It takes remarkably little time for a Washington witticism to become a cliche. Such has been the case recently with the quip that Donald Rumsfeld may not have been a very good secretary of defense, but he is a remarkable secretary of war.

In early September 2001, the Defense Department looked to be in poor shape. Since Rumsfeld had taken over, there had been much talk of "defense transformation" -- the successor term for the "revolution in military affairs," a supposedly new way of waging war -- but little evidence of it. Senior military officers, excluded from early studies arranged by the new secretary, grumbled to the press that the Bush administration was treating them even worse than the Clinton administration had. This discontent may have reflected naive expectations that the Republicans would show up with genial smiles and open wallets, but the friction was real -- as was the irritation of many on Capitol Hill and in the press who were put off by Rumsfeld's caustic style.

A planned $18 billion increase in the defense budget, meanwhile, most of it earmarked for personnel costs and not envisioned as part of a series of further increases, fell far short of what many felt was needed to make up for the shortfalls of the previous several years. Difficult decisions such as the cancellations of major weapon systems appeared necessary yet were not forthcoming. In the works was a Quadrennial Defense Review that spoke of shifting away from a narrowly defined set of two major contingencies as the chief planning construct for the Pentagon, but there did not seem to be much of a link between words and programs.

Then came September 11. From the moment the secretary dashed out of the burning Pentagon to rescue wounded subordinates, perceptions of his leadership reversed. In a time of war, Rumsfeld's disagreeable brusqueness appeared as refreshing honesty; his uncomfortably hard edge became the kind of resolution required of a leader; his willingness to badger his generals was not an absence of diplomacy but a firm hand on the reins. From the object of polite derision at cocktail parties, he became the hero of satirical skits on Saturday Night Live. For the armed forces fighting the war on terrorism, meanwhile, money was suddenly no object. Plans for this year's defense budget increase shot to nearly $80 billion, and talk of hard choices was put off.

The jape about the two secretaries reflects a certain truth, one borne out by historical experience: the great war ministers of the past, such as Abraham Lincoln's ferocious secretary of war, Edwin Stanton, would rarely have succeeded in peacetime. But to understand defense policy as merely a function of personality would be inadequate. The pre-9ffi11 frustrations emerged not so much because of Rumsfeld's personal style as because of the inherent difficulties in guiding a behemoth such as the Defense Department onto a new course. And the post-9ffi11 amity obscures the tough questions that remain on the table: Is the Pentagon headed for fundamental change, and, if so, is that change necessary or even desirable? What lessons, if any, should its officials take away from the operations in Afghanistan? And what comes next if the war that flamed into the open on September 11 spreads and continues for years or even decades to come?


Peter Drucker, the great student of management, once suggested that "any job that has defeated two or three men in succession ... must be assumed unfit for human beings." After musing about then Secretary of Defense Robert McNamara, he concluded, "I am not yet convinced that the job of Secretary of Defense of the United States is really possible (though I admit I cannot conceive of an alternative)." Three decades later, Drucker's observation is even more apt than when it was first made, and students of the contemporary American military would do well to consider its implications rather than obsess over an individual's quirks and crotchets. The truth is that being secretary of defense is an impossible job -- if one hopes to change quickly the institutions of the armed forces rather than merely preside over them.

The reasons are legion, beginning with the American system of government's reliance on political appointees. What is needed is not merely a Defense Department head but leadership four levels down (the deputy secretary, undersecretaries, assistant secretaries, and deputy assistant secretaries, along with many of their support staff -- all told, a group numbering in the hundreds). This reliance is not a bad thing, but invariably, no matter what the administration, intermingled with the experienced, talented, and energetic appointees come the hacks, the political payoffs, and the incompetents.

More important, it has become increasingly difficult to staff an administration promptly. The exhaustive vetting associated with the confirmation process deters some of the best candidates available, as do compensation packages well below those available in the private sector and even universities. It frequently takes, as it has in the Bush administration, the better part of a year to staff the Pentagon, and then even the quick learners take another six months or more to learn the system, brush up on arcane jargon, and master obscure bureaucratic procedures.

