The first of the giants of American grand strategy during the Cold War lived to be the last of the giants. When George F. Kennan died a few weeks ago at the age of 101, none of his great contemporaries was left. Truman, Marshall, Acheson, Forrestal, Harriman, Bohlen, and Lovett had all preceded him in death years ago; and even Kennan's most formidable rival on matters of policy, his longtime friend Paul Nitze, died last fall at 97.

It is an appropriate moment, therefore, to assess what Kennan and his generation accomplished. For although great grand strategies are bounded by time and space, they also transcend time and space. They all arise, as the strategy of containment did, within particular periods, places, and sets of circumstances. And yet the adjective "great" implies relevance beyond context. It suggests that the strategy in question may serve as a guide in periods, places, and circumstances yet to come.

When Kennan returned to Washington in the spring of 1946, having riveted the attention of the United States government with the longest telegram ever sent from its embassy in Moscow, his first job was to design a course on strategy and policy at the National War College. "We found ourselves thrown back," he recalled, "on the European thinkers of other ages and generations: on Machiavelli, Clausewitz, Gallieni--even Lawrence of Arabia." Total war in a nuclear age would be "suicidal," but there was no American tradition of limited war. So it was necessary to explore other traditions: Montesquieu's view that "nations ought to do one another in peace the most good, in war--the least possible evil," or Gibbon's claim that the "temperate and indecisive conflicts" of the eighteenth century had been a strength rather than a weakness of that era. Kennan was relying here on the principle of transferability: that grand strategies from the past could suggest what to emulate and what to avoid in shaping grand strategies for the future.

It seems fair enough, then, to apply that standard to the strategy that Kennan himself devised after moving to the State Department early in 1947. To what extent might containment work in other periods, places, and sets of circumstances? Not at all, he seemed to suggest during the Vietnam War: "I emphatically deny the paternity of any efforts to invoke that doctrine today in situations to which it has, and can have, no proper relevance." The possibility that there could be strategies of containment--that his own strategy could spawn mutations of which he disapproved--left Kennan frustrated, apologetic, and often angry. The sensation, he recalled, was that of having "inadvertently loosened a large boulder from the top of a cliff and now helplessly witness[ing] its path of destruction in the valley below, shuddering and wincing at each successive glimpse of disaster."

By the end of the Cold War, however, the successes of containment had clearly outweighed its failures. There was no war with the Soviet Union, as there had been twice with Germany and once with Japan between 1914 and 1945. There was no appeasement either, as there had been in the years between the two world wars. Whatever the miscalculations, whatever the costs, the United States and its allies sustained a strategy that was far more consistent, effective, and morally justifiable than anything their adversaries were able to manage. Indeed, it is difficult to think of any peacetime grand strategy in which the results produced in the end corresponded more closely with the objectives specified at the beginning.

Students of strategy will therefore be studying containment for decades, even centuries, to come. Leaders will be applying its lessons in periods, places, and circumstances that nobody can now foresee. Transferability, however much Kennan resisted the notion, is unavoidable. But because the context can never again be that of the Cold War, not all aspects of that strategy are likely to transfer equally well.

Kennan himself suggested one such aspect: the requirement that the adversary to be contained share one's own sense of risk. Containment probably would not have succeeded against Napoleon or Hitler, he pointed out in 1947, because both gave themselves historical deadlines--determined presumably by their own mortality--for achieving their objectives. Sticking to timetables was more important to them than avoiding war. They lacked the caution that Marxism-Leninism had instilled in Soviet leaders: "the Kremlin is under no ideological compulsion to accomplish its purposes in a hurry," Kennan reminded the readers of Foreign Affairs that summer. Convinced that history was on their side, Stalin and his successors were prepared to be patient. And this bought the time that was needed for containment to demonstrate that they were wrong.

Nor is it clear that containment would have worked against states whose leaders believed in, as Sir Michael Howard has put it, "the inevitability of, and the social necessity for, armed conflict in the development of mankind." Such views were common in the late nineteenth and early twentieth centuries, a fact that helps to explain how so many great powers could have blundered so easily in 1914 into a Great War. But that global conflict, and the one that followed in 1939, profoundly shook "bellicist" assumptions; and the use of atomic bombs in 1945 shattered them. Quite apart from the presence of a cautious adversary, therefore, there was in the postwar era a far more favorable psychological climate than had previously existed for developing measures, such as containment, short of war.

