRC) facilities in Kabul.  Despite the ICRC’s being asked to provide coordinates for all its facilities in Afghanistan prior to American military action in the country, details of the Kabul compound were not passed on to the U.S. military.  However, in the wake of the Oct. 16 raids, CENTCOM was advised of the location of this facility.  Despite this, a subsequent raid on the ICRC compound took place on Oct. 26.  This second raid saw two B-52s each drop three 2,000 JDAMs on the warehouses, while an F-18 inadvertently dropped a 500-pound GBU-12 bomb on a residential area some 700 feet south of the warehouses.  Initial indications were that the GBU-12’s guidance system malfunctioned, but the raid itself was attributed to a human error in the targeting process.
Nov. 11, 2001: A UN Convoy traveling to Bamian was struck by debris when the cliffs alongside a road were bombed in an effort to deny the route to enemy forces.  There were no casualties in the convoy.  A CENTCOM report released on April 10 claimed that the convoy had not coordinated its movement with coalition forces and was not displaying UN identification markings.
Nov. 26, 2001: Five U.S. troops were injured near Mazar-e-Sharif when an F-18, working close air support (CAS) in conjunction with a ground controller, injured five U.S. Special Forces personnel with a Joint Direct Attack Munition (JDAM).  The raid also killed at least six allied Afghan fighters, who were working with the American troops to quell a revolt by al Qaeda/ Taliban prisoners at the Qala Janghi fortress.  The initial findings of a CENTCOM investigation suggested that the accident was the result of ‘procedural errors in the transmission and application of friendly and enemy coordinates.’4  The U.S. Special Forces team in question was being overseen by a recently –arrived headquarters detachment when the raid happened.  Indeed it was a headquarters’ member who was in charge of calling in the air strikes.  This raises the question of whether one of the more experienced Special Forces soldiers could have avoided the friendly fire incident.  At least two of the Special Forces personnel were reportedly upset at not being tasked with calling in the air strike, and a third alleged that some in the headquarters unit were ‘trigger happy,’ adding that they seemed anxious to do some bombing.’5
Nov. 28, 2001: U.S. helicopters fired on a Navy Special Warfare Unit patrol near Forward Operating Base Rhino.  There were no casualties or damage to equipment incurred in the incident.  A subsequent U.S. Naval Forces Central Command (NAVCENT) investigation was conducted, and a communications and identifications procedure amended to avoid any such incidents in the future.
Dec. 5, 2001: A B-52 providing CAS under the guidance of a ground controller dropped a JDAM near Sayd Alim Kalay, killing three U.S. troops and five allied Afghans and injuring over forty.  It was later ascertained that the changing of a Global Positioning System (GPS) receiver’s battery in the middle of the American air raid caused the incident, which almost killed the just-appointed Afghan Interim Authority leader, Hamid Karzai.  The receiver defaults to display its own coordinates after the battery is replaced, something the operator either did not know or overlooked in the heat of battle.  While the latter is understandable if tragic, the former should raise serious questions about the training U.S. forces receive on such equipment.  A Special Forces’ source familiar with the incident disclosed that a newly-arrived Air Force Tactical Air Control Party (TAC-P) called in the air strike in conjunction with a headquarters officer.6  On previous missions the operational detachment – rather than a headquarters element – tended to call in air strikes, usually via a more experienced and better trained controller.
Jan 23, 2002: Coalition forces raided two suspected al Qaeda/Taliban compounds near Hazar Qadam.  Two locals were killed at one compound, 14 at the other. One American was injured.  It was later ascertained that no al Qaeda/ Taliban personnel were at either target, and the 27 people detained during the raids were subsequently released.  According to CENTCOM, the deaths occurred as U.S. Forces were fired upon first.  The intelligence information that led to the raid came from American sources, causing U.S. Defense Secretary Donald Rumsfeld to dismiss suggestions that American forces may have been intentionally mislead by Afghan rivals of those targeted.  In an attempt to calm Afghan tempers, the CIA later distributed at least $1000 to each of the families who suffered a loss during the raids.  The Agency was not involved in the raid itself and was not made aware that the operations were to be carried out.7
March 2, 2002: As Operation Anaconda began, an allied-Afghan/U.S. convoy came under what was originally thought to be an enemy mortar attack near Terghul.  Several allied-Afghans were killed, as was a U.S. Special Forces soldier.  However, it later emerged that an American AC-130 Spectre Gunship was engaging what it believed to be an enemy convoy at the same time the allied-Afghan/U.S. convoy was coming under attack.  CENTCOM commander, Gen. Tommy R. Franks, directed his Special Operations component to investigate any relationship between the two incidents.
March 6, 2002: Coalition forces attacked a suspected al Qaeda leadership target traveling in a vehicle near Shikin.  The attack left 14 dead and one wounded. The dead included eight adult males, three adult females and three children.  The Pentagon later defended the targeting of the vehicle, claiming that it was in a known enemy area and that there was reason to believe that it contained al Qaeda personnel.

These episodes do not constitute the only instances of ‘friendly fire’ during the Afghan war.  Many civilians have died in the campaign at American and coalition hands.  According to a study by Professor Marc Herold of the University of New Hampshire, 3,767 Afghan civilians were killed by U.S. air strikes between October 7 and December 7 last year.  While these figures have been disputed in some circles, Herold claims they are conservative and that a much more realistic figure for civilian Afghan casualties would be around 5,000, as he omitted to count those who died as an indirect result of such raids or who died later of their injuries.8  The exact number of civilian casualties in Afghanistan will probably never be known, however it can be assumed to be high.

