Operation Anaconda in Afghanistan A Case Study of Adaptation in Battle

Case Studies in Defense Transformation Number 5 Richard Kugler

i

Sponsored by the Office of the Deputy Assistant Secretary of Defense Forces Transformation and Resources

Prepared by the Center for Technology and National Security Policy

Report Documentation Page Form ApprovedOMB No. 0704-0188 Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

1. REPORT DATE 2007 2. REPORT TYPE

3. DATES COVERED 00-00-2007 to 00-00-2007

4. TITLE AND SUBTITLE Operation Anaconda in Afghanistan. A Case Study of Adaptation in Battle

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) National Defense University,Center for Technology and National Security Policy,Fort Lesley J. McNair BG 20,Washington,DC,20319

8. PERFORMING ORGANIZATION REPORT NUMBER

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES The original document contains color images.

14. ABSTRACT

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT

18. NUMBER OF PAGES

27

19a. NAME OF RESPONSIBLE PERSON

a. REPORT unclassified

b. ABSTRACT unclassified

c. THIS PAGE unclassified

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

The views expressed in this article are those of the authors and do not reflect the official policy or position of the National Defense University, the Department of Defense or the U.S. Government. All information and sources for this paper were drawn from unclassified materials. Dr. Richard L. Kugler is a consultant to the Center for Technology and National Security Policy. His specialty is U.S. defense strategy, global security affairs, and NATO. He advises senior echelons of the Office of Secretary of Defense, the Joint Staff, and the interagency community. An operations research analyst and political scientist, he holds a Ph.D. from the Massachusetts Institute of Technology.

ii

Introduction In his memoirs, American Soldier, former U.S. Central Command (CENTCOM) Commander General Tommy Franks, USA, (Ret.) portrayed Operation Anaconda, as an “absolute and unqualified success,” but one in which the original U.S. military battle plan “didn’t survive first contact with the enemy.”1 General Franks’ apt portrayal provides the framework for this case study of Operation Anaconda, which took place in the Shahikot Valley of eastern Afghanistan during early March 2002. The goal of Operation Anaconda was to root out enemy Taliban and al Qaeda forces that had gathered in this valley following their earlier defeats in the initial three months of the war in Afghanistan. In order to achieve this goal, U.S. commanders crafted a complex and sophisticated battle plan involving a “hammer and anvil” attack by U.S. and friendly Afghan ground forces into the valley. This battle plan unraveled on the first day when enemy resistance proved fiercer than originally anticipated and friendly Afghan forces failed to carry out their march into the valley, thereby leaving deployed U.S. infantry forces to face the enemy alone. Success was achieved when U.S. forces switched tactical gears by calling on air strikes, in larger numbers than originally planned, to work with the ground forces to suppress and destroy the enemy. Originally planned as a three-day battle with light combat, Operation Anaconda turned out to be a seven-day battle with intense combat and was officially terminated only after 17 days. Operation Anaconda, which lasted from March 2–18, was successful because up to several hundred enemy fighters were killed and the rest fled the Shahikot Valley, leaving it in the control of U.S. and allied forces. U.S. casualties totaled eight military personnel killed and over 50 wounded. Success was achieved because the U.S. military showed a capacity to adapt by employing joint operations and modern information networks to surmount a surprising and difficult challenge. As a result, this battle was the last time that year that enemy forces chose to engage U.S. forces in major combat in Afghanistan. In the aftermath, nonetheless, came controversies about several issues, including original intelligence estimates, the U.S. command structure, the initial reliance upon friendly Afghan forces, the armament of U.S. Army light infantry forces, and networked air-ground coordination of air strikes against enemy positions. In the months after Operation Anaconda, many of the problems encountered there were corrected by the U.S. military, and they did not reappear when Operation Iraqi Freedom, the invasion of Iraq, was launched in early 2003. Even so, the events of Operation Anaconda, the biggest pitched battle of the Afghanistan war, deserve to be remembered, as do its positive and negative lessons for modern-era force operations and defense transformation. The following pages endeavor to recount the battle’s key features, its initial frustrations, and its ultimate success.2 1 See General Tommy Franks (Ret. USA), American Soldier (New York: Regan Books, 2004), 379. 2 This case study is based on research conducted by CTNSP/NDU, including interviews with senior U.S. military officers that participated in the battle. It also draws upon other published literature on Operation Anaconda.

1

2

The Military Setting: How Earlier Battles Set the Stage for Anaconda Operation Anaconda was an outgrowth of earlier events during the war in Afghanistan. Major U.S. combat operations in Afghanistan—Operation Enduring Freedom—began October 7, 2001, less than one month after the September 11 terrorist strikes against the United States. In endeavoring to topple the Taliban government and destroy al Qaeda, U.S. forces fought a new type of war in Afghanistan. The distant, landlocked country prohibited an immediate infusion of large ground forces. Instead, CENTCOM relied heavily upon special operation forces (SOF) teams that employed satellite radios, lasers, global positioning system (GPS) and Predator unmanned aerial vehicles (UAVs) to designate targets for air strikes, which were provided by cruise missiles and a combination of strike aircraft, including AC-130 gunships as well as strategic bombers and tactical fighters that delivered precision munitions, such as Joint Direct Attack Munitions (JDAMS). This new-era air operation of precision strikes and information networks was blended with ground operations by friendly Afghan forces from the Northern Alliance to form an effective campaign for the ambitious mission at hand.3 At first, U.S. combat operations commenced slowly because of the time needed to deploy forces to the region, establish suitable bases and logistic support, and secure support from friendly governments, including Uzbekistan and Pakistan. But by October 19, SOF teams and Central Intelligence Agency (CIA) operatives were in place with Northern Alliance forces, and CENTCOM was able to command a posture of nearly 400 aircraft and 32 ships, including two U.S. aircraft carriers. Over the next two months, major success was achieved. U.S. air strikes quickly destroyed Taliban bases, headquarters, air defenses, and logistic support. On the ground, Northern Alliance forces were lightly armed and outnumbered by the enemy by a margin of two-to-one. Supported by U.S. precision air strikes, nonetheless, Northern Alliance forces steadily overpowered Taliban and al Qaeda resistance. Key towns in northern Afghanistan—including Taloqan, Konduz, Herat, and Mazar-e Sharif—fell over a three-week period. On November 9, Kandahar, the enemy’s last urban stronghold in southern Afghanistan, fell. On November 13, the enemy abandoned the capital city, Kabul, without a fight. By December 22, U.S. officials were attending a reception in Kabul celebrating the victory and installation of a new pro- American government under Hamid Karzai. Although enemy forces had been routed on the battlefield and the Taliban regime removed, the war was not yet over. Intelligence reports showed that remnants of Taliban and al Qaeda forces were gathering in Afghanistan’s eastern White Mountains, near the Pakistan border. In particular, large numbers of enemy appeared to be assembling in the Tora Bora region, to include Osama Bin Laden and other top al Qaeda figures. A major attack was launched on Tora Bora November 30. It included the now-familiar combination of U.S. SOF, strike aircraft, and friendly Afghan ground troops. This time, 3 For an analysis of the Afghanistan war, see Stephen D. Biddle, Afghanistan and the Future of Warfare: Implications for Army and Defense Policy (Carlisle, PA: Strategic Studies Institute, U.S. Army War College, 2002.)

