Drones—Unmanned Combat Aerial Vehicles (UCAVs)—are the wonder weapons of today's wars. UCAVs have been credited with striking the convoy carrying Muammer Gaddhafi; killing al-Qaeda's Abu Yahya al-Libi and Anwar al-Awlaki; eviscerating the Taliban's ranks and other militants in the Afghanistan-Pakistan theater; and hitting targets from Asia to Africa—all without putting their pilots in harm's way.

Given this record, it's no surprise that drones are beginning to dislodge manned aircraft from the central military role they have played since World War II. The drone revolution promises many benefits, but there are also drawbacks to this nascent unmanned air force—drawbacks that the nation and its policymakers have barely begun to assess.

The appeal of UCAVs is simple: they minimize risk. By keeping pilots and ground crews far from harm's way, drones make military operations dramatically less dangerous. The missiles that hit Gaddhafi's escaping convoy, after all, were fired not by an artillery grunt marching through the desert or an F-15 pilot lingering overhead, but by a remote-control warrior sitting safely in a nondescript building outside Las Vegas.

At a time when national defense is coming under increasingly severe budgetary pressure, the fact that drones are much cheaper than other weapons systems is also important. A Predator drone costs $4.5 million. The F-35 and F-22 fighter planes cost $159 million and $377 million each, respectively, while a B-2 stealth bomber has a per-unit cost approaching $2 billion. Moreover, training costs for a UCAV controller are less than a tenth of those for a traditional combat aviator. In 2011 the Air Force trained more drone pilots than fighter and bomber pilots combined.

In 2000 Congress expressed the desire that "within ten years, one-third of U.S. military operational deep strike aircraft will be unmanned." The Pentagon didn't reach that ambitious goal but is still making rapid strides:

• Combat air patrols by drones increased 1,200% since 2005.
   
• In the past decade, the U.S. drone fleet swelled from 50 aircraft to 7,500, though the vast majority of these drones are not UCAVs. The fleet of combat-class drones—drones capable of striking targets—is expected to grow to 650 by 2021. The Pentagon's plan is to double the drone fleet by 2020, even as the size of the manned fighter and bomber force shrinks. 
   
• Annual drone strikes in Pakistan increased from one in 2004 to 117 in 2010, when they peaked. The Brookings Institution estimates that as many as 2,769 militants have been killed by UCAV strikes in Pakistan. 
   
• The Air Force envisions deploying swarms of drones networked together to "operate in a variety of lethal and non-lethal missions at the command of a single pilot"—as many as five drones per pilot.
   
• Greater human ingenuity in equipping machines to gather, interpret, and respond to data requires less real-time decision-making by humans. The Air Force wants its next-generation, Long Range Strike bomber to be "optionally manned," and UCAVs equipped with "autonomous attack systems" are on the horizon.
   
• Drones are not, an Air Force report reminds, "limited by human performance or physiological characteristics." They can handle gravitational forces and speeds humans cannot, and never get wounded, sick, exhausted, bored, thirsty, or hungry. Drones deprive the enemy of human targets and, with the advent of nuclear-powered drones, offer the possibility of nearly endless operation.

In short, UCAVs are a startling advance in military technology. Drone warfare, however, has sobering implications—moral, legal, and strategic.

More Casual War

Concerning the first, UCAVs make war far less risky for the warrior who guides it and the nation he fights for. After all, the loss of a drone is the loss of nothing more than metal.

This is good for pilots, but may be bad for our republic. "More willing to lose is more willing to use," as Daniel Haulman of the Air Force Historical Research Agency puts it. Reduced risk affects policymakers' decisions. Military operations involving manned warplanes force policymakers to consider broader consequences and serve as a check on their war-making power. The prospect of losing an American life—and justifying that to the nation—does and should give the commander-in-chief pause. If there are no Americans at risk, however, it's less likely that the parents, spouses, children, or congressional representatives of those pulling the trigger are going to raise a fuss.

President Eisenhower weathered international humiliation after the Soviets brought down Gary Powers's U-2. President Kennedy was pressed to go to war when Rudolf Anderson's U-2 was shot down over Cuba. President Clinton had to deal with a hostage crisis abroad and withering attacks in Congress when Michael Durant's Blackhawk was shot down in Mogadishu. The loss of drones, by contrast, never posed a crisis or even a problem for President Obama.

Without those checks—without those dangers—presidents will be tempted to order military action more frequently and more casually.