This sense of shared risk persisted throughout the Cold War, which is why the adjective remained attached to the noun. It did not matter whether Democrats or Republicans occupied the White House, or whether reformers or reactionaries inhabited the Kremlin: all feared a third world war. All had societies to defend and states to preserve. Total war had ceased to be a means by which those objectives could be accomplished, even if limited wars were still possible. It is hardly surprising, then, that Kennan and his war-college students read Clausewitz, for it was his great principle that the use of force must never become an end in itself: "The political object is the goal, war is the means of reaching it, and means can never be considered in isolation from their purpose." No major leader during the Cold War would have disagreed.

That fact suggests a second limitation on containment's applicability beyond a Cold War context, which is that it was a state-based strategy. It depended not only on the fear of all-out war, but also on the existence of identifiable regimes that could manage the running of risks short of war. This, too, was consistent with Clausewitz: where else could the capacity to constrain force come from if not the state, the entity created, at the dawn of the modern era, to monopolize the means of violence? To imagine Clausewitz apart from the state is to imagine a boat without water. Might the same be said, then, of containment? Can that strategy function in an environment in which states are no longer the principal threats to be contained?

The attacks of September 11, 2001 posed that question for the United States in the starkest possible terms. The Bush administration was quick to conclude that Cold War strategies--containment and the deterrence that accompanied it--would not have worked against Al Qaeda. How does one contain someone who, before striking, is invisible? How does one deter someone who, in the act of striking, is prepared to commit suicide? These problems led Bush to announce a new grand strategy of pre-emption. The United States would henceforth act multilaterally where possible, but unilaterally where necessary, to destroy terrorists before they could hit their intended targets. The purpose was to defend states against stateless enemies.

But Bush's strategy was less of an innovation than at first it seemed to be. Pre-emption had never been ruled out during the Cold War. No American president, in a nuclear age, would have knowingly risked another Pearl Harbor. Nor was Al Qaeda an entirely stateless enemy. Osama bin Laden ran it from Taliban-controlled Afghanistan, against which the Bush administration swiftly and successfully retaliated in the fall of 2001. Its first clear act of pre-emption also took place against a state, Iraq, in 2003. One purpose was to frighten the leaders of any other states who might be harboring terrorists or thinking about doing so: but that was deterrence, with a view to countering an anticipated danger. Pre-emption by the Bush administration's logic, then, led back to containment. It did not replace containment.

Yet there is another way of understanding September 11 that might indeed make containment obsolete. It comes from claims that the attacks could only have occurred because of a diminished capacity on the part of all states to control what happened within their territories and across their borders. If September 11 initiated a new age of insecurity in which the actions of only a few individuals could endanger entire societies, then strategies of containment as traditionally conceived would be of little use. Containment presumed threats from states seeking to survive. It was never designed for movements seeking martyrdom. Preemption in such situations, the argument runs, may be the only feasible option.

A third limitation on containment's relevance beyond a Cold War context has to do with the persistence, throughout that conflict, of something obviously worse than American hegemony. It is clear in retrospect that the United States retained a preponderance of power-- in all of the categories that constitute power--throughout the last half of the twentieth century. But it did so more often by invitation than by imposition: as long as the Soviet Union was the alternative, there was always something worse, in the eyes of most of the rest of the world, than the prospect of American domination. That minimized the "friction"--to use Clausewitz's term--that hegemony might otherwise have generated.

With the end of the Cold War, however, the unintended advantage that the Soviet Union had given the United States disappeared--as did the urgency of cultivating allies and neutrals who, if neglected, might defect to the other side, or at least threaten to do so. Multilateral consultation diminished steadily during the administrations of George H.W. Bush and Bill Clinton, not because the principle was objectionable, but because the practice seemed less necessary than it had during the Cold War. George W. Bush's administration inherited unilateralism. It did not invent it.

But it certainly intensified unilateralism in several ways: through tactless diplomacy with respect to the Kyoto Protocol on global climate change, the International Criminal Court, and the Anti-Ballistic Missile Treaty; through the casualness with which it brushed aside offers of help from NATO allies in invading Afghanistan; through its single-minded determination to overthrow Saddam Hussein despite widespread opposition within the international community; and through its reluctance to acknowledge, having conquered Iraq, that it had no clear idea what to do there. All of this led to an unprecedented loss of support throughout the rest of the world for the United States and its foreign policy objectives. The view seemed to be emerging that there could be nothing worse than American hegemony if it was to be used in this way.

If this trend continues, the basis for American power will indeed have shifted from invitation to imposition, a very different context from the one in which containment arose during the Cold War. When Kennan wrote in 1947 that "the United States need only measure up to its own best traditions," he assumed that those traditions would have greater appeal beyond American borders than would those of the Soviet Union and the international communist movement. He was right: the existence of such rivals provided an eminently realistic reason for Americans to respect their own ideals and to try to reflect them, for the most part successfully, in their actions. But if the United States now ceases to do that--if it creates a new tradition of imposed rather than invited power--then it should hardly be surprised to find little that might transfer from the strategy of containment that produced its own pre-eminence.