3

4

however, the Afghan troops were not from the battle-tested Northern Alliance, but instead from a local Pushtun militia under a warlord named Hazrat Ali. The U.S. bombing was intense, but the Pushtun militia failed to perform effectively. The result was that although several hundred enemy troops were killed or captured, a larger number escaped into Pakistan. Reports suggested that Osama bin Laden and his aides were among those who escaped. They were able to flee largely because insufficient allied ground troops were available to block likely escape routes. At the time, no sizable U.S. ground forces were deployed to Afghanistan. In many quarters of the U.S. military, Tora Bora was seen as a frustrating failure and a lost opportunity that drew attention to the need for more U.S. ground forces if similar operations were to be launched in the future. In important ways, the battlefield at Tora Bora set the stage for Operation Anaconda the following March. The Intelligence Picture: Enemy Forces in the Shahikot Valley In mid-January 2002, U.S. officials began receiving intelligence reports suggesting that enemy forces, including al Qaeda, were assembling in the Shahikot Valley. This valley was a natural place for the enemy to regroup its forces after its earlier defeats. Located in Paktia province, which borders Pakistan, the Shahikot Valley is about 80 miles southeast of Kabul, and 18 miles south of Gardez. At an altitude of 7,500 feet, it runs on a southwest-to-northeast axis. The valley is relatively small, about five miles long and two and a half miles wide. On the valley’s floor are four small towns: Marzak, Babulkhel, Serkhankhel, and Zerki Kale. Surrounding the valley are high mountainous ridges. On the western side is a humpback ridge called the “Whale,” which is four miles long, one mile wide, and almost 9,000 feet high. On the eastern side is a high ridgeline culminating in the south at a peak called Takur Ghar: at an elevation of 10,469 feet, it is the highest peak in the valley. Jutting into the valley’s southern end is an arrowhead-shaped ridgeline called the “Finger.” With its high ridgelines and limited access routes, the valley provided seclusion and natural protection to the gathering enemy fighters.4 The Shahikot Valley is relatively easy to defend. Its high ridges provide many natural caves, crevasses, and other protected locations for guerilla fighters to establish positions capable of raking the valley floor and access routes with gunfire. Twice during the 1980’s, the Soviets mounted assaults against the Shahikot Valley with attack helicopters, artillery, and infantry. On both occasions they withdrew in retreat, driven back by fierce resistance. In 2002, the U.S. military possessed capabilities, especially modern information networks and precision strike weapons, which surpassed the weaponry fielded by the Soviet Army. But the rugged terrain, high altitude, and cold foggy weather had not changed, thereby making it hard for an attacking force to operate. Events were to

4 For a journalistic analysis of Operation Anaconda, see Sean Naylor, Not a Good Day to Die: The Untold Story of Operation Anaconda (New York: The Berkeley Publishing Group, 2005). See also: Mark G. Davis, Operation Anaconda: Command and Confusion in Joint Warfare, thesis, School of Advanced Air and Space Studies (Air University, 2004); Christopher F. Bentley, “Afghanistan: Joint and Coalition Fire Support in Operation Anaconda,” Field Artillery, September-October, 2002.

5

show that the Taliban and al Qaeda fighters of 2002 intended to take full advantage of the valley in mounting an effort to repulse the latest attacker. As U.S. military officials began contemplating an assault on the Shahikot Valley in late January 2002, they faced a major drawback—lack of good intelligence on the number of enemy fighters and their weaponry. U.S. officials relied upon multiple sources of intelligence, including human intelligence, communications intercepts, and overhead reconnaissance. Even so, getting accurate and reliable intelligence was a serious problem. The difficulty owed partly to the rugged inaccessible terrain, and partly to the enemy’s skill at concealing and camouflaging itself. The lack of good intelligence was not due to any lack of effort. Indeed, during the two weeks before the battle, U.S. SOF sent several reconnaissance teams to get as close to the valley as possible without being detected, in the hope of getting a better picture of the enemy. Even this effort failed to produce an accurate, definitive picture. Judging from outward appearances, the valley seemed largely devoid of people: only small clusters were periodically detected. Confronted with this uncertainty, U.S. officials did their best to generate estimates of the enemy’s strength. Initially, intelligence estimates suggested that anywhere between 100– 1000 enemy fighters could be in the valley. But gradually, a consensus emerged that the number was likely 200–300 fighters with light weapons. In addition, the intelligence estimate suggested that 800–1000 Afghan civilians were living in the four villages on the valley floor, complicating any effort to bring U.S. firepower to bear without causing major civilian casualties. The picture created thus was not one of a formidable force, but instead a relatively weak and demoralized foe merged with a larger civilian population. U.S. officials contemplated how the enemy might react to an American assault composed of ground forces and precision air strikes. The consensus was that the Taliban and al Qaeda fighters, readily defeated in the past and facing overwhelming firepower again, likely would not put up staunch resistance. A common belief was that they were mostly living in the valley’s villages, rather than deployed in the surrounding mountains and ridgelines. Another common belief was that they would try to flee the valley, and if that failed, try to negotiate a peaceful surrender. Thus, an invasion of the Shahikot Valley was deemed likely to take the form of some intense fighting at the start, followed quickly by police action aimed at arresting enemy fighters while ensuring innocent civilians were not harmed. For this reason, U.S. officials preferred to rely upon friendly Afghan soldiers to enter the valley floor and perform the arresting: they were judged better able than U.S. troops to separate al Qaeda fighters from innocent civilians. The role of U.S. ground troops was to block escape routes created by narrow passageways through the mountains on the valley’s eastern side. An extended, bitter battle for control of the valley was not on the minds of U.S. officials who designed the Operation Anaconda plan. Subsequent events showed that the U.S. attack on the Shahikot Valley was based on a faulty intelligence estimate. The actual number of fighters was considerably higher, perhaps 700–1000. They were more heavily armed than thought: they had heavy machine guns, rocket-propelled grenades (RPGs), mortars, and even a few artillery pieces. Most were not living in the villages, but instead were heavily deployed in the mountains and