The promise of risk-free war, the temptation to gain all the benefits of military success with none of the costs or consequences, not only makes it easier to go to war, but to continue one. Paul Miller, a former National Security Council official, observes that "endless war is unacceptable and dangerous."

Specifically, Miller asks, "When, and under what conditions, will the U.S. government stop using drones to bomb suspected terrorists around the world?" The drone war is an outgrowth of the post-9/11 campaign against terrorist organizations and regimes—a campaign authorized by the Use of Force Resolution of September 18, 2001. That measure directed the president "to use all necessary and appropriate force against those nations, organizations or persons he determines planned, authorized, committed or aided the terrorist attacks that occurred on September 11, 2001, or harbored such organizations or persons, in order to prevent any future acts of international terrorism against the United States by such nations, organizations or persons."

Does the sweeping mission to prevent any future terrorist attack against America mean the president is authorized to use all necessary force against states and sub-state actors that had some connection to 9/11, or does "such" mean similar ones that might commit the next terrorist attack? It would be a stretch to say that the September 18 measure authorized today's autopilot war against targets in Pakistan, Yemen, Somalia, and beyond. Those targets may indeed be enemies of and threats to the United States, but few targets of today's drone war "planned, authorized, committed or aided" 9/11. Underscoring this point, the Washington Post reports that a growing number of drone strikes in Yemen have targeted "lower-level figures who are…seen mainly as leaders of factions focused on gaining territory in Yemen's internal struggle." September 11 provided the motive, and drones provide the means, to wage war—or something war-like—against any geopolitical entity we believe poses a terrorist threat. But the feasibility of risk-free perma-war does not establish its wisdom.

International Watchdogs

"Just wars are limited wars," philosopher Michael Walzer argues. "[T]heir conduct is governed by a set of rules designed to bar, as far as possible, the use of violence and coercion against noncombatant populations." The world is taking notice that drone war is hard to square with this moral standard, which brings us to the legal issues raised by UCAVs. "Drone attacks do raise serious questions about compliance with international law," according to Navi Pillay, United Nations High Commissioner for Human Rights, who worries about "indiscriminate killings and injuries of civilians." The Brookings Institution estimates, for example, that some 400 non-militants have been killed along with the 2,000-plus militants killed by drones in Pakistan. The use of drones to cripple al-Awlaki's branch of al-Qaeda killed dozens of other people, many of them apparently not affiliated with al-Qaeda, including a 16-year-old relative of al-Awlaki born in Denver.

This should raise concerns. According to the New York Times portrait of the inner workings of the drone war, the Obama Administration has embraced a controversial method for determining civilian casualties that "counts all military-age males in a strike zone as combatants…unless there is explicit intelligence posthumously [!] proving them innocent." Equally worrisome, the president is described as "at the helm of a top-secret ‘nominations' process to designate terrorists for kill or capture." He studies "mug shots and brief biographies" of possible targets, approves "every new name on an expanding ‘kill list,'" "signs off on every strike in Yemen and Somalia and also on the more complex and risky strikes in Pakistan," and often decides "personally whether to go ahead" with a drone strike.

To many Americans, that sounds like a commander-in-chief fulfilling his primary responsibility (albeit through greater personal involvement than we might expect). But international observers might see something far more sinister in this portrait. Consider the Obama drone protocols in the context of the Rome Statute, which spawned the International Criminal Court. The statute considers launching an attack "in the knowledge that such attack will cause incidental loss of life or injury to civilians" as a war crime and defines a "widespread or systematic attack directed against any civilian population" as a crime.

The United Nations Human Rights Council (UNHRC) also has raised questions about the drone war's legality based on sovereignty issues, noting that targeted drone strikes violate a state's sovereignty if the state doesn't consent to the attack or if the strike is not a case of self-defense under the U.N. charter. We know that many UCAV strikes have not had the consent of Pakistan or Yemen. And although the U.S. government contends that a state of war has existed since September 11, 2001—a reasonable and defensible contention—the UNHRC argues that it is "problematic" for the U.S. to show "it is in a transnational non-international armed conflict" beyond Afghanistan.

This is not an argument in defense of international watchdogs tying America down. The bureaucrats who roam the U.N. may refuse to recognize America's special role, but by turning to Washington whenever civil war breaks out, nuclear weapons sprout up, or genocide is let loose, they are tacitly conceding that the United States is, well, special. Washington has every right to kill those who are trying to kill Americans. But the brewing international backlash against the drone war reminds us that means and methods matter as much as ends.

