We have seen the future of effective military command and control and it is only made possible by speed. In the future, adversaries will increasingly rely on machines rather than people for basic functions like surveying the battlespace, distinguishing friend from foe, and formulating options for strikes. To keep pace, the US military is developing a new mechanism for command and control, calling it the ‘Joint All-Domain Command and Control system.’ This new system will have two organizing principles aimed specifically at increasing decision speed: pre-decisional functions (i.e., surveying, identifying, and formulating) will be automated through AI-enabled technology, and decision-making itself will occur at the lowest level possible (because of ethical and pragmatic limitations, today’s AI is nowhere near the point of being able to make decisions). Since people – not machines – will remain responsible for real-time choices about the use of force, the only way for a military to decide quickly in a complex battlespace is to diffuse decision authority throughout the organization, rather than concentrate it in the hands of a few.
Despite this vision of future decision-making occurring at low levels, preparation for the command and control of the future has concentrated on the operational or higher levels of warfare. The Air Force, for example, has created a new career field devoted specifically to the operational level – the ‘13 Oscar’ multidomain command-and-control officer. Separately, after unveiling its multidomain operations concept, the Army began exploring the formation of two new ‘field army’ headquarters occupying an echelon above Army divisions and corps. These efforts indicate that, in building the command and control of the future, the military is working from the top down, prioritizing the operational and strategic levels above the tactical level.
But this isn’t the only way. Senior leaders would be wise to consider a bottom-up approach, which would offer four distinct advantages: improved decision speed, superior inter-service integration, a greater likelihood of solving the most difficult command-and-control problems, and the best chances for the survivability and resiliency of the holistic command-and-control system.
Four Reasons to Build Bottom-Up
Let’s start with decision speed. A core principle of military theory holds that more rapid decisions – provided they are not so hasty as to be rash – have inherently more value than slower decisions. Deciding faster puts the adversary in a reactive position. With people responsible for choices about the use of force, decision speed requires diffused decision authority. The trouble with working top-down is that it risks transferring the more cumbersome decision-making practices of traditional command and control – higher levels hold decision authority while tactical levels act – onto the command and control of the future. Of course, some traditional command and control already embraces the idea of tactical level decision-making, called ‘mission command’ in the Army or ‘centralized control, decentralized execution’ in the Air Force. Despite the intentions of such directives, however, reality is often different, especially at the intersection of domains. Today, for example, what sounds like a simple matter of deciding what air assets will support which ground-combat unit is the result of a tedious back-and-forth between the overall commander of a given region and the commander of the air units in that region. In the future, the military would do well to have this apportionment decision an outcome of many decisions at the lowest level possible, made fast and frequently over the course of a conflict, rather than the result of one decision at the flag-officer level. The same will be true for decisions at the intersections of other domains, many of which the military is only beginning to imagine, and for which it has neither defined mechanisms nor practiced in exercises.
Second, the bottom-up approach increases the likelihood that the resulting command-and-control structure will be joint in practice and not just in name. Our previous research demonstrates that the further one gets from the battlefield, the less likely it is that genuine inter-service cooperation will occur. It is a simple matter of the structural impediments to joint acquisition that have accumulated under the Department of Defense’s zero-sum budget game and Washington politics. At the tactical level – with its imperatives of surviving and defeating the enemy overshadowing bureaucratic politics – cooperation happens quickly and fluidly. But because they originate under the control of D.C. politics, systems designed for use at the operational level or higher are more likely to be ‘service-parochial’ and incompatible with other services’ systems than those designed for tactical use, or cobbled together in the heat of battle. The implications for investing in command-and-control systems are clear: Jointness requires bottom-up experimentation and development.
