This essay is adapted from Defense Budgeting for a Safer World: The Experts Speak, a new publication by the Hoover Institution Press.
It is impossible to separate technology from military power. For thousands of years, the states best able to invent and adopt technologies have found advantages on the battlefield: whether it be the longbow, the trace italienne, or the steam engine. However, picking the right technologies and deciding which ones to invest in and how much to allocate for these systems is a complicated set of guesses, a process fraught with uncertainty. This uncertainty is particularly acute when it comes to emerging technologies, in which decision makers must balance educated guesses about the impact of technology with their feasibility—ultimately trading off priorities between known technologies and the promise of future technologies. The states best able to predict the future of technology on the battlefield and make the right investments are ultimately the strategic winners.
Despite the complexity and uncertainty of these decisions, too often technology is treated as a simple variable: greater investment in technology leads to better outcomes on the battlefield. Therefore, the competition for military power becomes about capacity. In fact, many political scientists use economic capacity and GDP as proxy variables to predict the most capable militaries—the states that invest in the right technology with the greatest resources generate the greatest military power. And yet, it is often not the case that capacity determines who “wins” the race to master emerging technology or even that greater innovation occurs within the militaries with the largest budgets. Instead, how militaries decide to invest, adopt, and integrate military technologies is just as important as any state’s inherent capacity to produce military technologies.
Understanding how human interaction with emerging technology influences military effectiveness is pivotal for militaries as they set out to deter and defeat their adversaries. It is also a pressing issue for the US military, facing a daunting peer competitor in China and staring down a revisionist Russia in Ukraine. A growing list of technological buzzwords vie with one another and traditional platforms for priority within a budget increasingly crowded with competing requirements.
Let’s examine one large and heterogeneous group of technologies, unmanned systems, as a case study to glean lessons about how the Department of Defense (DoD) responds to emerging technologies. What lessons might we learn from the decisions made over the past fifty to one hundred years? Are there patterns, or best practices, that may help us better build budgets for today and in the future? We then can apply the lessons and best practices from this case study to current challenges with cybersecurity, space, missiles, and software.
Service identity matters—a lot
Capacity and technological development have been important to which technologies ultimately succeeded. But these were often secondary factors behind organizational identities, beliefs, policy entrepreneurs, and exogenous shocks to the system.
First, perhaps the greatest determinant of when and why some unmanned technologies have succeeded while others have failed is how the technology interacts with organizational incentives: namely, armed-service identities. This is not a new phenomenon. Famously, Carl Builder’s 1989 study on American military strategy, The Masks of War, identifies distinct service-based personalities within each military branch that define their attitudes toward technology.
The Navy’s focus on tradition and independent command at sea leads to decisions prioritizing the Navy as an institution. As the only service operating air, sea, and land forces, the Navy has the most subcultures and also views itself as the service least reliant on joint escapades.
The Air Force is insecure about its independence and therefore advocates a doctrine that emphasizes strategic airpower and prioritizes technology over the individual service member. As Builder writes, the Air Force “sees itself as the embodiment of an idea, a concept of warfare, a strategy made possible and sustained by modern technology.”
In contrast, the Army is focused on personnel and has “roots in the citizenry,” making it a late adopter of technology and an advocate for personnel-heavy doctrine over the technology-focused efforts of the Air Force.
Builder focuses on US service culture to explain DoD weapons and doctrine choices but also discusses subcultural identities that derive from operational specialties. For instance, in the Air Force, fighter pilots once vied for influence against bomber pilots; more recently, pilots of unmanned systems have subdivided into a culture distinct from that of manned fighter and bomber platforms. The Navy has an even more codified set of organizational identities, with three specialties organized as separate personnel and manning structures: surface-warfare officers, submariners, and aircrew. Previous work suggests that specialties that face replacement by emerging technologies may be less likely to adopt them and may actively fight back against the proliferation of these systems through budget choices, doctrine development, and personnel choices.
In exploring investments in unmanned systems, service and occupational identities have been the primary predictors for which technologies ultimately succeeded. It required significant external intervention—a war, Congress, or influential policy entrepreneurs—to overcome these identities. When the unmanned system threatened the service’s identity, it was more likely to fail. When the unmanned system didn’t have an advocate within the service, it was more likely to stagnate. Joint endeavors, often foisted by Congress in a top-down attempt to streamline or consolidate investment, were almost always unable to gain enough support from within services to survive over multiple budget years.