Containment cannot be expected to succeed, therefore, in circumstances that differ significantly from those that gave rise to it and sustained it, and within which it eventually prevailed. Kennan's objection to invoking it "in situations to which it has, and can have, no proper relevance," in this sense, makes sense. He never claimed that the pre-containment strategies he studied at the National War College could be wrenched from their historical contexts and applied uncritically in the early Cold War. And yet he obviously did believe in selective transferability: otherwise there would have been no point in studying the grand strategists of the past. It is worth considering, then, what aspects of containment might remain relevant in a post-Cold War, post-September 11 world.

One has to do with a kind of intellectual geography. Kennan's strategy of containment mapped out a path between dangerous--even deadly--alternatives. Despite the persistence of a multipolar international system, the dominant trend in thinking about strategy through the end of World War II had been one of bipolar extremes: war or peace, victory or defeat, appeasement or annihilation. The idea that there could be something in between--neither war nor peace, neither victory nor defeat, neither appeasement nor annihilation--had never been clearly articulated. It had been implicit, as Kennan noted, in the strategies of earlier eras, but it had disappeared with the advent of mass mobilizations, lethal technologies, and the total wars they made possible. Imagination itself had failed, making the first half of the twentieth century a period of unprecedented violence among the great powers. There seemed to be no middle ground.

The second half of the twentieth century turned out to be very different. Despite the emergence of a bipolar international system, the dominant trend in thinking about strategy was one of avoiding extremes. Nuclear weapons had something to do with this, but so did the idea of containment, which preceded the Soviet-American nuclear stalemate by almost a decade. When seen in this context, containment was a feat of imagination, made all the more impressive by the bleak circumstances in which it originated. The transferable lesson here is a psychological one: that any strategy in which the only choices available are deadly, dangerous, or otherwise undesirable requires re-thinking.

A second transferable principle follows: that enemies should be encouraged to defeat themselves. The idea goes back at least to Sun Tzu. It pervades Clausewitz. It is what Marx and Lenin expected would happen to capitalism, the internal contradictions of which were supposed to bring about its collapse. Both Kennan in the late 1940s and Reagan in the early 1980s reversed this logic, insisting that it was Marxism-Leninism, not capitalism, that carried within itself the seeds of its own destruction. The United States could, through its actions, increase the strains under which the Soviet Union and its allies operated; but in the end the inefficiencies of command economies, the absence of political accountability, and the improbability that an internationalist ideology could indefinitely suppress nationalist instincts would cause communism's demise.

Exploiting contradictions also makes sense in a post-September 11 era, because the interests of terrorists and those of the states that support them are not always the same. Terrorists have no economic program; states in an increasingly interdependent world must have one. Terrorists substitute intimidation for representation, a bargain that has not proved sustainable in a democratizing age. And states seek to survive even if terrorists do not: even rogue states have an interest in preserving the international state system, because they have no way of knowing what might replace it. These contradictions are at least as striking as the ones within the communist world of which the practitioners of containment took advantage, and over the past few months in the Middle East their consequences have become unmistakable.

How does one know, though, when such opportunities might arise? Answering this question brings up a third transferable principle from the strategy of containment, which is that history is more useful than theory in anticipating what is to come. Kennan's insights during the early Cold War went well beyond the conventional wisdom of the time. He insisted that Stalin was not another Hitler; that an authoritarian state need not be impermeable to external influences; that an ideology based upon a deterministic view of the past could miscalculate the future; that international communism would not remain monolithic; that war and appeasement were not the only choices open to the United States and its allies in dealing with the dangers that confronted them.

It is not at all clear which theories might have yielded such conclusions. They came instead from Kennan's reading of Gibbon on the Roman Empire, from his knowledge of the history and the culture of Russia, from his own crash course on the great grand strategists while at the National War College--and even from works of imagination, as when he cited Thomas Mann's Buddenbrooks to make the point, about the Soviet Union, that "human institutions often show the greatest outward brilliance at a moment when inner decay is . . . farthest advanced." Formal theory, in seeking universal validity, too often disconnects itself from the flow of time. It pays insufficient attention to how things became what they are, which usually offers the best clue to what they will become. History, in contrast--and also literature--distills past experience in such a way as to prepare one for future uncertainties.

The process is intuitive, even impressionistic, involving the ability to see that a situation is "like" one or more that have existed in the past, and that it is worth knowing how the analogous cases were handled. It requires the self-confidence to be selective, the self-discipline to be clear, and a certain amount of self-dramatization to get one's point across. (What else could an eight-thousand-word telegram have been?) It benefits from insights into human nature, which theory rarely provides. It is ironic that Kennan is remembered today as one of the founders of "realism" within the field of international relations, for he never considered himself to be a theorist at all. He was by temperament, by training, and later in life by choice a historian. And he would have liked to have been a novelist.