For more than a year, therefore, the department is in the hands of a civilian leadership that is understaffed, overworked, and desperately behind. The natural resort of such civilians is to fall back on the bureaucracy, and the military staffers above all. From a harried political appointee's perspective, they are a godsend: masters of the details of their jobs, hard working, cheerful, bright, deferential. The disdainful phrase "captured by the bureaucracy" altogether underestimates the subtlety and power of the process that prevents new administrations from trying hard to reorient the military, let alone succeeding.

As long as the fundamental premise of military competence holds, any attempt to change the armed services by some dramatic organizational coup is doomed to fail. It is tough enough to sack failed generals, terminate programs that produce weapons that do not work, and dismantle bureaucracies responsible for battlefield disaster. A civilian leadership attempting to redirect a military that is obviously successful -- and America has been winning its wars for more than a decade -- is destined to have a rough time of it.

Bureaucratic norms and pressures have an almost overwhelming force. For example, even so apparently reasonable and modest a measure of the early Rumsfeld Pentagon as the creation of a senior executive council foundered. The council would have allowed the secretary of defense to exert influence through the secretaries of the different services. This move would have been useful because, under the current structure, the service secretaries have tended to defend their individual services' programs and perspectives against external attack rather than acting in coordination with their boss.

Absent such reforms, unsurprisingly, the services and existing military bureaucracy dominate Pentagon affairs -- which is why the Bush administration's future defense plans, even with their dramatically increased size, basically conform to and confirm pre-Bush administration thinking. Most of the old systems scheduled for procurement did very well in the 2003 budget, including some, such as the Army's Crusader self-propelled howitzer, that had publicly been labeled as likely candidates for sacrifice on the altar of transformation. There are some new emphases -- missile defense, most notably, and space-related operations -- but in general the new budget is very much a Clinton plan, only far better funded.

The reasons for this inertia are obvious. Because of the sunk costs looming over decision-makers, the domestic and international trauma likely to be caused by the termination of large systems, and the sheer momentum behind long-standing programs such as the Joint Strike Fighter or the v-22 Osprey airplane, cutting major programs inevitably inflicts pain on powerful interest groups and thus requires exceptional political will.

For more than a decade, furthermore, the armed forces have suffered a procurement drought. Republicans and Democrats alike have maintained an essentially static budget, a large and reasonably well-paid force, and an extremely active military presence overseas. They have paid for that budget by skimping on purchases of new equipment, with the result that the U.S. arsenal is now filled with aging airplanes, tanks, and ships. Major replacement programs, such as a new bomber, may take a decade or more to come on-line, so even those officers who know that many of the items now being acquired are hardly cutting-edge are not willing to forgo them.


All this inertia looks like a recipe for distressing immobility, and in some ways it is. Before 9/11, more than one disappointed advocate of transformation had written off the administration's defense policy, and most of those skeptics found no reason to change their views after the release of the 2003 budget. But they may have underestimated important countervailing forces, some of which became apparent in the Afghanistan war (as indeed they had in earlier conflicts as well).

The first is sheer size. Even pre-9/11 U.S. defense planning envisioned annual expenditures of more than $310 billion -- that is, roughly ten times the British defense budget, or considerably more than twice the defense spending of the rest of NATO put together. Inevitably, embedded in such a large budget will be the funds -- substantial in absolute if not relative terms -- for innovative projects. The Navy, for example, had reluctantly accepted financing to convert two ballistic-missile-firing submarines into a version of the arsenal ship (a submerged barge for long-range fire support) long advocated by military reformers. The Air Force put most of its weight behind big, traditional aircraft programs, but it was also willing to spend hundreds of millions of dollars developing long-range unmanned aerial vehicles (UAVS), including some that might eventually substitute for airplanes with pilots. And all the while, the boring acquisition of routers and servers, networks and data links, radios and global positioning system (GPS) receivers was laying the base for the networked war that U.S. forces ended up waging in Afghanistan.