In other words, drones may not deliver on their promise of risk-free war because their use poses political and strategic risks. "In 17 of 20 countries," a recent Pew survey found, "more than half disapprove of U.S. drone attacks targeting extremist leaders and groups in nations such as Pakistan, Yemen and Somalia." The ongoing drone war feeds "a widespread perception that the U.S. acts unilaterally and does not consider the interests of other countries."

Distant and Remote

Fair or not, the world's verdict on the drone war recalls something British physicist Patrick Blackett wrote amid the Allied bombing raids on Germany at the end of World War II. Blackett worried that the Allies had developed a "Jupiter Complex," which historian Paul Johnson describes as "the notion of the Allies as righteous gods, raining retributive thunderbolts on their wicked enemies." The Allies concluded, Johnson explains, that strategic bombing "was the best way to make the maximum use of their vast economic resources, while suffering the minimum manpower losses."

UCAVs take the logic of the Jupiter Complex to its ultimate conclusion: maximum use of economic and technological resources with zero manpower losses and zero risks—all buffered by the virtual-reality nature of the delivery system. Being seen in such a light—as distant and remote in every sense of the word, especially in waging war—has serious consequences. Power-projecting nations are following America's lead and developing their own drones to target their distant enemies by remote control. Some 50 countries have drone programs underway. It is doubtful that Americans will sleep better if the way our government has used and justified drones to date becomes the Golden Rule by which all other countries deploy them, much less the opening bid in an international race to the bottom that culminates in a planetary free-fire zone.

For one thing, many of these nations have been far less discriminating in employing military force than the United States—and far less skillful. Indeed, drones may usher in a new age of accidental wars. Consider that America's Predator, Reaper, and Global Hawk drones crash more than any other aircraft in America's fleet—nine drones are lost for every 100,000 hours flown. That's almost triple the ratio for the rest of the Air Force's fleet. If the best drones deployed by the best military fail this often, the accident rate for mediocre drones deployed by mediocre militaries will be much higher. Such failures could trigger international incidents between India and Pakistan, North and South Korea, or China and any number of its wary neighbors.

Equally worrisome, drones are a cheap alternative to long-range, long endurance warplanes. Yet despite their low cost, drones can pack a punch. Owing to their size and range, they can conceal their home address far more effectively than the typical, non-stealthy manned warplane. Recall that the possibility of drone attacks was cited to justify the war against Saddam Hussein's Iraq.

Proliferation

Of course, cutting-edge UCAVs like the Predator and Reaper have not fallen into un-deterrable hands—at least not yet. But if history is any guide, they will. Such is the nature of proliferation:

• Venezuela has hired Iranian missile engineers to build a drone fleet. Venezuela's drones are nowhere near the Predator class of drones. But given the rhetoric and actions of Caracas and Tehran, it's fair to conclude that a) their goal is to have the ability to threaten America with UCAVs and b) they are closer to that goal today than before they launched their collaboration.
   
• After watching the U.S. drone war from a distance—and Georgia's use of drones up close—Russia turned to Israel for a batch of drones in 2009. Russia is earmarking $13 billion for drone development between now and 2020, including "automated strike aircraft." 
   
• China has at least a dozen drones on the drawing board or in production. The Pentagon's recent reports on Chinese military power detail "acquisition and development of longer-range [drones]…for long-range reconnaissance and strike"; acquisition of the Israeli HARPY drone system; development of UCAVs to enable "a greater capacity for military preemption"; and interest in "converting retired fighter aircraft into unmanned combat aerial vehicles." At a 2011 air show, Beijing showcased one of its newest drones by playing a video demonstrating a pilotless plane tracking a U.S. aircraft carrier near Taiwan and relaying targeting information. Underscoring the validity of worries about the proliferation of drones, the Washington Post quotes a Chinese defense-industry official saying, "The United States doesn't export many attack drones, so we're taking advantage of that hole in the market."

Even if the spread of UCAV technology doesn't harm the United States in a direct way, it seems unlikely that swarms of semiautonomous, pilotless warplanes roaming about the earth, striking at will, crashing here and there, and sometimes simply failing to respond to their remote-control pilots will do much to promote a liberal global order. It would be ironic if the promise of risk-free war offered by drones spawned a new era of danger for the United States.