The next president will need “more than tough talk or calls to carpet-bomb civilians,” declared President Obama in his final State of the Union address this month. He delivered a few well-placed barbs aimed at Republican presidential candidates and slammed Ted Cruz for his hard-line position on ISIS. After deflating the Texas senator’s call to bomb ISIS “back to the Stone Age,” President Obama said Cruz’s brand of bombast “might work as a TV sound bite, but it doesn’t pass muster on the world stage.” It was an interesting critique coming from a president whose Air Force flew 21,000 sorties over Iraq and Syria last year—9,000 of which involved “weapons release.” Altogether, more than 20,000 U.S. bombs were dropped in Iraq and Syria last year. If that doesn’t count as a massive bombing campaign, then what does?

Obama’s strategy of bombing ISIS isn’t working. It isn’t that the bombs aren’t having any effect—they are, and they’re probably more catastrophic than the Pentagon will admit—but the act of bombing itself isn’t accomplishing what it’s suppose to: that is, reducing the number of ISIS terrorists while preventing as many civilian casualties as possible. Operation Inherent Resolve’s spokesman, Colonel Steve Warren, recently said: “If you’re part of ISIL, we’ll kill you. That’s our rule.” But the size of ISIS has remained relatively static, around 30,000 since 2014. And as far as avoiding civilian casualties, the Pentagon late last year gave the suspiciously low estimate of six civilian casualties during the entire bombing campaign. An independent investigatory group puts the number at something around 100 times that. The bombings may be helping troops on the ground—the Kurds and the Iraqi army—but is the assistance the United States is providing cost-effective? So far, the air war against ISIS has cost Americans around $5.5 billion, or about $11.2 million per day.

Many think bombing is a cheap and effective alternative to spilling American blood, but there’s a counter-history of bombing that suggests it isn’t. In fact, instead of bombing campaigns being a safe and cheap default strategy, they are only practically useful and morally feasible within very narrow parameters. America has a misplaced faith in bombing campaigns.

The first bomb was dropped out of an airplane on November 1, 1911, just outside of Tripoli by the Italian pilot Giulio Cavotti, according to Sven Lindqvist in his odd and magisterial book A History of Bombing. Cavotti dropped four bombs, Danish hand grenades really, and thereby ushered in the air-to-ground attack. The Italian-Turkish War, a bid for the recently united Italian state to pick apart the corpse of the Ottoman Empire, also saw the first airplane shot down by rifle fire. Attack from the air and air defense were born and raised together, locked in an evolutionary competition.

The modern bombing campaigns of the mid-twentieth century—the “total war” that is experienced now through black-and-white photos of former cities reduced to either silent rubble or a washed-out incandescent glow—began properly during World War II. Lindqvist notes, “The principle for what was going to happen in Dresden and Tokyo at the end of the Second World War was already formulated at the beginning of the First.” The British mathematician F.W. Lanchester wrote in his 1915 book, Aircraft in Warfare, that total war, including the firebombing and destruction of entire cities, was an eventuality that the West’s military establishment needed to prepare for, even if queasy “sentimentalists” thought that the murder of 5 million peaceful civilians was just “the figment of a diseased imagination.” The crux of Lanchester’s reasoning was that the West should destroy its enemies’ cities before its enemies destroyed the West’s.

Already existing fully formed in the Western martial imagination, then, World War II gave Lanchester’s musings about total war from the air a stage on which to play out. In bombing Germany, U.S. and British air power had different animating principles. Both wanted to demoralize the population, foment uprising, and bring about Germany’s political, social, and technological collapse—but while the British focused on bombing civilian populations, the Americans focused on military and industrial targets. The bombing wasn’t accurate, especially at the beginning of the war. Even by 1944, only roughly 7 percent of bombs hit their targets. In fact, not until late in the war, when targeting became more sophisticated and more bombers were able to get through German air defenses, was Nazi industrial might even affected by bombing. For most of the war, carpet-bombing Germany tended to be expensive and inaccurate, exacting a high cost in Allied planes and qualified pilots. Even its role in the final Allied victory is dubious. Air-power expert Robert Pape argues that it was the territorial losses more than the relentless bombing that led to the Nazis’ industrial deterioration. Further, Pape says the war would have ended at relatively the same time, in relatively the same manner, even if there had been no bombing at all.

But the efficacy of strategic bombing is only half the question. There’s also the moral impact. In Germany, windows were blown out by incendiary charges that ignited fires that burned so hot the streets turned liquid. Some bombs were on time delays, blowing up hours after an initial round of explosions, when emergency crews had started to respond. A.C. Grayling, author of Among Dead Cities, wrote in The Guardian that, although the United States should be credited for diverging from the Royal Air Force by bombing German military and industrial sites instead of civilians, the Americans took to bombing Japan with far greater moral ambivalence. In less than a year, Grayling writes, the Americans killed as many Japanese civilians with aerial bombardment in a single year than German civilians were killed during the entire war. It took two atomic bombs before the Japanese finally surrendered. “[B]ombing civilians is not only immoral, but ineffective,” writes Grayling. “It takes nuclear weapons, delivering absolutely massive civilian extermination, to have the desired effect of reducing a people to submission.”

Of course, that’s a lesson that remained unlearned by the time America went to Vietnam. The United States dropped a combined total of 7 million tons of bombs on Vietnam, Laos, and Cambodia. That’s more than twice the number of bombs dropped during World War II. Half the bombing in Vietnam was performed by small planes, like the F-4 Phantom, which dropped napalm on South East Asia. More than 400,000 tons of it. “Overkill” was the mantra for the United States in Vietnam: Bomb a path to victory. It didn’t work. America bombed Vietnam for six years longer than it bombed Japan—and it still lost.

U.S. bombs have become more accurate since Vietnam. They’re “smart” now. But their accuracy, as far as avoiding civilians goes, is predicated almost entirely upon intelligence—knowing whether or not civilians are in a specific building or even in the vicinity of an attack. Even America’s pre-invasion “shock and awe” bombardment in Iraq—meant primarily to destroy Saddam Hussein’s political grip—was far more destructive than anticipated, possibly killing thousands of Iraqi civilians. The campaign was supposed to harness cutting-edge technology to cleanly decapitate control from Hussein and his cohort; instead, it precipitated the destruction of civil society. As General David McKiernan put it: “No Iraqi army, no Iraqi police … No local or national government organizations. Ministries didn’t exist.” It’s another example of the illusion that bombing can achieve all things in all circumstances—whether it’s the complete and utter devastation of the German industrial state, the squashing of a popular uprising in Vietnam, or what amounted to the multibillion-dollar assassination of a handful of men in Iraq. And now precision bombing is supposed to defeat ISIS cheaply and efficiently. Bombing has become America’s go-to tactic for dealing with international crises.

After nearly a century of aerial bombing, Grayling writes, the lesson learned by today’s militaries is simply to “not do body counts.” That’s certainly true of recent U.S. bombing campaigns: Estimated civilian casualties are just long-shot guesses. Bombing is alluring because it’s deceptively cheap. But the costs are all too real—in civilian deaths, operational costs, and energy sunk into a strategy that hasn’t accomplished much. Why does the United States continue to bomb? Pape lays the blame on “bureaucratic interests and political pressures for cheap solutions to difficult foreign policy problems.” Coercion by force is difficult, especially against a quasi-guerilla army. There is no easy, cheap, or quick solution, and there is no avoiding the financial and moral costs of combat.