By and large, American generals and admirals today are, as senior military officers usually have been, competent and conservative. But underneath them exists a cadre of energetic young officers whose familiarity with information technology and willingness to experiment with it represents something more radical. Increasingly, they have both operational experience and advanced technical education, and they are slowly but surely dragging the system along with them. Whatever stodginess and hierarchy may exist in the U.S. military, the truth is that the majors and lieutenant colonels, and even the sergeants operating the hardware, have a certain power to press their superiors, and they use it.

In this respect, the younger generation is powerfully reinforced by American culture. A poll of military officers will almost invariably show them committed to the revolution in military affairs, or defense transformation, or some other form of radical change -- even when their practice and views (about the importance of heavy armor formations, for example) would suggest a very different array of beliefs. This is not so much hypocrisy as it is false consciousness. As Winston Churchill once noted, one should never underestimate "the enormous and unquestionably helpful part that humbug plays in the social life of great peoples." Like American pragmatism, American eagerness to change, experiment, and modernize pervades society so thoroughly that it cannot help saturating the military as well. A generalized belief in the intrinsic value of transformation can thus coexist with, and will gradually overcome, an attachment to traditional weapons and ways.

As the military has become increasingly aware of the achievements of the private sector in exploiting advanced technology, moreover, it is increasingly willing to emulate it. Once military logisticians realized that Wal-Mart and Federal Express put them to shame in efficiency, it was only a matter of time before they would, and did, figure out ways to adopt similar practices. If even a modest heating-oil distributor uses GPS data and efficient routing software to track and schedule its trucks, there is no reason why the U.S. military cannot have similar systems embedded in most vehicles deployed in the Balkans. Indeed, now it does have them, along with a completely different view of the battlefield than ever before.

It is a sociological truism that military organizations reflect the societies from which they emerge, and here at least the United States is no exception. However jerky the transmission belt, the qualities of the modern American economy -- its adventurousness, spontaneity, and willingness to share information -- eventually reach the American military. Just as the teenager who grew up tinkering with automobile engines helped make the motorized armies of World War II work, so do the sergeants accustomed to playing video games, surfing Web pages, and creating spreadsheets make the information-age military of today effective.


All this said, the truth is that even the generous increases in defense spending now planned for 2003 and later will not eliminate the problems that existed before September 11. Nor will they address the new conditions in which the Pentagon operates. The fundamental problems plaguing weapons procurement and infrastructure maintenance have not been solved, as much of the new funding will go to war-related costs and increased spending on personnel. The 2003 budget boosts procurement by more than ten percent, or almost $8 billion, and such spending is projected to rise about $30 billion more by 2007. Yet unless some of the large procurement programs currently on the books are drastically scaled back or eliminated, they will burst through even that huge figure.

The pathologies of the military procurement system, moreover, remain intact. After the first months of the Afghan war, for example, there were reports of severe shortages in many of the precision-guided weapons used so effectively there, such as the Joint Direct Attack Munition (JDAM) -- a relatively inexpensive modification for conventional bombs that enables satellite-guided precision delivery from virtually any weapons platform under all conditions. This was hardly the first time such shortages had occurred. American arsenals of precision weapons were also depleted after both the 1991 Persian Gulf War and the 1999 Kosovo campaign. The shortage in Afghanistan thus represented the third consecutive failure of the Pentagon to stockpile the right munitions or create a system for surge production of them when needed. The 2003 budget jacks up ammunition spending sharply, but the traditional bureaucratic logic of buying platforms and assuming that the weapons will follow is only slowly yielding to the imperatives of war.

