Swarming (military)

At the siege of Samarkand, Spitamenes used Bactrian horse archers in effective swarming attacks against a relief column sent by Alexander the Great. Bactrian horse archers surrounded various Macedonian phalanxes, staying out of range of their melee weapons, and shot arrows until they had no more. The archers would then withdraw to a supply point, but another swarm of horse archers would sometimes replace them, and sometimes attack elsewhere. The Bactrians eventually caused the phalanx to break formation, and destroyed it. Alexander recognized his forces could not directly combat horse archers, but that the horse archers needed resupply of provisions, horses, and arrows. Alexander split his forces into five columns and began building fortifications in the areas where the Bactrians had resupplied. Eventually, his anti-swarm tactics worked: cut off from resupply, the Bactrians had to meet the Macedonian phalanx, which were vastly superior in melee. Alexander made it priority to engage guerillas or other light mobile forces. Spitamenes was effective as long as his force were mobile, and he had adequate communications with mounted couriers. Once he was forced into direct battle with heavy forces, he lost the advantages of his forces and was defeated quickly. At the Battle of the Jaxartes River, Alexander once again faced swarming tactics from an army of Scythian horse archers. Alexander sent a unit of heavy cavalry ahead of his main line. As expected, the Scythian horsemen surrounded the detached cavalry. At the right moment, Alexander's cavalry reversed direction and pushed half of the Scythians straight into the main phalanx of Alexander's army, where they were slaughtered. Upon seeing this, the remaining half of the Scythian army retreated from the battle.

Mongols under Genghis Khan practiced an equivalent of swarming, partially because their communications, which used flags, horns, and couriers, were advanced for the time. Also one of the standard tactics of Mongol military was the commonly practiced feigned retreat to break enemy formations and to lure small enemy groups away from larger groups and defended positions for ambush and counterattack. Genghis Khan used the Yam system, which established a rear line of points for supplies and for remounts of fast-moving couriers. The remount system allowed horsemen to move much faster than the couriers of opponents without them. These couriers kept the Mongol senior and subordinate commanders informed, such that they could make fast decisions based on current information. In modern terms, the courier system provided the means of getting inside the opponent's OODA loop. With fast communications, the Mongols could make decisions not just on what they could see locally, but with that information oriented within the overall situation. They could then decide and act while the enemy were still waiting for information. Outnumbered Mongols could beat larger forces by faster communications, which allowed units to withdraw and regroup while other groups continually stung the enemy, withdrew in turn, while the earlier group again hit the enemy.

In the 1980s, the Soviets developed an 'Operational Maneuver Group' (OMG) for a fast armored thrust deep into NATO defences east of the Rhine River. An OMG was expected to exploit strategic surprise with a force equal or greater than an armored division, featuring up to 700 tanks, 500 IFVs, and a substantial number of helicopters. As a countermeasure, NATO considered neutron bombs but their use was politically controversial. NATO instead devised a plan to slow the thrust with a swarming counterattack, called Dynamic Density, which used single-seat Small Military Aircraft (SMAs) operating autonomously in pairs with infantry ATGWs such as Milan, their pilots being infantry.[9] One aircraft would carry the Milan post and four missiles, the other the night-vision sight and four missiles (two of which might be anti-helicopter), and the tactics would be shoot and scoot. The SMA, known as the Dragoon, was evaluated and highly praised by the MoD's test pilots at Boscombe Down, its STOL performance and ease of handling making it ideal for this role.[10] Large numbers would be needed and 5,000 was suggested as sufficient to ensure that swarming would be successful against a force as large as an OMG. Once it was recognized that success could be claimed with a significant deceleration, other novelties were introduced, among which was Synthetic Density which required the SMAs to distribute pneumatic models (fitted with radar reflectors) of tanks and artillery along the OMG's MLA, these requiring the room to be put down and time to be lost before progress could be resumed.

The proposal was published in the Journal of the Royal United Service Institution and a couple of years later a shorter article suggested that if the Soviets themselves had used SMAs in swarms in Afghanistan their COIN operations against the Mujaheddin would have been far more successful.[11] Contemporary Western armies in Afghanistan can readily accept that swarming at the tactical and operational levels is appropriate, but the physical structure of the country rules out the currently available fighting vehicles. However, networked and swarming SMAs, again 5,000 in number, all armed with laser designators for the second echelon of conventional ground attack aircraft, would constitute force multipliers with a substantial impact.