In contrast, technological innovation was spurred when Congress pitted services against one another to develop platforms. In particular, the Air Force’s consummate need to validate its existence meant that when pitted against another service for control over an unmanned system or mission, the Air Force was willing to adopt the technology even if it countered its core service identity. For example, Theodore von Kármán (a prominent rocket scientist during World War II and colleague of Hap Arnold, the first general of the Air Force) explained how the Air Force wrested control over ballistic missiles, explaining, “We used the term ‘pilotless aircraft’ to cover all types of missiles, so as to prevent the project from falling into the hands of the Army.” Decades later, the Air Force’s desire to control missions also led it to invest in and adopt remotely piloted aircraft at a far greater level than any of the other services—despite the power of the fighter-pilot identity.
While service identities led to different unmanned-technology trajectories within each service, in general, they created an incentive across services to focus on manned platforms—which services prized as core to their identity—over unmanned munitions and support equipment like bombs, missiles, communications, or intelligence assets, and finally over unmanned platforms.
Powerful narratives
In many cases, the only reason unmanned technologies survived organizational incentives and service-identity bias was that they became part of a larger narrative. Narratives about the future of technology and beliefs about how technology might affect the future battlefield have been key to convincing both service chiefs and Congress to preserve investments in technologies that might otherwise have been cut. In particular, two core belief narratives have dominated US defense discussions about technological investment in the past fifty years. These beliefs become more influential to policy when they are championed by an enterprising individual, particularly one with power within the services.
The first belief, technological determinism, is that technology exists within a linear understanding of history in which technology punctuates equilibrium to create revolutionary advances in military effectiveness. Unmanned systems are a part of that linear progression as a component of the most recent information-technology revolution. According to these beliefs, technology is the primary agent of change. It is, therefore, the responsibility of the United States to harness the power of unmanned technologies to leapfrog adversaries by creating campaigns of speed, situational awareness, and decisive advantage.
The second set of beliefs—derived primarily from the US experience in Vietnam and from midgrade officers who dominated US defense thinking after the Cold War—is about casualty aversion and force protection. It holds both that the US public is casualty intolerant and that public opinion is important for the military to achieve strategic success. Public opinion about the loss of troops constrains decision makers and influences the choice of military tools on the battlefield. By removing US personnel from the battlefield, unmanned technologies provide a technological solution to the constraints decision makers believe are imposed by the American public’s casualty intolerance.
From the outside in
Outside catalysts also play a key role in shaping defense technology. Wars drive innovation. They provide both an impetus and increased funding for technology, serving as immediate proving grounds for technology that might otherwise be favored (or rejected) by services. This has happened time and time again for unmanned systems. For example, torpedoes—which showed early promise—languished after World War I. Faced with declining defense budgets, the Navy moved the systems to the bottom of their priority list, behind the new aircraft, submarines, and aircraft carriers the service was clamoring for. As author Robert Gannon wrote, when World War II began, the Navy was left with “a tiny dribble of beautifully crafted torpedoes, barely less erratic than their World War I forefathers, produced by an organization corpulent, sluggish, and not so much consciously resistant to change as physically and emotionally unable to.” Only World War II forced the Navy to restart torpedo development and introduce far more capable ones.
Similarly, the Vietnam conflict drove tactical investments like unmanned aircraft and autonomous munitions, which were ignored under the peacetime dominance of the Strategic Air Command. More recently, 9/11 led to the armed remotely piloted aircraft, a phenomenon that has dominated unmanned system investments in the past two decades. As then–secretary of the air force James Roche recounted, there was a lot of resistance to arming the Predator before 9/11, but then “two buildings fell over,” and suddenly arming the Predator didn’t seem nearly as risky or revolutionary as it had before.
In some cases, it wasn’t just war that provided an exogenous shock to technology. While much of the Cold War was dominated by service-identity and organizational competition, the 1957 launch of the Soviets’ Sputnik satellite and the nuclear threat created impetuses for investments in emerging technologies that might otherwise have failed. The Sputnik moment also created a window of opportunity for the Air Force. After Sputnik, the Atlas missile program and its descendant, the Titan, sprinted forward technologically as pushes to respond to the Soviet lead in space ensured a solid budget allocation. Congress, and to some extent, the executive branch, play important external catalyst (and confining/shaping) roles for technological development. While Congress is derided for retaining weapons that support industries within representatives’ districts, it often plays an important role in saving technology that would otherwise not be funded by the services. Congress and various presidential administrations, for example, saved most of the ballistic and cruise missile technology that now exists in the US arsenal.