The difficulty in bringing adequate spending to bear on UAVS is a good example of the Pentagon's bureaucratic rigidities, what Rumsfeld recently described as "people who come in with approaches and recommendations and suggestions and requests that reflect a mindset that is exactly the same as before September 11." The Afghan war has demonstrated the enormous value UAVS can provide by spending hour upon hour monitoring enemy positions, transmitting video and other data back to command posts and attack aircraft. It has also shown how they can be used as platforms to deliver lethal attacks. And yet the Predator UAV, one of the technological stars of Afghanistan (and Kosovo), was judged not "operationally effective or suitable" by the Pentagon's office of testing and evaluation in October 2001. This determination had less to do with the qualities of the Predator than it did with the extraordinary standards for effectiveness set by the department. It was a classic case of impossibly demanding requirements causing the Pentagon to disparage its own systems, creating pressure to defer adequate acquisition of what is good today in a perpetual quest for the extraordinary system that will do anything and everything tomorrow. In the 2003 budget, total UAV spending -- including research and development as well as acquisition -- comes to not much more than a billion dollars, even as the department plans to spend more than $4.5 billion buying f-22 fighters at $200 million each, and more than $3 billion to get Navy F/A-18 fighters at $70 million each.

The logic of the war, or set of wars, upon which the United States has now embarked, furthermore, will pose new challenges beyond classic military conservatism. The United States, like most imperial powers before it, will find it extraordinarily difficult to withdraw from engagements abroad. The Pentagon would dearly love to leave the former Yugoslavia to the supervision of its European allies, which together have a GDP comparable to that of the United States. Yet the political truth remains that American presence is the indispensable catalyst for NATO action, so thousands of U.S. soldiers will continue to police an uneasy peace in the hills of Bosnia and Kosovo. Rumsfeld yearns to withdraw an American infantry battalion from the Sinai desert, where its military skills rust as it presides over a decades-old peace between Egypt and Israel. Yet doing so now, as war metastasizes in the Levant, will be blocked by objections from both the State Department and the local parties.

The Air Force, meanwhile, has been wearing itself out flying combat patrol over U.S. cities, and homeland defense will not be cheap even though the National Guard will bear most of its burdens. U.S. forces have begun constructing semipermanent bases in South and Central Asia, and it is unlikely that the United States will ever revert to the minimal presence it maintained there before last fall. The tempo of U.S. activity elsewhere in Asia has also picked up, with American forces engaging Muslim guerrillas in the Philippines. Other operations in Yemen and elsewhere remain a possibility, and a full-scale war with Iraq looms on the horizon. In the background, Chinese military modernization proceeds apace.

What, then, are the defense requirements of the future? They begin with the capability to project effective military power rapidly to most locations on the planet. This shift implies a bias toward systems that can either deploy very quickly, without local bases or substantial pre-positioning of supplies, or that can hover within range, hopefully below the threshold of local political visibility. Some kind of new bomber is clearly needed, whether manned or unmanned, based on older designs or not. So, too, are guided missile submarines or other versions of arsenal ships, and the kinds of ground forces that can use advanced technologies to call in decisive long-range fire, as the special forces have done in Afghanistan.

American forces are, on the whole, extremely flexible at tactics but more rigid institutionally. Although some stirrings are visible, the services have generally been unwilling to explore moving beyond old forms of organization (e.g., divisions in the Army, fixed-wing aviation in the Marine Corps, or large ships in the Navy). The coming decade, however, will generate even greater stress than the previous one. The armed forces are destined to engage in all-out conventional war and peacetime training of foreign forces, spectacular raids and humdrum military government, operational offense and tactical defense, deterrence and preemption, violent assault and patient vigilance. To handle all these tasks successfully, far broader and deeper transformation will be necessary than has occurred until now. The department's pre-9ffi11 battles will therefore have to be refought.


The common use of the term "post-Cold War era" indicates a failure by students of international affairs to characterize today's world. The Rumsfeld Pentagon, in its early presentations, always highlighted the unpredictability of the international environment, and it had a valid point: one might usefully call the past dozen years "the age of surprises." The U.S. government has been surprised by the end of the Warsaw Pact, the disintegration of the Soviet Union, the Iraqi invasion of Kuwait and the ensuing Persian Gulf War, the Asian financial crisis, the Indian and Pakistani nuclear detonations, and now the events of September 11. There is no reason to think that the age of surprises is over, and there are many reasons to think we are still at its beginning.