6

ridgelines. They occupied already prepared, concealed positions that gave them good fields of fire throughout the entire battle zone. Their machine guns and mortars were pre- sighted to provide accurate fires on critical points. The villages were almost deserted, with few fighters and even fewer civilians. Moreover, the Taliban and al Qaeda fighters had no intention of ceding the valley without a fight or surrendering. Indeed, they intended to make a decisive stand in the hope of bloodying American and friendly Afghan troops. When the Anaconda battle began, they declared a ‘jihad.” Instead of enemy troops fleeing the valley, sizable numbers began entering the valley from outside, primed to fight the Americans. Inaccurate intelligence of enemy capabilities and intentions thus was the first thing that went wrong during Operation Anaconda. As U.S. and friendly Afghan troops approached the Shahikot Valley on March 2, they were expecting a battle involving light, brief combat followed by an easy, overwhelming victory. The entire operation was to be over in three days, and major fighting was to take place, if at all, only on the first day. What transpired was the opposite: a hard, pitched battle that dragged on for several days, during which the enemy put up fierce resistance and was dislodged only after intense U.S. bombing. This dramatic, unexpected change in enemy resistance was a main reason why, as General Franks said, the original battle plan did not survive initial contact with the enemy. The U.S. Military Command Structure for Anaconda: Multi- Headed A hallowed principle of war is unity of command that is military campaigns and battles should be commanded by a single senior officer with the authority and staff assets to blend the operations of all components into a single, cohesive plan. At the time Operation Anaconda began, unity of command had not been established because the U.S. military presence in Afghanistan was not yet fully mature. The earlier phases of the war had been conducted without a joint commander and command staff in Afghanistan. U.S. combat operations were directed by CENTCOM, led by General Franks, working through two main subordinate commands: Coalition Forces Land Component Command (CFLCC) and Coalition Forces Air Component Command (CFACC). CFLCC was led by Army LTG Paul Mikolashek and CFACC was led by U.S. Air Force Lt Gen Michael Moseley. Both CFLCC and CFACC were located in the Persian Gulf, and they directed Afghanistan force operations from there. To the extent a ground commander existed in Afghanistan, it was Army COL John Mulholland, who led Task Force Dagger of the Army’s Fifth Special Operations Group, which provided SOF teams that worked with friendly Afghan troops and performed spotting for most of the U.S. air strikes. But outside Mulholland’s command were other SOF units and CIA operatives working in other parts of Afghanistan and/or performing different missions. When SOF forces on the ground requested air strikes, the decision of whether and how to carry them out was in the hands of CFACC, which worked through its Combined Air Operations Center (CAOC), also located in the Persian Gulf. In previous battles, modern satellite communications had permitted command and control of the air operation from such a long distance with successful results.

7

From mid-January to mid-February, interest grew in assaulting the Shahikot Valley and employing sizable U.S. ground combat forces as part of the operation. In response, Generals Franks and Mikolashek reached the conclusion that a U.S. tactical commander was needed in Afghanistan. They turned to MG Franklin Hagenbeck, commander of the Army’s 10th Mountain Division. Shortly before, General Hagenbeck and his division staff had deployed to K-2, a U.S. base in Uzbekistan that had been used during the earlier stages of the war. In mid-February, Hagenbeck and his staff deployed to Bagram Air Base, north of Kabul in Afghanistan, from where they were to command the assault on the Shahikot. As the highest-ranking U.S. military officer in Afghanistan and directly subordinate to CFLCC, General Hagenbeck had the seniority to command the assault on the Shahikot Valley. But he did not arrive at Bagram with his full division staff or all of his combat units. At the time, 10th Mountain Division (which is assigned only two maneuver brigades instead of the standard three) had about one-half of its division headquarters and a brigade combat team in Kosovo, a battalion task force in Bosnia, and a battalion task force in the Sinai. This left General Hagenbeck with only one-half of his normal division headquarters, neither of his assistant division commanders, and only a single infantry battalion for operations in Afghanistan. General Hagenbeck was able to increase manning of his division headquarters by drawing upon replacements from elsewhere in the Army, but at a cost of continuity and experience with 10th Mountain operations. In particular, new personnel helped populate his intelligence staff and his air liaison staff for coordinating air-ground operations with CFACC and CAOC. Expertise in both of these areas was necessary for Operation Anaconda. In an effort to strengthen General Hagenbeck’s authority, CENTCOM named his headquarters Coalition and Joint Task Force (CJTF) Mountain and gave it command and control authority over Operation Anaconda. This meant that General Hagenbeck commanded his own divisional units, the 3rd brigade of the 101st Air Assault Division (which had recently deployed to Kandahar, Afghanistan), Task Force Dagger, and other SOF units, for example Task Force K-Bar and Task Force Bowie. But General Hagenbeck did not have command over Task Force 11, a key SOF unit for advanced force operations, such as reconnaissance and strikes against high value targets, which remained under separate operational command. Task Force 11, its subordinate SOF teams, and assigned Army Rangers were slated to play an important role in Anaconda even though they remained outside General Hagenbeck’s command structure. In addition, General Hagenbeck was not granted command over the U.S. air component forces—from the Air Force, Navy, and Marines—that were slated to provide support to Anaconda ground operations. Instead, these forces remained under command of CFACC and its CAOC. Indeed, authority for every strike sortie and target remained in the hands of CFACC and CAOC, which applied tactical principles, knowledge of available air forces, and CENTCOM rules of engagement (ROEs) to the decisions at hand. General Hagenbeck and his ground commanders could only request air strikes, not order them. During the planning phase for Operation Anaconda, this bifurcated air-ground command structure was not deemed troublesome because a major close air support (CAS) effort was considered unnecessary. But matters were to change when the battle unfolded, an