Congress and the executive branch also play an important role in influencing budget cycles, which serve as critical junctures for technological trajectories. Investments in research and development made in big-budget years create a path dependency for technologies during subsequent lean years that can lead to suboptimal technological trajectories. Emerging technologies are more likely to be cut by the services during lean budget years, making outside intervention more important to saving technologies that the services do not otherwise prefer. It also may leave the military with a glut of technology that isn’t optimized for the current context.
The defense industrial base rarely innovates alone. There are very few cases of successful technological innovation started by the defense industrial base without a requirement from the Department of Defense. Instead, the tale of unmanned technologies is replete with examples of civilian innovations that fail to find customers within the military until a large exogenous shock forces the military to revisit (and sometimes resurrect) these technologies.
What this means for new technologies
Investment in unmanned systems reveals a few lessons. First, the status quo for technologies is that they will succeed or fail based on how well they fit into a service’s identity. To negate these biases, it is important to have the services compete against one another when necessary. Also, leaning on Congress and the executive branch can exert important top-down pressure when processes stagnate. Narratives, especially those propagated by successful policy entrepreneurs, can help overcome service biases. But perhaps the biggest implication of research into unmanned systems is this: a general self-awareness within the DoD of when technologies might inadvertently be set up to fail because they don’t have champions in the budget process could lead to better acquisition and development processes for emerging technology.
Cyber technologies—especially those developed for defending the nation—will need such a champion to ensure priority within defense budgets. In fact, the charismatic Army general Paul Nakasone has waged a persistent and successful battle to influence legislation to protect the Cyber National Mission Force (CNMF) and cyber command. How has cyber command been able to do this? Cyber command has crafted a narrative about its role and its mission’s uniqueness compared to those of the services. Cyber command also began a public-facing effort to advertise previously covert efforts.
Related, and perhaps even more complicated, are investments in information technology and Joint All-Domain Command and Control (JADC2). The information-technology structure of the military is already divided into armed-service segments (the Air Force owns its network and data, etc.)—which makes basic information technology (and cybersecurity) upgrades difficult to carry out. JADC2—an effort to combine these networks to communicate and share data seamlessly across the services—will be extremely difficult to implement. However, Congress may give the services an incentive to work together or prioritize information-technology funding by threatening to allow one service to run the entire JADC2 program—creating an impetus for innovation where the default is for stagnation.
There also are implications for the future of US conventional missile capabilities. Why hasn’t the United States invested as much into hypersonic missiles as it probably should have? Why does the United States not have a larger, more sophisticated arsenal of conventional strike surface-to-surface (or even ship-to-ship or ship-to-shore) missiles? Part of the explanation for falling behind states such as North Korea and Iran in conventional strike options, and China in hypersonic missile options, is context, and the US-Soviet relationship in particular. Arms-control agreements between the United States and the USSR limited much of the conventional strike arsenal; even cruise missiles (which were not explicitly a part of strategic arms-control agreements) were used as part of a negotiating tactic for the United States trying to limit nuclear arsenals.
Organizational interests also handicapped the US conventional strike development. The Army had abdicated its stake in long-range missiles completely by the 1980s to the Air Force, which was happy to commandeer the mission but also was not interested in investing in missiles that didn’t fit into the nuclear mission of Strategic Air Command, nor the conventional campaigns fought by Tactical Air Command. Meanwhile, the Navy (like the Air Force) lost interest in conventional strike missions that might threaten the bread-and-butter aircraft of aircraft carriers, while submarines, generally focused on strategic strike missions, were underprioritized as a conventional missile strike option.
These organizational and contextual complications meant that the focus was on missiles as munitions that could be carried by existing platforms like aircraft and destroyers. Even when leaders recognized the need for emerging technologies like hypersonic missiles, there were few service imperatives to invest in systems that didn’t fall neatly into organizational niches and that threatened the role of favored platforms.
Technology does not exist without human intervention. We cannot simply invest more in technology and expect it to lead to victory on the battlefield. Instead, how technology shapes the winners and losers in war is a result of the process by which organizations, individuals, and beliefs create and use that technology in the first place.