Third, the technological problem of conducting all-domain control is most troublesome at the tactical level. The future environment will be one in which every sensor is connected to every shooter in every domain. In this environment all echelons of war will go from being complicated arenas to truly complex by a rigorous complexity-theory definition of the latter term. Navigating this complexity will be a big challenge regardless of the level of war, but will grow especially daunting in a tactical setting where any all-domain control system would have to be mobile at the level of small teams and would have to function in close proximity to adversaries. Moreover, the tactical level may be best suited for solving this problem: tactical-echelon units get the most practice in combat-realistic simulations of what the new style of warfare will look like, and are therefore much more likely to rapidly innovate systems that are interoperable and practical, provided the services allow this innovation to take place. Candidly, as members of an Air Force warfighting specialty best suited to serve in this capacity, we admit some cultural preparation will be required. Eagerness to prioritize the command-and-control portion of the tactical air-control party mission on equal footing with the maneuver-centric portion of the mission has not been a popular endeavor. But here again tactical development is a boon, because it could serve to overcome this institutional inertia.
Last, emphasizing the tactical level now is the best way to maximize the survivability of systems. While centralized control has proven effective in the past, in a major conflict with a peer adversary, centralized nodes will be vulnerable. Building architecture dispersed among tactical nodes is one means of buying insurance at the operational and strategic levels. Current thought is already pointing in this direction – the Air Force’s great-power conflict concept, for example, envisions tactical nodes as a means of making an entire network survivable against a peer adversary.
The (Tactical) Way Forward
Moving forward, the military – led by what the military calls in typical unwieldy jargon the ‘Joint Cross-Functional Team for Joint All-Domain Command and Control’ – can better prepare for the future with initiatives emphasizing tactical development. It should start by educating and equipping tactical elements across the joint force for all-domain control. Operators should receive familiarization training in domains with which they do not currently interact, and subject-matter experts from each domain should be assigned to tactical command-and-control units. The military should also create direct links between tactical operators and the industry partners who will develop the AI-enabled technology of the future. When it sanctions experiments with future systems, the military should require involvement and feedback from tactical elements. This type of bottom-up development is anathema to the structural and political limitations that shape most military acquisition, and will require committed senior leaders from at least two separate services who are willing to make concessions. This approach, however, is feasible – in the 93rd Air Ground Operations Wing, our current unit and the one responsible for the preponderance of conventional tactical air-control party airmen, we have participated in joint experiments with the express intent of gathering feedback from tactical operators. We have found that industry partners are eager for this kind of feedback, that tactical operators are happy to give it, and that the confluence promises to build more pragmatic and survivable networks in the end.
Overall, the arguments here speak to a more general debate ongoing within the military: As it reforms itself, whether in command-and-control structure or something else, is it better to drive change from the bottom up or the top down? Of course, all levels of war are important and proficiency at each level can determine effectiveness in the others. Tactics can drive strategy and, at the same time, no level of tactical proficiency can make up for poor strategic decision-making. Tactical innovation ought to link with operational aims, top-echelon commanders’ priorities, and knowledge of how the higher echelon needs to receive and process data. Otherwise, tactical innovation becomes a trivial novelty with no meaningful warfighting application.
Our argument evokes the building of the transcontinental railroad. The railroad could have been built from the East, where capital and labor were concentrated at the time, or from both the East and the West. Building from both sides required adjustment to meet in the middle, but that approach forced the builders to deal with Western-specific problems like mountain passes sooner, and ultimately resulted in a faster timeline. Likewise, our argument is not that the tactical level is more important than the operational and strategic levels, simply that the tactical level offers some of the best ways to solve the litany of problems facing the military as it transitions to the command and control of the future. At this stage of development, we believe that emphasizing the tactical level will build more ‘railroad’ faster. Working bottom-up is the best way to increase decision speed, maximize joint cooperation, develop all-domain command and control, and increase the survivability of future command and control, all of which are critical requirements for higher echelons.
With such a big set of problems ahead, it is reasonable for the military to start with what it knows. In a world of scarce resources, however, each sensible step the military takes is also a step it doesn’t take. Ingrained habit and organizational power suggest the first step will be from the top. Prudence suggests the weight of effort should be on the bottom.