8

intense CAS effort became necessary, and the bifurcated command structure resulted in strained operations that left both Army and Air Force officers frustrated. Finally, General Hagenbeck did not have command authority over friendly Afghan forces slated to participate in Operation Anaconda. As occurred earlier in the war, these forces worked closely with U.S. SOF teams, principally Task Force Dagger. But they were not under the command of Task Force Dagger. They operated according to their own command structure, principles, and designs. The common practice was to establish a consensus among SOF teams and friendly Afghan units before an operation began. Earlier, this practice worked successfully, and Northern Alliance forces proved reliable in carrying out their commitments. The consensus process was employed again in the period before Anaconda, but it was no guarantee that friendly Afghan forces would fully perform the roles assigned to them in the U.S. battle plan. Because they were not commanded by General Hagenbeck or Task Force Dagger, the friendly Afghan forces had the freedom to depart from the original plan if, in their judgment, circumstances so dictated. Events were to show that they exercised this freedom when trouble was encountered in their march to the Shahikot Valley, in ways that damaged the entire operation and increased the danger to U.S. forces. Thus, the reality was that as the plan for Operation Anaconda took shape in late February, it did not benefit from unity of command. Instead, the command structure was multi- headed. General Hagenbeck and Task Force Mountain headquarters commanded most of the U.S. ground forces in the battle, but not all of them. They did not command the U.S. air forces assigned to support the battle plan. Nor did they command the friendly Afghan units that were intended to provide a main axis of attack in the battle plan. This multi- headed command structure was not deemed a problem when the original Anaconda battle plan was forged in ways that fitted its components together in a cohesive manner. But the command structure did become a problem when the original plan broke down and new plans had to be quickly realized among multiple forces that responded to multiple commanders. U.S. and Friendly Afghan Forces for Anaconda: Assets and Deficiencies A second guiding principle of warfare is mass: the need to have sufficiently large and well-equipped forces for an operation. In the weeks preceding Operation Anaconda, U.S. ground forces were arriving in Afghanistan, but the buildup was constrained by Washington’s desire to keep a relatively low military profile to prevent the appearance that a massive, Soviet style occupation was taking place. No formal manpower cap was established, but in his memoirs General Franks stated that the CENTCOM goal was to keep the U.S. military presence from rising above 10,000 troops.5 This troop level translated into deployed SOF units plus the equivalent of one to two U.S. Army brigades along with combat support and combat service support units. As a practical matter,

5 See Franks, American Soldier, 324.

9

General Hagenbeck was able to draw upon a light infantry battalion of the 10th Mountain Division, commanded by LTC Paul LaCamera, and the 3rd Brigade of the 101st Air Assault Division, commanded by COL Frank Wiercinski. The 3rd Brigade, called Rakkasans (a title inherited from World War II), deployed only two of its three battalions to Afghanistan. The rest of the 101st Division remained in the United States, where it was needed for other potential contingencies. Because all three of the battalions from both divisions had only two of their companies present, the U.S. ground force for Anaconda totaled six infantry companies. This yielded a total of about 600 infantry troops. When combined with the roughly 400 friendly Afghan troops also available, plus about 200 SOF troops from other countries, the result seemingly was sufficient manpower to carry out the assault on the Shahikot Valley and block escape routes from it. Nonetheless, this ground posture suffered from important deficiencies that were to become apparent only after the assault was launched and events took a turn for the worse. In order to keep a low profile, the infantry battalions of the 101st and 10th Mountain Divisions were sent to Afghanistan without any tanks, infantry fighting vehicles, or artillery. These heavy weapons were deemed unnecessary by CENTCOM—air forces presumably could make up any unanticipated deficiency in firepower—and keeping them out of Afghanistan lessened any comparisons to the tank-heavy Soviet Army. 101st Division commanders requested that a battalion of AH-64 Apache attack helicopters be deployed, but they were granted only a single company of eight AH-64s. In addition, they were given only 13 Chinook lift helicopters and eight Black Hawk utility helicopters: less than the normal number. Support troops also were kept low: the brigade had a signals package, a military police battalion, and some other units, but lacked its normal complement of logistic support. This limited presence was tailored to allow the 3rd Brigade to perform the missions that were deemed likely to occur, but it left the brigade lacking the heavy weapons and equipment that could be needed for intense combat against serious opposition. The same constraint applied to the battalion from the 10th Mountain Division, which also lacked armor, artillery, and attack helicopters. The result was that together, all three U.S. Army battalions had a fairly large number of infantry troops, but they were lightly armed—mainly with rifles, machine guns, and some mortars. Their only heavy firepower came from a small number of AH-64 Apache attack helicopters and from a few AC-130 gunships, which were under the command of SOF units and flew only at night. The absence of tanks and artillery meant that they could not turn to the heavy weapons normally employed by the Army, along with infantry and helicopters, to wage major combat operations and generate firepower against targets located more than a few hundred meters away. Their mortars in theory gave them some capability for delivering firepower at longer ranges, but when the assault was launched, the Army units, not expecting heavy combat and constrained by the limited number of lift helicopters, brought only a few light-to-medium mortars and only a single heavy, 120 mm mortar with a limited supply of ammunition. Problems also afflicted the friendly Afghan troops available for the assault. They too were light infantry, lacked artillery and tanks, and had only a few mortars. In addition,

10

they lacked vehicles for traveling to the Shahikot Valley: U.S. officials had to assemble a diverse pool of civilian trucks, pickups, and other utility vehicles for the job. Most important, these troops were not from the Northern Alliance, whose forces were battle trained, tested, and capable of serious operations. Instead, they were mainly local Pushtun militia from the so-called “Eastern Alliance” (a loose collection of Pushtun militias) under command of a local warlord named Zia Lodin. Lodin seemed a good, respected leader and was eager to cooperate, but he and his subordinate commanders lacked experience in big battles. Individually, the Afghan fighters were familiar with infantry weapons and many had seen guerilla combat before. But they were not organized into a formal force. Nor were they trained for the maneuver-and-fire operations typically performed by combat units. In the weeks before Operation Anaconda was launched, U.S. SOF advisers set about to organize these forces and provide them training in combat operations. The force that emerged was organized into a single battalion with three companies, which were subdivided into platoons and squads along U.S. lines. But normally, a period of two or three months is required to train such units to the point where they can perform complex combat operations. The time available for training was less than that: only a month or so. U.S. SOF advisers made the best use of this time and expressed belief that this Afghan force could be relied upon to perform its missions during Operation Anaconda. Yet the reality was that the Pushtun militia were not yet fully ready for intense combat against a tough opponent. As the date to launch Operation Anaconda approached, General Hagenbeck and his staff thus presided over a coalition force composed of U.S. Army troops and friendly Afghan fighters, plus some SOF from other countries. On paper, this ground force may have provided sufficient manpower to perform the mission, but it was almost entirely light infantry, supported by only a few attack helicopters and gunships. It lacked armor, artillery, and heavy mortars. Although the U.S. component was superbly trained and ready, the Afghan component, which was to provide the main effort for the attack, was not fully trained for demanding fire-and-maneuver tactics or multi-unit operations. Perhaps this coalition force would have been amply capable of performing the mission that was anticipated: a brief period of intense combat followed quickly by police action aimed at arresting enemy fighters in the valley and as they tried to escape through mountain passes. But the battle that actually emerged was far more demanding, and it exposed the deficiencies of both parts of this force. It also compelled U.S. commanders to turn to air forces to provide the firepower needed to defeat the enemy: a development that had not been anticipated and for which preparations were not fully made in the period before Operation Anaconda was launched.