In addition, Afghan civilians have often found themselves targeted in ground raids intended to detain al Qaeda/ Taliban fighters, as have allied-Afghan troops.  A recent instance of the latter occurred on May 31 near Gardez, when U.S. troops engaged a group of friendly Afghan fighters, killing three and wounding two others.  This incident occurred in the village of Khomar Kalay, when a team of U.S. Special Forces surrounded a compound reported to contain al Qaeda members.  The Americans were unaware that approximately 22 allied-Afghans from the neighboring Logar province had already began a search of the compound.  The incident occurred when one of the Afghans appeared to aim a grenade launcher at the Americans.

Such instances should not be taken as an indication that friendly fire is in any way peculiar to the ongoing operations in Afghanistan.  Such occurrences have been depressingly familiar throughout the history of warfare, and even senior commanders have not been immune to the firepower of their own troops.  Among the most famous incidents of friendly fire in America’s military experience was the fatal shooting of Lt. Gen. Thomas J. “Stonewall” Jackson by his own men on May 2, 1863.  Nor is fratricide something peculiar to wartime deployments: one of the worst recent examples involving Americans took place over Iraq in 1994 when two U.S. Air Force fighter pilots unwittingly shot down two U.S. Army Black Hawk helicopters, killing 26 personnel.  However, it was the Persian Gulf War that brought home the problem of friendly fire most graphically, with friendly fire accounting for 24 percent of those Americans killed in action and 15 percent of those wounded.9  By comparison, 7 (possibly 8) out of 23 coalition hostile fire casualties in the Afghan campaign (a figure approaching 35 percent) have been caused by fratricide.  The relatively low numbers of coalition casualties in Afghanistan to date (22 out of 32 Purple hearts awarded by March 2 were for wounds inflicted by friendly fire) correspondingly affect the friendly fire/ hostile fire ratio.  Nonetheless, such figures indicate that friendly fire continues to be a disturbing aspect of warfare.

This is perhaps more true today than in the past. Amplified media coverage and a more sophisticated public opinion increase pressures on political and military leaders to ensure that killing friendly troops or civilians is as limited as possible.  Ongoing initiatives to do just that include the work undertaken by the U.S. Joint Combat Identification Evaluation Team (JCIET), a subordinate unified command of the U.S. Joint Forces Command, headquartered at Elgin AFB, Florida.  JCIET completed a series of field evaluations at the end of April in which over 4,000 U.S. and allied military personnel participated.  The aim of the exercise, the sixth of an annual series, was to test all aspects of combat identification, such as systems, doctrine and training, providing feedback that will assist participating militaries’ in lessening the incidence of fratricide.  Much of the attempt to cut down on friendly fire revolves around technological solutions such as Identification Friend or Foe (IFF) systems which will enable troops to differentiate between allied and enemy troops.  However, technology does not offer a panacea for fratricide, with some commentators estimating that IFF could, at best, lessen friendly fire incidents by no more than fifty percent.10  Moreover, while friendly troops may conceivably be equipped with IFF systems, this is unlikely to be true of friendly civilians.  More importantly, as JCIET acknowledge, fratricide arises from many factors which are not easily ameliorated by technology.  This is especially true in the case of ground fighting, where the fog of war tends to be even more amplified than in the more ‘sterile’ environments of aerial or naval combat.

Causative factors in fratricide include difficulties imposed by terrain; visibility; type, size, and scale of operation; high technology; carelessness; stress of combat; and lack of training, discipline, fire control and coordination.  Such instances of friendly fire that have occurred in Afghanistan to date would appear to confirm this diagnosis.  Unfortunately, such factors are perennial and, while it is to be hoped otherwise, the casualties suffered by the PPCLI are unlikely to be the last such incurred in the war against terrorism.  While a point of diminishing return may someday be reached with the research into reducing friendly fire losses, this is still some way off.  JCIET and similar initiatives will not banish the scourge of fratricide; they will help limit it.  As such they should be encouraged and widened.


1 “Department of Defence Dictionary of Military and Associated Terms,” Joint Publication 1-02, 12 April 2001 (As amended through 9 April, 2002), p. 178.

2 Also see, Lieutenant Colonel Charles R.Shrader, “Amicide: The Problem of Friendly Fire in Modern War,” Combat Studies Institute, December 1982.  This suggests the word Amicide for describing friendly fire.  While graphically applicable, Shrader’s term could possibly be more usefully employed for cases of friendly civilian casualties, to differentiate these from casualties inflicted upon friendly troops.

3 United States Central Command, “Status of Investigations During Operation Enduring Freedom,” 29 March, 2002.  Online at

4 “Status of Investigations During Operation Enduring Freedom.”

5 Quoted in, John Donnelly, “A Special Few Led to Afghan Success: Elite U.S. Forces Overcame Glitches,” Boston Globe, 31 March, 2002.

6 See, Jules Crittenden, “Report: Air Controller Called in Friendly Fire,” Boston Herald, 27 March, 2002.

7 “U.S. Copes With Treachery in Picking Afghan Targets,” Reuters, 8 February, 2002.

8 “Afghanistan’s Civilian Deaths Mount,” BBC Online, 3 January, 2002.

9 Figures quoted in, Rick Atkinson, “Crusade: The Untold Story of the Persian Gulf War,” (New York: Houghton Mifflin, 1993), p. 315.

10 Major Charles F. Hawkins, “Friendly Fire: Myths and Misperceptions,” Proceedings, June 1994.


By Mark Burgess
CDI Research Analyst

Printer-Friendly Version



1779 Massachusetts Ave, NW, Washington, DC 20036-2109
Ph: (202) 332-0600 · Fax: (202) 462-4559