The 2001 Quadrennial Defense Review season is hard upon us, but this year's energies are being expended in a new context. Previous exercises were mainly about preserving Service resource shares in a mostly stagnant intellectual and budgetary environment. This year there is, first and foremost, the major review undertaken by Defense Secretary Donald Rumsfeld, a review that promises-or threatens, depending on one's point of view-a significant reallocation of resources among the Services.

Additionally, the defense dollars available to the Services for existing missions are being squeezed from two sides: by a large bill for national missile defense, and by a large tax cut. What has not changed is the art of special pleading. Thus have some commentators claimed anew that the NATO bombing campaigns against Yugoslavia in 1995 and 1999 herald a new era of warfare in which precision weapons and airpower alone promise swift and decisive results with little loss of life or collateral damage. As historian John Keegan has put it, not only did the Kosovo campaign demonstrate that "war can be won by airpower alone", it also provided a showcase for the Western world's "superior technology and higher public morality."

It's just not so. On the one hand, the weapons themselves are still hardly free of major glitches, as the failure in February of Navy Joint Standoff Weapons used in attacking Iraqi air defenses showed. Even the newest technology is susceptible to inclement weather, enemy countermeasures or, in this case, just wind. More important than technical problems with the weapons are misconceptions about how tactics accumulate into a strategy for their use. A close examination of the 1999 Kosovo air campaign demonstrates that, talk of humane warfare notwithstanding, air bombardment remains an extremely destructive action that is most effective in achieving strategic goals when targeted against the civilian elements of a society.

Unpleasant as this conclusion may be, it should not surprise anyone familiar with military history. It is not uncommon that new technologies which appear to have decisive application on the battlefield soon end up being turned upon civilians. In 1870, for example, the Prussians decided that the quickest and ultimately the most humane way to reduce French fortresses was to shell the civilian population with modern artillery until it forced the garrison to surrender. Giulio Douhet incorporated this and similar lessons into his post-World War I theories about bombing cities. American airmen in the 1930s, however, developed a different approach based on the promise of precision attacks. Studying New York City as a model, they concluded that destroying only seventeen targets within its transportation, water and electrical systems would render the city uninhabitable without inflicting mass casualties. They expanded their concept of exploiting key vulnerabilities in the economies of industrialized nations and developed a coherent precision-bombing doctrine that has shaped the evolution and application of American airpower ever since. Resulting Air Corps studies asserted that the principal mission for airpower was "the attack of those vital objectives in a nation's economic structure which will tend to paralyze the nation's ability to wage war", while service school texts proclaimed:

"Direct attack of the civil populace is rejected as an air objective due to humanitarian considerations."

Unfortunately, practice has let theory down. Though technology has continued to advance, public expectations and U.S. Air Force promises about airpower's decisiveness and accuracy have advanced faster. As a result, key decisions about the application of military force in most American wars in the air age have been shaped by an overestimation of airpower's effectiveness against military and industrial targets, and disappointing results have led repeatedly to the escalation of aerial operations against civilians-confirming Douhet's theories and confounding America's precision bombing enthusiasts. Such escalations have long-lasting implications. It may be, for example, that current North Korean programs to develop ballistic missiles are motivated by memories of the destruction of most of their cities and towns by American bombing between 1950 and 1953.

Recent air operations over Yugoslavia repeated the pattern of the Korean War: anticipatory claims of decisiveness, followed by disappointment, followed by escalation against civilian targets. Frustrated by seemingly interminable peace talks and the failure of aerial interdiction, American airmen adopted a strategy they called "Air Pressure": coercion through the destruction of key dual-use civilian-military targets. These targets eventually included hydroelectric power facilities, almost every city and town in North Korea, and irrigation dams for rice fields. Again in Kosovo there were high expectations for what airpower, along with the newest precision-guided munitions and information warfare, could accomplish. While airpower was in the end the primary offensive arm that produced a settlement without risking U.S. and allied ground casualties, the results were not at all those envisioned when the campaign started.

When the bombing commenced, Pentagon planners and State Department spokesmen admitted that they did not expect airpower alone to force President Slobodan Milosevic to surrender Kosovo. Consequently, President Clinton announced that the operation had three primary objectives: to stop the ethnic cleansing in Kosovo; to prevent an even bloodier Serb offensive against civilians there; and to "seriously damage" the Serb military capacity to do such harm. Bombing did not achieve any of those goals; indeed, it exacerbated the assault against Albanian Kosovar civilians as Serb ground forces responded to the high-tech aerial assault with a low-tech ravaging of the province.

As to seriously damaging the Serb military capacity, NATO peacekeepers subsequently discovered that initial estimates of the degradation of Serbian forces from air attacks were vastly exaggerated, primarily due to extensive Serbian use of decoys and deception. NATO officials quickly reduced initial claims of tanks destroyed from 122 to 93, and were then forced to admit uncovering only 26 "kills" when all was said and done. Yugoslav vehicle commanders, it seems, proved quite adept at hiding in villages, using the surrounding community and inhabitants as human shields.