Developing the Anaconda Battle Plan: Hammer and Anvil During the three weeks before the assault on the Shahikot Valley was launched, Task Force Mountain and its subordinate units conducted a concerted effort to prepare a sound battle plan. Careful attention to details was needed not only because Operation Anaconda was to be the biggest set-piece battle of the Afghanistan war, but also because it involved many moving parts, all of which had to be coordinated together. In his memoirs, General Franks said that when he first was briefed on the impending plan, he was not satisfied

11

that it was sufficiently joint and properly integrated. But after the plan was improved, he became satisfied, and he approved it February 17, as did General Mikolashek.6 On February 20, General Hagenbeck issued his formal operations order for the plan. The assault was originally scheduled to be launched February 25, but was changed to February 28 and then to March 2 in response to bad weather and to provide more time to train Zia’s force. Task Force Mountain selected the name Anaconda for a reason. An anaconda is a large constricting snake of the boa family that coils around its victim, thereby crushing him to death. U.S. military planners conceived the idea of surrounding the Shahikot Valley with several concentric rings that would block enemy entrance to and exit from the valley. The outer rings were to be composed of U.S. SOF, friendly Afghan troops, and special forces from Australia and several other participating nations. The assault on the valley itself, however, did not take the form of a ring. Instead, it was to be a classic “hammer and anvil” plan. The day before the battle, several SOF teams under Task Force 11 were to establish themselves on high ground at the northern and southern ends of the valley, thus enabling them to see the entire valley, observe the arrival of U.S. and friendly Afghan troops, and help provide spotting of targets for air strikes. Shortly before the arrival of ground troops on D-Day, a limited number of U.S. air strikes were to be conducted against 13 targets, principally machine gun positions on the Whale and other ridgelines. The ground assault on the valley was then to be conducted at dawn by two task forces. Task Force Hammer, Zia’s force guided by U.S. SOF troops, was to move by trucks, drive south along the road from Zermat, pass the Whale on its western side, and then enter the valley from the south at a position known as the “Fishhook”: the terrain between the Whale and the Finger. At the same time, Task Force Rakkasans—U.S. infantry troops from the 10th Mountain Division and the 101st Air Assault Division—was to fly by Chinook helicopters into the valley, enter from the south, and land along the slopes of the valley’s eastern ridges. As its name implies, Task Force Hammer was to function as the “hammer” of the battle plan. Its mission was to advance across the valley floor—called “Objective Remington”—and engage enemy forces and destroy or capture them. Meanwhile, Task Force Rakkasans was to act as the “anvil” by establishing blocking positions at the passes leading through the mountains on the valley’s eastern side. U.S. military planners decided to establish seven blocking positions, which were named (from south to north) Heather, Ginger, Eve, Diane, Cindy, Betty, and Amy. They further determined that four infantry companies—roughly 400 troops—would be needed to populate these seven blocking positions with enough troops to secure them. The battalion of the 10th Mountain Division, with two companies, was assigned to the three southern blocking positions. One of the 101st Air Assault Division’s battalions, with two companies, was assigned to the four northern blocking positions. This deployment left the second battalion of the 101st Air Assault Division, with two companies, as an uncommitted tactical reserve, capable of responding to unexpected developments and to adaptations in the plan. 6 Ibid., 378.

12

U.S. military planners regarded successful interaction of both the hammer and anvil as critical to a victorious outcome. The hammer’s job was to attack the enemy on the valley floor, and the anvil’s job was to prevent the enemy from fleeing, thus leaving Taliban and al Qaeda fighters exposed to the attack by Zia’s troops. Both the hammer and the anvil were required to arrive at the same time, and to perform like clockwork, trapping the enemy while avoiding the risk of friendly fires against each other. Accordingly a great deal of effort was expended to ensure that the necessary resources were in place and that the plan could be fully executed as designed. U.S. SOF forces trained Zia’s troops for the battle, assembled sufficient trucks and vehicles to drive them to the valley, and reconnoitered the Zermat road in advance. Meanwhile, U.S. forces practiced helicopter lift and transit operations, established helicopter landing zones and refueling points, and flew in large amounts of fuel, ammo, and other stocks in order to keep Operation Anaconda properly supplied. Considerably less attention, however, was devoted to preparing for tactical air support of Operation Anaconda in event that large numbers of air strikes became necessary. This was the case largely because the operation was viewed by Task Force Mountain and CFLCC as being mainly a ground assault in which air forces would play only a minor supporting role. Thus, the ground and air components did not enter the battle with an accurate common operational picture of how events actually would transpire, and of the critical role that the air forces would be compelled to play. CFACC and CAOC were not involved in the initial planning and officially became aware of the impending operation only when General Hagenbeck’s operations order was issued February 20. Air Force liaison officers assigned to Task Force Mountain were more involved, but did not exert a great deal of influence over the plan. Initially they proposed heavy saturation bombing of the valley for two days before ground forces entered it, but although this idea was supported by CFACC/CAOC, it was rejected by Army officers because it would compromise surprise, thus enabling enemy fighters to flee before the ground assault began. As a result, the air component’s main job was to airlift supplies to Afghanistan, provide strike aircraft for the limited number of targets to be attacked on the first day, and assign an adequate number of on-call CAS aircraft to orbit the battlefield for the three- day operation in event their use became necessary. CFACC and its subordinate units performed these functions. On the morning Operation Anaconda was launched, U.S. combat aircraft and command-and-control aircraft were circling over the Shahikot Valley area in greater numbers than previous weeks, available to respond to calls for support. General procedures for providing air strikes in support of ground forces had been established. Even so, the ground-oriented nature of the battle plan meant that before D-Day, neither U.S. ground forces nor U.S. air forces had engaged in the kind of close, careful cooperation and joint planning that normally would have been deemed necessary to mount a major CAS operation from the onset of the battle onward. Prior to Operation Anaconda, U.S. air forces had mainly been conducting pinpoint strikes against fixed targets, not performing CAS strikes against moving targets in close proximity to U.S. ground forces. Neither the 10th Mountain Division nor the 101st Air Assault Division had recent combat experience in joint operations requiring CAS strikes in support of maneuvering ground units. U.S. ground forces entered the battle unaware of

