Technological advances often give rise to new types of weapons, but the achievement of lasting breakthroughs in fighting power requires organizational and doctrinal innovation as well. Invention of the internal combustion engine more than a century ago, for example, led to the tank and airplane. Yet these weapons systems did not realize their potential until the 1930s, when the Germans concentrated their armor into panzer divisions and articulated a blitzkrieg doctrine that tightly coupled maneuver forces on the ground with attack aircraft above. Today, the U.S. military is fielding awesome new technologies, but it is still far from figuring out the right organizational structures and doctrines for best applying them.
Advanced information technologies have revolutionized U.S. forces' abilities to communicate swiftly, monitor enemy movements in real-time, operate vehicles remotely—on land, at sea, or in the air—and guide weapons in a way that effectively decouples range from accuracy. Yet, only modest attempts at organizational and doctrinal innovation have been tried.
The U.S. Air Force is experimenting organizationally by creating "composite" wings and tailored "air expeditionary forces" that mix different types of air platforms in the same tactical combat units. A concomitant new doctrinal emphasis on supporting advanced ground operations is bringing modern air power tantalizingly close, after so many decades, to realizing its fullest war-winning potential. The Marines have also engaged in field exercises in which the units of maneuver have been radically altered by creating autonomous units as small as eight-man squads. The Marines (not to mention special operations forces) understand that connectivity coupled with air mastery greatly empowers even the smallest combat formations.
For the most part, though, the bulk of the U.S. military is still wedded to heavy ground divisions and aircraft carrier battle groups. Almost all the technological changes of the past two decades have been folded into the Pentagon's existing understanding of war, summed up in the doctrine of "AirLand Battle." This concept of operations—originally intended for use against Russian forces if the Cold War ever got hot—is but a small upgrade to the aforementioned World War II-era blitzkrieg doctrine. Indeed, Norman Schwarzkopf's "left hook" in the Iraqi desert in 1991 was a virtual clone of Erwin Rommel's panzer sweeps across the North African desert in 1941.
Meanwhile, the world keeps moving into the age of networks. Networking means much the same for the military as it does in business and social-activist settings, not to mention among information-age terrorists and criminals: monitoring the environment more broadly with highly sophisticated sensors; expanding lateral information flows; forming and deploying small, agile, specialized teams; and devolving much (but not all) command authority downward. But it also has a doctrinal implication that these other types of actors are learning faster than the U.S. military: It's a good idea to become adept at "swarming."
Swarming is a seemingly amorphous but carefully structured, coordinated way to strike from all directions at a particular point or points, by means of a sustainable "pulsing" of force and/or fire, close-in as well as from stand-off positions. It will work best—perhaps it will only work—if it is designed mainly around the deployment of myriad small, dispersed, networked maneuver units. The aim is to coalesce rapidly and stealthily on a target, attack it, then dissever and redisperse, immediately ready to recombine for a new pulse. Unlike previous military practice, battle management is now mainly about "command and decontrol," as networked units all over the field of battle (or business, or activism, or terror and crime) coordinate and strike the adversary in fluid, flexible, nonlinear ways.
Early examples of swarming appeared with the great mounted armies of the 7th century Muslims and the 13th century Mongols, both of which mastered the technique of omnidirectional attack. In modern times, British fighter planes swarmed from dispersed airfields all over southeastern England to harry massed Luftwaffe formations during the Battle of Britain, while at sea German U-boats were widely distributed when scouting, then converged to attack allied convoys. What's different today is that advanced sensing, communication and weapons guidance technologies make swarming possible in any terrain, against any opponent, 24/7.
While the American military remains officially wedded to AirLand Battle, its latest field campaigns exhibit the beginnings of a potential "BattleSwarm" doctrine. In Afghanistan in the fall of 2001, slightly more than 300 special forces soldiers, who were networked with each other and with various air-based attack assets, quickly toppled the Taliban. These same elites did it again in much of Iraq, striking all over the country from the outset, saving the oilfields in the south, knocking out the Scud Box in the west, coordinating with the Kurds in the north and securing the approaches to Baghdad.
Will the U.S. military build on these first steps toward developing a truly networked "swarm force"? To best counter the adversaries bedeviling us in Iraq and Afghanistan, and those we may confront in other terror-war theaters, it is advisable to innovate along these lines. Right now, many military leaders are attracted to the concept of "network-centric operations," a vision of wiring together all our sensors and shooters. In some circles, however, swarming is being viewed narrowly, as a specialty notion, associated mainly with the use of autonomous (i.e., artificial intelligence-driven) systems. But as a deeper vision emerges and fixations on technology ease, serious questions will be raised about how best to give network-centric concepts operational life through organizational and doctrinal innovation. When these systemic questions get some traction, it will become evident that swarming is a big part of the answer.
John Arquilla and David Ronfeldt are Rand Corp. analysts who have been writing together for many years.
This commentary originally appeared in Aviation Week & Space Technology on September 29, 2003. Commentary gives RAND researchers a platform to convey insights based on their professional expertise and often on their peer-reviewed research and analysis.