There are some things defense planners can know with some confidence: that the United States faces a larger and more durable threat from radical Islamists than it previously understood, and that in China it confronts a regional rival. They know that America looks out on an international order that is, for better and for worse, global and dynamic. This order's language is English and its lifeblood is the massive and instantaneous flow of information, but its economic surges and ebbs have left many societies sorely disadvantaged, elites embittered, and state structures weakened. Defense planners know above all that the United States will and must play the central role in ordering this international system, a function for which it is both uniquely suited and uniquely disabled. And they know that although the United States has an unmatched array of allies, many of them resent its dominance and willingness to act independently even though they will not pay the price or make the effort to substitute for it.

These probabilities should be reflected in defense programming -- in the kinds of forces that the U.S. stations in the Pacific, for example, or the emphasis placed on recruiting and equipping special operations forces able to operate in the Arab world. But the uncertainties that lie ahead are equally significant. The pace and intensity of the current conflict will depend partly on the behavior of America's opponents. A salvo of biological agents launched at Israel from Iraq could spark a Middle Eastern war before the United States is ready for one. The shaky Pakistani regime could collapse or become embroiled in a war with India that could lead to a nuclear holocaust. Collapsing states could drag the United States into wars it has studiously avoided. Technological breakthroughs could conceivably nullify important elements of American military strength. Terrorists could successfully use weapons of mass destruction against the United States.

The Pentagon now combines a planning device of two major theater wars with what is called "capabilities planning." Yet even the latter, a somewhat abstract statement of requirements, is not fully adequate. To meet the range of known and unknown challenges the United States faces, the military will need to cultivate two functions that it has largely neglected during the last busy decade and before: mobilization and professional education.

The increasing use of reservists and National Guard personnel for routine military operations not only puts stress on the lives of those called up for months or even a year at a time. It also runs the risk of exhausting their willingness to serve and indeed undermines the very idea of the citizen soldier. In the past, mobilization meant calling up young men for conscript service, and it is conceivable (if highly unlikely) that this could be necessary once again. More likely, however, will be the need to tap internal reservoirs of military personnel in an emergency -- which is why, to take a petty example, it may not have been altogether wise to replace military sentries at major posts with cheaper contract guards. At the same time, the United States may find itself making even greater use of quasi-military contractors to do a great deal of what looks like military business -- for example, flying and maintaining UAVS.

Moreover, the Pentagon will have to find better ways of creating an industrial infrastructure that can increase the production of critical munitions and platforms in short order. This change may require different procurement strategies, and perhaps even a reversion to the World War II concern for having systems that could be mass produced -- a far cry from the increasingly capable, exotic, and expensive platforms of today's military. How to create a variable-sized military, which the country has not had for several decades, will be a continuing challenge.

Above all, the twenty-first-century U.S. military will require an officer corps of unprecedented versatility and intelligence. One great source of American strength in recent decades has been the excellence of its military training system. By and large, that strength remains, although it also suffered in the 1990s from underfunding and the pressure of deployment schedules that disrupted unit cohesion and the ability to hone perishable collective skills. The practices and outlook of the military toward advanced civilian and military education, however, have not kept pace with the rest of the training system. Technical degrees are generally rewarded. Advanced work in the social sciences and humanities, however, is often regarded as a ticket to be punched rather than an opportunity to grow, and younger officers are often effectively punished rather than rewarded for pursuing their intellectual ambitions. And yet never more than today has there been a need for officers who can think broadly and creatively, who can learn swiftly about unfamiliar regions of the world, and who will fall prey neither to cliches nor to comforting assumptions about societies, military organizations, or war itself.