13

the intense air support that they would soon be calling upon the air forces to provide. U.S. air forces entered the battle not expecting the urgent requests for major, sustained help from the ground forces that would soon be directed at them. As a result, not all of the assets needed to launch a major CAS effort were in place on the day Anaconda began, and the necessary steps to ensure effective collaboration between the ground and air components had not fully been taken. Events were to show that when the call came for a major CAS effort, it began raggedly in some respects, but improved as the battle progressed. Task Force Mountain did not approach Operation Anaconda blind to the possibility that unexpected events might occur. It prepared “branches and sequels” that provided variations on the main battle plan, and it assembled reinforcements—two infantry companies and on-call CAS—that could be committed if necessary. But in the harried days before D-Day, it attentions primarily were devoted to developing the main battle plan. Task Force Mountain was focused on the demanding task of assembling the assets for the coming planned assault, not on preparing and mobilizing resources for the kind of major adaptation that proved necessary. A major improvised CAS effort, and use of ground reinforcements, likely would not have been necessary had the hammer-and-anvil ground operation unfolded as planned. But this was not to be the case mainly because the hammer—Zia’s Afghan troops—failed to carry out their attack into the valley, and instead withdrew from the battle on the first day. Hammer and anvil attacks have been standard military doctrine for many years, and often have proved successful. Modern information networks have made them increasingly feasible by permitting constant communications among widely dispersed units. But they also can be risky ventures because they split the attacking force into two or more parts, and expect them to launch coordinated attacks against an enemy that enjoys the advantage of its forces gathered in one place. When one part of the attacking force fails to perform its mission, the other part can be left outnumbered and vulnerable to the full wrath of the enemy. This was to be the case in Operation Anaconda, when Zia’s unexpected withdrawal left deployed U.S. infantry troops alone on the valley’s eastern slopes, vulnerable to concentrated enemy fires. The Battle Plan Unravels: Troubles in the Shahikot Valley Operation Anaconda began unraveling even before H-Hour at 6:30 a.m. March 2, when Zia’s force of friendly Afghan fighters and U.S. ground troops were scheduled to arrive in the valley. One hour before, a B-1 bomber, a B-2 bomber, and two F-15E fighters were scheduled to bomb the 13 enemy targets in the Shahikot Valley, three of them on the Whale. The aircraft arrived on schedule, but as they began their bombing, they received a radio message, possibly from a SOF unit on the ground, to cease bombing because friendly troops were endangered. The aircraft halted their operation, with only about one- half of their targets bombed. Meanwhile, Zia’s force, Task Force Hammer, guided by U.S. SOF personnel, was encountering trouble in its drive to the Whale. A column of about 35 trucks and other

14

vehicles left Gardez about 1:30 a.m., which in theory provided enough time to reach the Fishhook by 6:30 a.m. The dirt road, however, was primitive, and several trucks, driven by unskilled Afghan fighters, slid off the road and/or broke down. Gradually the truck column became strung out over a distance of three miles. Along the way, several vehicles, led by U.S. SOF personnel, left the column in order to drive eastward to the north end of the Whale to a position known as the “Guppy” in order to establish a blocking position to prevent enemy fighters from fleeing through a pass. As they approached the Guppy, they came under fire from a U.S. AC-130, which mistook them for enemy personnel. A U.S. SOF soldier and two Afghani troops were killed and another SOF soldier and 12 Afghans were wounded. When the main column became aware of this incident, it sent several vehicles to the Guppy to provide help, and then continued on its drive toward the Whale. At about 5:30 a.m., it paused—near the village of Carwazi, two miles from the Whale—to witness the planned U.S. air bombardment of the Whale. Evidently, the Afghan soldiers were expecting a massive saturation bombardment lasting nearly an hour. When only three bombs were heard exploding on the Whale, they began to become demoralized. The truck column then recommenced its slow drive toward the Whale, with headlights on in order to see the road. About one mile from the Whale, the column unexpectedly came under fire from enemy mortars on the Whale. The troops dismounted from their trucks and began returning fire with the few mortars in their possession. Their U.S. SOF guides wanted the Afghan troops to walk on foot the remaining distance to the Fishhook, but Zia and his fighters balked. Instead, they initially hunkered down at their current position and remained there for a period. Their SOF guides called for CAS strikes against the enemy, but only one aircraft was available because, by this time, most U.S. aircraft were busy supporting the U.S. Army troops that had entered the Shahikot Valley on the eastern side and were taking heavy enemy fire. Further demoralized, disorganized, and under enemy fire, the friendly Afghan fighters began retreating to Carwazi. They stayed there until mid-afternoon. At this juncture, Zia ordered his troops to return to Gardez, their original home base 18 miles from the Shahikot Valley. That day, Task Force Hammer took 21 casualties: one U.S. soldier killed and one wounded, and three Afghan troops killed and 14 wounded. The departure of Zia’s troops ended any prospect that the hammer of the original battle plan would perform its mission. The effect was to expose the anvil, U.S. Army troops, to the full wrath of the enemy. The first U.S. Army presence in the Shahikot Valley was scheduled to be six AH-64 Apache attack helicopters. One of them was grounded with maintenance problems, but the other five arrived. Their mission initially was to patrol the valley and check each landing zone for crew-served weapons and enemy troops. After the Chinook helicopters had arrived and deposited the ground troops, the Apaches were to provide fire support to the troops of Task Force Rakkasans. From the moment they swept over the valley, they began taking intense enemy fire from rifles, machine guns, and RPGs. They immediately began flying racetrack patterns over the valley, firing at enemy positions on the eastern ridges, the valley floor, and the Whale. They were scheduled to be on station for about 2 ½ hours, but within 90 minutes, two Apaches were so badly damaged that they had to leave the valley for their home base. The other three Apaches remained, delivering fire in

