The Sinews of Multi-Domain Battle
What is the most important capability the joint force can develop for multi-domain operations? It is not land based artillery engaging ships, not applying surface-to-air missiles to create local air dominance, or even using submarines for anti-air warfare. These buzzworthy techniques are fascinating; but the reality of “extending combined arms across all domains” is more mundane and quite difficult.
If U.S. forces are to master multi-domain battle we must first focus on tactical data flow through networks that form the connective tissue between sensors, decision makers, and shooters in disparate domains. Advanced physical employment configurations are simply new machinations of existing tactics; human-machine teaming is the quantum leap. Current tactical links are unable to seamlessly support high-speed actions between platforms, much less services. Even if the data could flow, the joint force also lacks coordination mechanisms in the other services to control these effects. Looking forward, existing data streams must be adapted, now, in order to be prepared for the employment of artificial intelligence on the horizon.
First, it is necessary take a step back to consider what’s radically new and therefore where we need to adapt. The idea of controlling land from the sea or vice versa is as old as the Peloponnesian War. Our weapons aren’t so different from any other age as to change the nature of battle. Lasers and rail guns are still direct fire weapon systems, just with much greater range than their black powder or bowstring predecessors. Perhaps the “precision revolution” is where we should look? Microchip aided weapons like anti-ship missiles haven’t changed a whole lot since the Falklands in 1982. Putting one on the back of a truck instead of on an airplane is novel, but not all that additive. Regardless of how precise, the locality of a weapon still restricts whether it can hit a target and how fast it can reposition. These factors, as well as cross-domain, more powerful, or smarter weapons, are all limited by classical mechanics. The emergence of “joint collaborative human machine battle networks” is the light speed, relativistic event that will mark the Third Offset as something new in warfare.
The power of microchips that led to a ‘revolution in military affairs’ and the Air-Land Battle doctrine that followed, is magnified by the connections between those silicon wafers. It is the effectiveness of these networks that should be our focus in the efforts to design multi-domain battle as the defining concept of the coming generation. The networks we were handed in the last generation of military technology, though ground breaking, are not necessarily joint and are certainly not collaborative. They are subject to a mess of proprietary limitations, antiquated interface standards, and misaligned capability developments that degrade the common operating picture and prevent system integration. To break these problems down we’ll use the U.S. Pacific Command (USPACOM) commander’s example of fighter jet cueing Army artillery to kill enemy ships.
The planned Army Tactical Missile System (ATACMS) upgrade can hit ships out to 300km, but any organic radar would have a shorter range for detecting surface targets. This means you will need to acquire the target with an off-board sensor, most likely the radar on an airplane, and perform a type of integrated fire control. Unfortunately, the Link-16 terminals those airplanes use are not necessarily optimized for the Army’s artillery radios. Newer fighters offer little hope of a solution in that the cutting edge F-22 is limited to receive only Link-16 traffic and the Joint Strike Fighter has similar drawbacks. Until recently, the Navy’s F-35 couldn’t exchange targeting information with its own ships. Internet protocol solutions have the potential to bridge these gaps but the message formats would still likely not agree with the data configurations that ground systems are designed around. If the most expensive weapons project in history can’t “talk” to surrounding systems, we’ll likely find that this theme defines other end products of our defense industrial complex.
While the DoD strives to create an open architecture network solution to system data exchange, the rest of the community can look to resolving the human side of multi-domain problems. In our ‘coastal artillery’ example, sensing a target that is in a different domain from the shooter also requires a third party to orient the sensor and shooter. The Joint Terminal Attack Controller (JTAC), as the tactical extension of the Joint Theater Air Control System (JTACS), has done a marvelous job vectoring close air support to ground combat forces for the last several decades. Who will do this for close ground support to maritime forces, what radio will he use, what calls and commands? How will this tactical need be answered from the operational level of war?
Very soon, if it hasn’t happened already, a company commander will need to “see” and “engage” an IP address in his immediate vicinity in order to effect a combat maneuver. Whether the digital target is attacking friendly communications, passing observation of U.S. forces, or “botting” out false news, the conventional Army has no U.S. Cyber Command (USCYBERCOM) Battlefield Coordination Detachment (BCD) applying the commander’s needs to a similarly non-existent USCYBERCOM “Air Tasking Order.” Whatever the digital and organizational solutions are, these difficult interoperability issues must be addressed quickly because there is an exponentially larger change on the horizon.
As AI transitions from research concept to engineering solution, it will automate the routine mental work of Intelligence Preparation of the Battlefield and Military Decision Making Process that are so critical to mission command. It will be taught rather than programmed. It may even learn to defeat human fighter pilots in air combat. The catch is that it will require massive amounts of data, perhaps hundreds of thousands of times more information than a human or group of humans can receive. We have to think about how to get the data to these systems now, otherwise AI will be great at tracking social media but terrible at generating weapons effects. Existing tactical network limitations hinder any cross-domain work we can do now and will cripple any future AI solution. Deputy Secretary of Defense Work has suggested that the future is Ironman. We’ll need something closer to Tony Stark’s J.A.R.V.I.S., and without intervention, he‘ll be more hindrance than help.
The U.S. Army Training and Doctrine Command (TRADOC) commander has called for “a lot of interoperability, a lot of interconnectivity” for multi-domain operations. Unfortunately, the community used the breadth of the concept to promote their favored projects from future vertical lift to healthcare rather than the weak links in the background. Even the Army Capabilities Integration Center (ARCIC) pamphlet on multi-domain mentions the physical and cognitive dimensions of the battlefield but misplaces the third dimension, information. With respect to autonomy, the most recent acquisition report on the topic acknowledges that more data is better but does not recommend efforts to get information out of existing links.
Getting information through dissimilar or solitary systems to facilitate integrated fire control is the really hard work of our new multi-domain concepts. Joint channels that have yet to be created must guide that information flow. Newton’s laws limit all the physical systems on which the discussion now dwells, the data that can empower them is restricted only by relativity. From the first century B.C. until the recent past, money formed the sinews of war. In a battle that occurs at the speed of light, data carries the day.