Swarming ties in well with the theories of the military strategist John Boyd, the "high-low mix" in which a large number of less expensive aircraft, coupled with a small number of extremely capable "silver bullet" aircraft, had the effect of a much larger force. Boyd's concept of quick action is based on the repeated application of the Boyd loop, consisting of the steps

• Observe: make use of the best sensors and other intelligence available
• Orient: put the new observations into a context with the old
• Decide: select the next action based on the combined observation and local knowledge
• Act: carry out the selected action, ideally while the opponent is still observing your last action.

Boyd's concept is also known as the OODA Loop, and is applicable to all military operations, as well as to civilian competition from sports to business.

These are a realization of Boyd's theories. A swarming case is any historical example in which the scheme of manoeuvre involves the convergent attack of five (or more) semiautonomous (or autonomous) units on a targeted force in some particular place. "Convergent" implies an attack from most of the points on the compass."[1]

Prevention of fratricide, as well as the ability to make ad hoc swarming attacks on targets of opportunity, is one of the major goals of combat data networking among units down to the level of individual tanks and soldiers. Blue Force Tracker is an early vehicle-level synchronization system,[12] also operating in helicopters.[13] These systems are still new and undergoing considerable improvement. One fratricide incident in Afghanistan came from the users not understanding that their target designation device reinitialized, after battery replacement, to the position of their designator, not of the target. If the bomber had a beacon that gave the crew the precise location of the friendly troops, that would have been another way of avoiding attacking one's own troops.

Chinese soldiers swarm a UN-held position during the decisive Battle of the Cheongcheon River in the Korean War.

Swarming is advantageous to less powerful countries and groups, because it allows them to balance their disadvantage in firepower and numbers. Despite being less technically advanced, Communist forces made good use of swarming in Asia during the Cold War. The Chinese were able to make up for their lack of firepower by attacking from all sides and then quickly advancing to the rear. The Vietcong were famous from attacking from all directions out of nowhere and then quickly disappearing. When they did come into close contact, they used a technique called "hugging the belt", which meant they were too close for the US to employ air and artillery support.[17] If the attackers "hugged" at several points, "pulsing" their attacks, they both neutralized external fire support, but also made it difficult for the US commander to know where to commit reserves.

Cordesman observes that swarming is a viable terrorist tactic against targets of opportunity.[18] Al-Qaeda, for example, uses a different form of swarming than those of advanced militaries, in which the general objectives of operational cells are agreed in a manner coordinated, but not continuously controlled by the core organization. Once the decision has been made on the general targets, the operational cells cut positive control links from the core, although they may still receive financial and other support. A signature of al-Qaeda operations has been multiple, near-simultaneous attacks, such as the several hijacked airliners in the 9/11 attacks, the closely spaced bombings aimed at US embassies in Tanzania and Kenya, and attacks on buses and trains in London. The attacks on trains in Spain had an additional dimension: not all the swarms were associated with al-Qaeda.

While John Arquilla, a professor at the Naval Postgraduate School, cites the ability to plan separate and widely dispersed attacks, coordinated by mobile communications that might originate from a cave on the Afghan-Pakistan border,[8] he does not emphasize the apparent al-Qaeda technique of releasing operational units to local control, once the policy is set.[19] See Clandestine HUMINT operational techniques.

The apparent al-Qaeda methodology of letting operational cells decide on their final dates and means of attack exhibit an operational pattern, but not a periodicity that could easily be used for an indications checklist appropriate for a warning center. Such lists depend on seeing a local pattern to give a specific warning.[20]

Semiautonomous swarming, in which the actors occasionally interfere with one another, is seen in attacks on computer networks by loose confederations of malicious hackers. On occasion, especially when the attack uses a botnet,[21] some of the units may try to overpower and control one another, as well as the target.[22] One of the observations of the Center on Terrorism and Irregular Warfare was that unfocused mass disruption was not a useful terrorist, and by extension general military, tactic.[22] The 9/11 attacks had symbolism. A cyberattack on a stock market would have symbolism. For the political purposes of the swarm, there has to be a symbol to which observers need to connect the purpose of the attack.