15

support of besieged U.S. ground troops. When their fuel was exhausted, they flew to their refueling and rearming station outside the valley, and then returned to the battle. They were also reinforced by the two Apaches that previously had not been committed. During the first day of Anaconda, they repeated this pattern several times, continuously returning to the battle to fire several hundred cannon rounds and rockets, plus a Hellfire missile. Their presence helped save many American lives that day, and they inflicted major damage on enemy positions, helping suppress fire against U.S. troops. At the end of the day, they returned to their home base, where inspection showed that all of them had been hit by multiple enemy bullets. They were able to continue operating only because the Apache was designed to be highly survivable in the face of small arms fire. U.S. ground troops arrived in the Shahikot Valley on schedule, landing at about 6:30 a.m.. The initial deployment included six Chinook helicopters carrying about 200 troops from both divisions: about one-half of the total invasion force. The remaining 200 troops of the invasion force were scheduled to arrive by Chinooks about 3 ½ hours later. Shortly after touchdown, the first wave of U.S. troops began coming under enemy fire, which grew in intensity. The 101st Division troops in the northern part of the valley encountered fire, but were able to move toward most of their blocking positions. Enemy resistance was especially fierce in the southern part, where troops of the 10th Mountain Division and the 101st Division’s brigade headquarters landed. The intense enemy fire drove many of the 10th Mountain troops to seek cover in a defensible position called “Hell’s Halfpipe,” located near BP Heather and Ginger at the southern end of the valley. For the entire day, the U.S. troops remained there, defending themselves under intense enemy fire. The scheduled deployment of the remaining 200 U.S. troops was temporarily postponed due to bad weather and the enemy fire. When night came, wounded troops were evacuated, and then the entire contingent in Hell’s Halfpipe was taken out by helicopter. On the second day, the remaining 200 troops were deployed along with a mortar unit that had several medium and heavy mortars. They were inserted into the northern part of the valley, where enemy fire was less intense, and proceeded to march southward in an effort to man all seven battle posts. Later, the Task Force Rakkasans tactical reserve of two companies was also committed into the valley, thus bringing the total deployment to over 500 troops. Over the course of the following days, during which they received growing CAS support, they took control of the valley’s eastern passes and established most of the seven blocking positions. Enemy resistance around BP Ginger, however, remained fierce and several days were to pass before it was secured by U.S. troops. No U.S. troops were killed in the fighting. Nonetheless, multiple troops were wounded, some of them seriously. The casualties could have been greater, but U.S. troops benefited from their helmets and body armor, and enemy fire was often inaccurate. Overall, the U.S. Army troops performed valiantly and effectively in a difficult situation, thereby carrying out their portion of the original battle plan, but over a longer time period than anticipated. Fatalities occurred the night of March 3–4 when Task Force 11 attempted to insert a reconnaissance team of Navy SEALS onto the peak of Takur Ghar, the highest point in the valley. A Chinook carrying the team arrived at Takur Ghar at about 2:30a.m., March 4. When the Chinook lowered its ramp, it immediately came under enemy fire and was

16

badly damaged by hits from two RPGs. The Chinook flew away in order to escape, but a SEAL, CPO Neil Roberts, fell out of the back ramp and landed on the ground. The damaged Chinook flew to a safe landing zone some distance away, and the team of Navy SEALS bordered another helicopter, and flew back to Takur Ghar in order to rescue Roberts (events were later to show that Roberts already had been captured and killed by the enemy). When the Chinook landed atop Takur Ghar, its disembarking SEALS again came under fire and one was killed. The survivors retreated to a defensible position and began returning enemy fire. Two hours later, another Chinook arrived, carrying a TF-11 Quick Reaction Force of nine Army Rangers. They too came under immediate fire: four were killed, others were wounded, and the Chinook was downed. Later, another Chinook landed with 35 Army Rangers. CAS support was committed to help kill and suppress the enemy. By early afternoon, the fight for Takur Ghar had been won, but at a cost of seven dead and six wounded. Battlefield Adaptation: Air Forces and CAS to the Rescue Had U.S. ground troops been left alone in the Shahikot Valley, on their own, they might have been in serious trouble for many days and taken considerably more casualties than was the case. Their lack of artillery and heavy mortars prevented them from generating suppressive fires against enemy troops that were firing on them with heavy machine guns and mortars from all directions, including from the Whale. But they were not left alone. As events began taking a negative course March 2, Task Force Mountain issued a call for help from U.S. air forces, which were not originally viewed as an integral part of Operation Anaconda, but now had become vital. That day, the Anaconda battle plan underwent a major adaptation. Task Force Mountain rejected the idea of abandoning the offensive and withdrawing U.S. troops from the valley. Instead, it switched from its original hammer-and-anvil focus to embrace a new plan that emphasized the massing of air fires in support of U.S. Army positions on the valley’s eastern sides. This adaptation proved to be the key to winning the battle, but as it was carried out, it encountered some frictions along the way. A decade or two earlier, the type of intensive CAS campaign launched during Operation Anaconda would have been impossible. After the Vietnam War, the Air Force and Navy turned away from preoccupation with CAS to emphasize air intercept, interdiction, and strategic bombardment. The Air Force procured the A-10 for CAS, but it was a small part of its inventory, which was dominated by jet fighters. In recent years, however, major improvements in their CAS capabilities had been achieved. An important improvement was the installation of advanced avionics that enabled jet fighter planes to conduct CAS strikes accurately—a big change from the aircraft of the Vietnam War. Creation of precision weapons, such as JDAMs, also made a big difference. Equally important was the development of laser target designators and GPS systems, which enabled precise identification of target locations from the ground or air, and between ground and air forces working together. Finally, creation of modern information networks permitted high-speed communications among air commanders, command-and-control aircraft, tactical combat aircraft, and ground forces. As a result, air forces at Anaconda were able

17

to promptly embark upon a major CAS campaign even though the original battle plan had not called upon them to do so. Task Force Mountain was able to draw upon air forces—provided by the Air Force, Navy, and Marine Corps—that totaled over 200 bombers, fighters, and combat support aircraft. Additional aircraft, including A-10s and AC-130s, were flown into the region in order to support the air bombardment campaign. Additional Army Apaches and Marine Corps helicopters also were sent. Over the following days, strike aircraft flew an average of about 65 combat sorties per day in support of U.S. ground forces. Over the course of the entire battle, a total of about 900 strike sorties and 2,000 associated support sorties were flown. U.S. aircrews performed with valor and skill throughout the battle. Benefiting from air-to-ground information networks, U.S. strike aircraft employed a combination of ordnance, including JDAMs, other precision-guided munitions, conventional iron bombs, rockets, and cannon fire. Many of the strikes were conducted at high altitude, but when necessary, low-level strafing attacks were made as well. The effect was to deluge the Shahikot Valley with a great deal of firepower, thus suppressing enemy fires and killing large numbers of enemy troops. In the initial days, this CAS effort encountered some troubles that reflected its improvised nature and rough edges that had to be worked out. Normally Army officers would have had artillery and mortars at their disposal, which could have been used to provide wide-area fires, including surge fires that can deliver a great deal of ordnance quickly. In this event, air forces could have been used for their normal purpose of delivering precision strikes against targets that could not be reached or suppressed by Army fires. At Anaconda, artillery and large numbers of heavy mortars were not available. As a consequence, air forces had to be employed not only for their normal purpose, but also to provide a substitute for artillery and mortars. The air forces were able to accomplish both tasks, but the process was not easy or quickly accomplished.7 One problem that arose during the CAS effort was command relationships. Although Task Force Mountain and its ground troops were the main consumers of the CAS campaign, CFACC and CAOC initially retained command authority for all strike sorties. As a result, requests for strike sorties from ground units had to be passed to CFACC/CAOC, or at least to their orbiting command and control aircraft, for permission. All requests for emergency CAS strikes were immediately granted, and most other CAS strikes were conducted within a few minutes of receiving the request. Getting permission for non-emergency strikes, however, occasionally proved time-consuming: up to 25–45 minutes in the opinion of frustrated Army ground commanders. Such delays were a product of multiple factors, including the need to: verify and prioritize targets, enforce ROEs, determine proper aircraft and munitions, determine ingress and egress routes, and manage the large volume of requests coming from the ground forces, which sometimes exceeded the number of aircraft immediately available. 7 For an historical perspective, see David E. Johnson, Learning Large Lessons: The Evolving Roles of Ground Power and Air Power in the Post-Cold War Era (Santa Monica, CA: RAND) 2006.