A fourth aspect of the strategy of containment that might be transposed to other contexts has to do with the relationship between coherence and accountability. Here Kennan was less than prescient: he worried that the volatilities of domestic politics would make it difficult to sustain a consistent grand strategy. He once compared democracy, famously but unwisely, to a "prehistoric monster with ... a brain the size of a pin." In a sense, he was right: there were repeated re-inventions of containment, with new administrations too mindlessly rejecting the ways their predecessors had applied it. When one looks at the American record throughout the Cold War, though, one cannot help but be struck by the extent to which the larger objectives of containment--avoiding the extremes of war and appeasement while waiting for the Soviet Union to change itself--remained the same, regardless of which party occupied the White House and which approach to containment each chose to embrace.

From that perspective, then, the volatilities look like course corrections imposed by the obligation of democratic accountability. The requirement to hold an election every four years may have made it hard to maintain consistency, but it was a safeguard against complacency, against the compulsion to continue failing strategies in the face of evidence that they were failing. To see the value of this, consider the performance of the Soviet Union, China, and the satellite regimes of Eastern Europe, where the only way to replace defunct strategies was to wait for their architects to die or to be overthrown. That did happen, of course, but not frequently enough to provide protection against the dangers of authoritarian autism--the tendency to persist in error that the absence of accountability encourages.

Nor did shifting approaches to containment impede another kind of accountability, which was the need to combine leadership with consent. It is striking that after four and a half decades of the Cold War, the alliances with which the United States began that contest were largely intact, while the Soviet Union had hardly any allies left. The prospect of something worse than American hegemony helps in part to explain this outcome. It is also the case, though, that the strategists of containment never underestimated the importance of allies. They worked hard to maintain multilateral consent for American leadership in waging the Cold War, without at the same time allowing the need for consultation to paralyze their alliances. Containment in that respect also sets a standard to which future grand strategists--perhaps even current ones--might aspire.

A final lesson from the past that will be usable in the future comes chiefly from Eisenhower, although Kennan agreed with it: it was that containment must not destroy what it was attempting to defend. Eisenhower's concern was that, in the effort to resist an authoritarian adversary, the United States itself might become authoritarian, whether through the imposition of a command economy or through the abridgment of democratic procedures. That never happened. Despite the military-industrial complex, the nation maintained its markets; despite McCarthyism, it sustained and ultimately strengthened civil liberties; despite the excesses of Vietnam and Watergate, the strategy of containment never came close to corrupting fundamental American values. They remained, at the end of the Cold War, what they had been at its beginning. The same can hardly be said of fundamental Marxist-Leninist values. So in this sense, too, containment was consistent with Clausewitz: it was an extension of war, diplomacy, and values by other means.

On February 22, 1946, sick in bed from the rigors of a Moscow winter and irritated as usual at the Department of State, Kennan summoned his secretary, Dorothy Hessman, and dictated an unusually long telegram. That document has a better claim than any other to having laid out the path by which the international system found its way from the trajectory of self-destruction it was on during the first half of the twentieth century to a trajectory that had, by the end of the second half, removed the danger of great-power war, revived democracy and capitalism, and thereby enhanced the prospects for human liberty beyond what they had ever been before.

An extravagant claim? Perhaps--but would anyone on that day have regarded the world as safe from the scourge of great-power war? How could it be, when it had not even been possible after World War II to convene a comprehensive peace conference? Or as safe from the dangers of authoritarianism? How could it be, when the Western democracies had to rely, during the war, upon one authoritarian state to defeat the others? Or as safe from a recurrence of economic collapse? How could it be, since there was no assurance that another global depression would not return? Or as safe from abuses of human rights? How could it be, when one of the most advanced nations in Europe had just committed the crime of genocide on an unprecedented scale? Or as safe from the fear that in any future war no one would be safe? How could the world be that either, with atomic weapons having been developed, and with little prospect that they would remain under exclusive American control?

What Kennan opened up, on that bleak day in Moscow in 1946, was a way out: a path that rejected the appeasement and the isolationism that had led to World War II, on the one hand, and on the other the alternative of a third world war, the devastation from which, in a nuclear age, would have been unimaginable. Fifty-nine years later there is much more than a long telegram--and a very long life--for which to be grateful.

By John Lewis Gaddis

Tory Story

By Geoffrey Wheatcroft

Head of His Class

By Alexandra Starr

After Containment

By John Lewis Gaddis

Page 1