Officers spend a great deal of time in the schoolroom, more than any other group of professionals. Yet there is no evidence that the heads of the war colleges are selected for their competence as educational leaders. No serious proposals exist for creating a military academy that would train field-grade officers from all the services in the new forms of operational art. There is no dramatic expansion of efforts to recruit the best and brightest from elite universities, no willingness to spend the relatively modest sums that would harvest a bumper crop of top-notch degrees in everything from Islamic history to military sociology.

One of the great military machines of the last two centuries, the German general staff, worked because it kept a tight link between its "thinking" and its "doing" elements. The war college in Berlin was a path to success in the general staff, which in turn set doctrine and strategy for the army in general. A slavish or simple-minded imitation of the German formula would be ill-advised, but the example bears pondering. In the future, the United States will need leaders different from those produced by the institutions of the Cold War -- different, in fact, from those who dominate the senior general-officer ranks today. And it will get the officers it needs only by reconstructing its educational system, to which senior leaders so rarely pay sustained attention.


Behind the need for new kinds of officers lie the changes that are taking place in war itself. The wars of the past were fought by armies organized, trained, and equipped for the kind of conflict that dominated international politics from the Peace of Westphalia in 1648 to the end of the twentieth century. This was war waged predominantly by states, for national or ideological purposes, and fought by armies that generally resembled one another.

When colonial or counterinsurgency conflict presented a different paradigm of war, states generally coped by adjusting or adapting their regular conventional forces. Some did well (the British in Malaya), some poorly (the Soviets in Afghanistan), and some turned in a mixed performance (the United States in Vietnam). Still, the dominant form of military power consisted of large forces equipped for an all-out contest of limited duration.

Today, however, war is changing, and an analogy with medieval warfare helps to show how. In the Middle Ages a variety of entities waged war -- states, to be sure, but also crusading movements, religious orders, principalities, and entrepreneurs. Today, the al Qaeda terrorist network and the military contractor Brown & Root are each manifestations of the dispersal of war beyond the exclusive precincts of the state. In contrast to the modern era, moreover, when war has flowed from realpolitik, national ambition, or ideological fervor, war in the Middle Ages emerged from an even broader array of motives: state or personal ambition, religious fanaticism, or sheer banditry. And where modern armies resemble one another closely in their organization and equipment, medieval military institutions looked as different as did the individual warriors -- the ponderous knights of western Europe, the highly specialized English longbow archers, or the doughty Swiss pikemen. Whereas modern warfare has tended toward well-defined beginnings and endings, finally, medieval war ebbed and flowed over decades and more, in pulses of violence, siege, and wary truce.

In one respect, however, such analogies break down. In the twenty-first century, characterized like the European Middle Ages by a universal (if problematic) high culture with a universal language, the U.S. military plays an extraordinary and inimitable role. It has become, whether Americans or others like it or not, the ultimate guarantor of

international order -- something quite different from what it was only a few years ago as the leader of a coalition of free states against the well-defined threats of the Cold War.

In the end, therefore, the dilemmas of U.S. defense policy today do not reflect the strengths and weaknesses of any individual, or even the institutional limitations of a giant and sometimes dysfunctional bureaucracy. They stem, rather, from America's profoundly ambivalent and only semiconscious acceptance of its unique and world-historical role. Whatever the pace at which the Pentagon adapts to that fact, it must do so, and the more swiftly and self-consciously, the better. When Rumsfeld snaps in exasperation at the "approaches and recommendations and suggestions and requests that reflect a mindset that is exactly the same as before September 11," he deserves a sympathetic hearing. The secretary of war may have gotten far better press, but the secretary of defense has the tougher job.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Eliot A. Cohen is Professor of Strategic Studies at Johns Hopkins University's Nitze School of Advanced International Studies and a book reviewer for Foreign Affairs. He is the author, most recently, of Supreme Command: Soldiers, Statesmen, and Leadership in Wartime.
  • More By Eliot A. Cohen