18

Strike aircraft were continuously available around-the-clock, but the geography of the Shahikot Valley impeded the CAS operation. Because the valley was only five miles long, thus creating a risk of mid-air collisions, it limited the number of aircraft that could be striking targets at any single time. Enemy targets in the mountains were normally small and hard to see. Orbiting aircraft relied upon the ground forces to identify targets and pass precise locations to them, in the form of GPS-generated coordinates, so that precision strikes could be made. Although several USAF officers and over 30 Enlisted Forward Air Controllers (EFACs) were assigned as liaisons to the ground forces, they were not always properly equipped. A number did not have laser target designators with them, and thus were unable to use GPS to determine precise coordinates for targets. As a result, they had to use tactical maps for this purpose, and many of their maps were not detailed enough to determine exactly where targets were located. When voice communications had to be used between ground and air forces, aircraft pilots sometimes had trouble visually acquiring the targets they were being asked to strike. On some occasions, air commanders overruled requests from ground forces for non- emergency CAS strikes because from their perspective the criteria for authorizing them had not been met. On other occasions, air commanders felt that the ground forces did not grasp the flexibility available to them under the existing ROEs. For their part, ground commanders sometimes felt they were placed in the position of having to create new ROEs in order to handle the situation. Sometimes, the problem of handling multiple radio frequencies inhibited ground units from talking directly with fighter aircraft overhead. Army units carrying only line-of-sight FM radios were hard-pressed to talk to other units when the terrain blocked their communication. This sometimes inhibited their ability to coordinate their requests for air strikes. Sometimes, the SOF units, which had satellite radios and were best trained and equipped to guide CAS sorties, had their voices drowned out by calls for strikes from Task Force Rakkasans troops. These frictions complicated the air-ground interaction during Operation Anaconda’s initial days. Moreover, enemy targets often were so well protected by the surrounding mountains and ridges that hitting them with strikes was difficult. Exact hits were often necessary, and even precision JDAMs sometimes were not able to achieve this accuracy. Even when exact hits were achieved, only a single small target—perhaps two or three enemy fighters manning a heavy machine gun—was normally destroyed. Hundreds of enemy fighters were deployed in the mountains and ridges, thus creating a very large number of small targets, each of which had to be attacked individually. A further complication was that enemy fighters often would scramble for the protection of caves when they sensed an impending air attack, only to re-emerge after the ordnance had been delivered. The process of rooting them out by air strikes thus was slow, frustrating, and time-consuming. Several days of intense air bombardment were needed before enemy fires began abating noticeably. As the Anaconda battle progressed, many of the difficulties experienced in the initial days by the CAS operation were solved or receded greatly. Misunderstandings about ROEs were ironed out, thus permitting strikes against virtually all enemy targets.

19

By Day 3, CFACC and CAOC had transferred authority for allocating sorties to TF Mountain, which was better able to identify, verify, and prioritize targets in the ways best suited to the needs of the ground forces. The arrival of additional A-10s, AC-130s, and Apache attack helicopters provided a larger number of aircraft that were originally designed for CAS missions, thus lessening reliance upon high-flying fighters and bombers for hitting difficult targets with JDAMs. Also important, combat pilots, ground controllers, and Army infantry troops steadily improved their capacity to work together smoothly and effectively. By March 9, the steady accumulation of lethal air strikes over a one-week period was clearly weakening the enemy—by destroying many targets and influencing other enemy fighters to flee the battle rather than be killed. Events on the ground also began taking a favorable course. Several medium and heavy mortars were airlifted to the deployed ground forces, thereby enabling them to increase their fires against distant enemy targets. By March 9, U.S. ground forces finally took control over BP Ginger, thus closing off the last remaining pass out of the valley eastward. Afghan troops also began re-entering the battle. On March 4, Zia’s troops were persuaded to return and occupy the pass near the Guppy, but they did not perform their originally planned maneuver through the Fishhook and onto the valley floor. This task was left to Northern Alliance troops under Gul Haidar, which employed tanks and infantry to sweep over the valley March 12. By this time, most enemy troops had either been killed or fled the valley. Even so, mopping up operations continued for another week. Afterward: Remedying Problems As General Franks said, Operation Anaconda was a success. Although its initial battle plan failed to survive contact with the enemy, it achieved success by performing tactical adaptation under fire. Operation Anaconda showed that although rooting enemy forces out of mountains could be difficult, it could be accomplished. The manner in which problems arose and were solved showed the value of planning for joint networked operations with air forces, rather than relying upon lightly armed ground forces to accomplish demanding missions alone. The experience also showed that in the information age of complex battle plans, ground forces need to be well-armed, especially when conducting dispersed operations. Afterward, the battle generated controversy when some officers publicly complained about the performance of both the ground and air components.8 The controversy abated when the Department of Defense took steps to enhance joint coordination in the planning and execution of tactical combat, including steps to better mesh the performance of ground and air forces. During Operation Anaconda, both components learned valuable

8 For articles on the controversy, see Elaine M. Grossman, “Was Operation Anaconda Ill-Fated from Start? Army Analyst Blames Afghan Battle Failing on Bad Command Set-Up,” Inside the Pentagon, July 29, 2004, and “Anaconda: Object Lesson in Ill Planning or Triumph of Improvisation,” Inside the Pentagon, August 19, 2004. See also, Robert H. McElroy and Patricia Slayden Hollis, “Afghanistan: Fire Support for Operation Anaconda, Interview with Major General Franklin L. Hagenbeck,” Field Artillery, September– October 2002.

20

lessons. Army officers learned the importance of fully including the air forces in developing battle plans, even when these plans do not necessarily envision major CAS strikes. Air Force officers learned the lesson that the air component must always be prepared for sudden CAS operations even when initial battle plans are dominated by the ground component. For both components, a key lesson of Anaconda is that because original battle plans often do not survive initial contact with the enemy, the capacity to adapt quickly through new forms of joint operations is vital. When Operation Iraqi Freedom was launched a year later, the interface between ground and air forces was considerably smoother—an indication that the lessons of Anaconda had been learned.

21