More generally, whether the air campaign motivated Serbians to eventually remove Milosevic from power in due course, or only strengthened his hand against a fragmented opposition in the short term, is still debated. The latter is more likely. What is clearer is that achieving any of NATO's primary objectives in Kosovo by airpower alone was made more difficult by the gradual escalation of the air attacks, which reinforced, by negative example, the lessons learned about the drawbacks of incrementalism in Vietnam, and ignored the lessons of Operation Desert Storm, a massive air effort that quickly paralyzed Iraqi forces and destroyed their ability to fight.

Such incrementalism, however, was only one aspect of a range of problems caused by the need to create consensus for the air campaign within the 19-member NATO coalition. It now appears that targeting in Kosovo was even more micromanaged by civilians than it was in Vietnam. The political and legal constraints resulting from the alliance's ornate and cumbersome decision system produced rules of engagement for pilots as strict as any seen in the history of war. The intention was to minimize civilian casualties; the result was very different.

Despite the layers of civilian control and the promise of precision weaponry, young pilots searching for targets on four-inch square monitors made mistakes. Fears of Serb air defenses kept aircraft at 15,000 feet or higher and further increased the difficulties of target identification, contributing to the tragic attacks on tractorloads of refugees near Djakova on April 14. The video replay of the accidental destruction of a Yugoslav train passing over a bridge just as it was struck by a NATO missile was highlighted on newscasts around the world. Still the Serbs did not yield. As General Wesley Clark expanded his target lists, the quality of his intelligence declined markedly, leading to further mistakes and episodes of futility. Outdated CIA maps led to the accidental bombing of the Chinese Embassy on May 7, but that was only the most spectacular and telegenic error of many.

These errors undermined support for the air war and put considerable pressure on NATO political and military leaders to achieve quicker political results. Clark was close to running out of militarily useful and politically acceptable targets when he secured approval for probably the most important raid of the campaign. The destruction of the transformer yards of the Yugoslav power grid on May 24 disabled everything from the air defense command-and-control network to the country's banking system. It did demonstrate nato's strength and dominance to both the political leaders and the civilian population in Serbia, but the price included the sudden cessation of the electrical supply to hospitals and water-pumping stations. Military lawyers made the implications clear to Clark; one recalled, "We'd have preferred not to have to take on these targets. But this was the Commander's call."⁷ All major Serb cities experienced extended power disruptions until a settlement was reached on June 10.

The Kosovo campaign thus featured anew many of the persistent shortcomings of American airpower. Gauging the decision-making process, vulnerabilities and will of targeted leaders again proved difficult. Problems with insufficient resources and allied sensibilities first limited the conduct of the air campaign, then propelled a sharp escalation of the assault on dual-use civilian-military targets as the conflict continued. Attempts at aerial interdiction of Yugoslav forces exhibited many of the same shortcomings that plagued operations and frustrated expectations in Korea and Vietnam.

And all this was despite the fact that airpower's efficacy in Kosovo-such as it was-benefited grandly from important assistance supplied by both traditional landpower and diplomacy. While a NATO ground offensive was not employed, open discussions about its possibility, and an apparent growing willingness to gather peacekeeping forces in the region, probably influenced Yugoslav leaders. The KLA was also moderately useful for a time in drawing Serb forces out into the open where aircraft could attack them. The Russians, despite their opposition to the NATO attacks, ultimately played a key role at U.S. behest in persuading the Serbian leadership to accept a settlement. They also assisted NATO by having declined to upgrade Yugoslavia's 1970s-vintage air defense system, for modernized antiaircraft technology would have downed more aircraft and tested coalition solidarity as casualties mounted.

Despite all this, Lt. Gen. Michael Short, joint forces air component commander for the Kosovo campaign, still chose to trumpet the success of air power once Milosevic had capitulated. In a breathtaking statement bespeaking both hubris and error, he averred that "NATO got every one of the terms it had stipulated in Rambouillet and beyond Rambouillet, and I credit this as a victory for air power." This was untrue, not least because the Rambouillet ultimatum demanded NATO access and portage throughout all of the Federal Republic of Yugoslavia, and this was not achieved by the war. As for hubris, let the facts brought above speak for themselves.

It is clear that whatever sort of victory nato's Kosovo air campaign brought, it was not achieved by "plinking" tanks or by causing strategic or operational military paralysis. Despite European attempts to restrain the attacks, a less-than-final settlement was achieved with the same sort of "imposed cost" strategy attempted in Korea and Vietnam, resulting in massive destruction of Yugoslavia's civilian infrastructure. Pentagon spokesman Kenneth Bacon admitted as much when he speculated that the main factor in Milosevic's acceptance of terms "was the increasing inconveniences that the bombing campaign was causing in Belgrade and other cities."

The more precise truth is that a yawningly broad definition of the term "dual-use" opened up a wide array of targets for NATO airmen, including bridges, heating plants and television stations. Black humor in Belgrade had it that even bakeries were valid targets for NATO aircraft because "soldiers also eat bread." When the 78 days of attack on cities, factories, power plants and other targets finally ended, 45 percent of Yugoslavia's television broadcast capability was degraded and a third of military and civilian radio relay networks were damaged. Petroleum refining facilities were completely eliminated. Seventy percent of road and 50 percent of rail bridges across the Danube were down. Destruction is what airpower does best, and it contributed to an enormous Balkan repair bill that the president of the World Bank fears will use up any money available to deal with other humanitarian crises in Asia and Africa.

There is also a good chance that Yugoslav civilian casualties exceeded military ones, suggesting that Mr. Bacon's choice of the term "inconveniences" did not quite capture the extant reality. This is particularly ironic considering the expectations for a bloodless war caricatured so well in Doonesbury cartoons and reinforced by NATO briefings on targeting accuracy. High NATO expectations for extremely low casualties on both sides helped convince reluctant coalition members to support the air campaign, and hence increased the political impact of each scene of civilian dead and wounded. Such expectations may also have played some role in persuading UN war crimes prosecutors of the International Criminal Tribunal for the Former Yugoslavia to assess evidence in December 1999 that NATO commanders had violated the laws of war with their air attacks.

Wrongheaded attempts to criminalize military activity aside, we must face the ethical implications of the fact that NATO airpower achieved success only because of the destruction wreaked on civilian infrastructure and the resultant noncombatant casualties. This does not necessarily make the application of such military force wrong; that judgment depends on a calculus-never easy to parse-of the greater and lesser evils to be considered in any given case. But it should dispel all illusions about the surgical bloodlessness of modern military technology, airpower not least.

In our ethical accounting, too, we need to factor in the role of precedent. One nation, in particular, quickly applied Kosovo's lessons on airpower to its own problems. Though they opposed NATO air strikes on Yugoslavia, the Russians soon applied the same tactics against Chechnya, trying to force the withdrawal of Islamic militants without a bloody and unpopular ground war. Russian newspapers declared that "the Kremlin is using the scheme NATO applied in Kosovo", and indeed, the Russian Air Force Commander, Col. Gen. Anatoly Kornukov, conducted NATO-style briefings narrating combat footage of precision munitions striking targets in Grozny. His forces were also conscious of the need to justify the destruction of dual-use targets. Russian aircraft destroyed Chechnya's cellular telephone exchange to prevent the rebels from talking to each other, bombed oil refineries to deprive the rebels of black market revenues, and presumably attacked the television station to stop the transmission of misleading propaganda-just like NATO did. As the damage to the civilian infrastructure and the civilian death toll mounted, more than 100,000 Chechens became refugees fleeing the violence.

Instead of demonstrating how airpower in limited war allows "righteous" states to restrain transgressors with a minimum of bloodshed on both sides, the Kosovo and Chechen conflicts show how aggressive belligerents with advanced technology can inflict massive destruction on others at low cost to themselves. They also show how modern technology has inclined to merge the civilian and military sectors of society to an unprecedented degree, creating a broader target spectrum that can be justified for attack. Instead of making war less likely and destructive, the power of new "precision strike" technology has done the opposite. It is now much easier to get domestic support to use force when all it requires is to launch a cruise missile or drop a precision bomb, for the expectation is that results will be clean, decisive and above all safe for the attacking side. When the results are neither clean nor decisive, it is easier to escalate a conflict as impatience grows, especially when there is a considerable technological mismatch between belligerents. So instead of heading for the sort of future envisioned by modern American airpower theorists, in which paralyzing attacks on military structures end wars quickly and with relatively little impact on the civilian sphere, we may be headed for that foreseen by Douhet, in which new weapons decide wars by inflicting maximum distress on civilian populations, which are inherently more vulnerable to the destructive power of modern technology than are military capabilities.

Particularly in lopsided contests, the fact of civilian vulnerability in turn raises the question posed by an anonymous European observer in the aftermath of the Kosovo war: "Now we know what Americans are willing to kill for; but what are Americans willing to die for?" It is a good question, for it gets at the choices and priorities that even great powers cannot-or, at any rate, should not-neglect, even if advanced military technology tempts us into thinking otherwise. Particularly for America, as another contributor to these pages once put it, "even a high-tech military offers no easy escape from the moral ambiguities that remain the lot of an imperial democracy."

This is not a new problem. During World War II, the Army Air Force staff created a model reply to answer any letters they received questioning the humanitarian application of their new technology. One section read:

We see air warfare as being different only in the range of its potential destruction. The air gives uncurbed bestial instincts a wider field of expression, leaving only humanity and common sense to dictate limitations. Law cannot limit what physics makes possible. We can depend for moderation only upon reason and humane instincts when we exercise such a power.

Would that such moderation and reason came in as ample a supply as the precision weapons in our arsenals.