John J. Mearsheimer, The Tragedy of Great Power Politics (New York: W.W. Norton, 2001), 448 pp., $27.95.

The absence of a sovereign in international politics, the root of what structural realists call the condition of anarchy, encourages states to look to their own security. Power, especially the power to wage war, is the means to this end, and states try to accumulate power because in the event of aggression by others, they must rely on themselves; they live in a self-help world. All structural realists agree on this point. Where they have come to disagree in recent years is whether states satisfice, in Herbert Simona's artful term, or maximize. Do they struggle for power until - for whatever reasons - they feel reasonably secure? Or do they struggle on until they achieve something like hegemony, the domination of their political environment (however they define both domination and environment)?

The satisficers are now called defensive realists; the maximizers offensive realists. John J. Mearsheimer, in The Tragedy of Great Power Politics, establishes himself as the leader of the offensive realists, for his is the first systematic, historically grounded statement of the theory of offensive realism. Informally, Mearsheimer's analysis also constitutes a test of that theory. The result is a superb book: the theory is stated with admirable clarity; evidence is marshaled by the battalion. Though the non-specialist reader must assimilate a slightly esoteric set of terms, there are only a few of them and they are so colorful that Mearsheimer need only explain them once. Few international relations theorists would introduce a new term into political science such as bait and bleed strategy; perhaps more should do so because the reader does not easily forget such evocative language. Anyone interested in international politics can read, enjoy, understand and argue with this book. It is not without problems, which will be taken up below, but first to the argument.

The Theory

Realists, or rather structural realists as they have been termed since the publication of Kenneth Waltz's Theory of International Politics in 1979, believe that broad patterns of behavior among the most powerful states in international politics over hundreds of years can be explained by the fact that all states must deal with the same problem: anarchy. This critical problem shapes the way states live. States must look to their own defense. They arm when they might prefer not to. They ally with countries they loathe. They might even start wars when they prefer peace. States work hard to make themselves secure. Nevertheless, many structural realists (defensive realists) believe that despite the condition of anarchy, states will stop well short of hegemony if geography or technology or social organization favors the defense. States seek power as a means to an end security. Starting from a world of fear and struggle, defensive realists come to a qualifiedly optimistic conclusion: that, at least in theory, conditions exist under which all states can be safe, even in anarchy. Wars could still happen for any number of reasons, but violent struggle for security may not be inevitable.

Mearsheimer also believes that states seek power to attain security, but he argues that so long as anarchy exists, no state can ever be truly secure. He asserts that even powerful states will seldom believe they have enough power to remain secure so long as others have significant power. Instead, states will keep trying to add to their power. They will try to maximize relative power by moving to increase their share of all the power in the world. Indeed, any state that gets the chance will bid for hegemony: it will try to make itself unchallengeable militarily by any state capable enough, and close enough, to ever threaten it with damage. This is a key disagreement between offensive and defensive realists. There is no happy ending for Mearsheimer, hence the "tragedy" of great power politics in his title.

Mearsheimer does not argue, however, that all states seek hegemony all the time. Most lack the capacity to achieve hegemony, and though they are interested in improving their relative power position, they are also interested in protecting their existing power position, which their attempts at aggrandizement may jeopardize. But what kind of hegemony do capable states seek? Mearsheimer argues that they bid for regional hegemony.

A regional hegemon seeks to be the most powerful state in its surrounding area, especially on land. If strong enough, it will wage war to finish off any near peer competitors around it. The power of a state derives from the size of its economy and its population, but these factors must be distilled into land power for a state to be a great power. Mearsheimer argues that land power is the most important kind in international politics because land power wins wars. Land power is based mainly on the size and quality of a state's army and, in modern times, the air and naval capabilities that contribute directly to the striking power of armies. Aspiring hegemons, therefore, seek to be the most capable land power in their region, and by a very substantial margin.

Aspiring hegemons cause most of the excitement in international politics: ultimately, other states that value their sovereignty have to stop them, but their efforts to contain aspiring hegemons are plagued with difficulties. When an aspiring hegemon starts to move, neighboring states will be slow to balance its expanding power. Neighboring powers will form balancing coalitions if they must, but their first preference is to see if some one power among them will do the heavy lifting. States pass the security buck to others if they can, hoping that another state will expend its own strength to stop or at least weaken the aspiring hegemon, leaving the buckpasser stronger relative to both. Mearsheimer observes that buckpassing is even more common than balancing, and on this point he tilts with “defensive realists who argue that balancing is the most common response to hegemonic ambitions. Offensive and defensive realists agree, however, that the least likely response is bandwagoning. States seldom join up with the expansionist program of the most powerful state in the system unless they are too weak to play either the balancing or buckpassing game. Bandwagoning, all realists agree, at best dooms a state to be the hegemon's final victim.

The choices a state has in dealing with an aspiring hegemon are influenced heavily by the constellation of power in the international system's bipolarity, balanced multipolarity or unbalanced multipolarity as tempered by the facts of geography. Like Waltz, Mearsheimer finds bipolarity, "a world with only two relatively equal superpowers", to be the most stable system. The two powers watch each other carefully, and neither can make gains without energizing intense balancing behavior by the other. Balanced multipolarity is the second most stable system; though it may be characterized by much diplomatic jockeying for advantage and some war, large-scale war should not occur. The least stable system is unbalanced multipolarity. One power emerges far ahead of the (regional) pack, and will soon be tempted to try to cement its advantage. This leads to hegemonic war (though this is not a term that Mearsheimer uses). Moreover, because other powers are usually tempted to buckpass rather than balance, the aspiring hegemon may be able to knock off some of its rivals one at a time until the survivors realize that only a great coalition can stop its march.

The concept of unbalanced multipolarity is one of Mearsheimer's key theoretical innovations. It is Mearsheimer's story of Europe's great wars. In contrast to defensive realists, Mearsheimer does not believe that there is enough variation in the inherent defensive or offensive possibilities of military technology in any given era to affect significantly the competitive behavior of states. Defensive realists believe that competition among land powers will be muted when the defense has the advantage. Mearsheimer believes that land power is land power, and that the power to defend and the power to attack are pretty much the same. To make this point, he carefully measures the power of states since the French Revolution. He shows convincingly that the great aspiring hegemons, Napoleonic France, Wilhelmine Germany, Nazi Germany and Soviet Russia, were significantly superior to their immediate neighbors when they embarked on their ambitious careers. Moreover, he shows that the young United States embarked on a similar career in North America, and for similar reasons. He notes, however, that the United States is modern history’s only successful regional hegemon; all others failed.

Are there any great powers that just stop? Mearsheimer observes that the insular powers, Britain and the continental United States, have not bid for Eurasian hegemony. The reason, to which he refers repeatedly, is the stopping power of water. The insular powers do not try to establish Eurasian continental hegemony despite the disparity between their power potential and that of the land powers. They do not do so because, in Mearsheimer's view, it is difficult for them to get ashore on continental land masses with a large army. By the same token, it is difficult for any state on the continental land mass to get at them. This is Mearsheimer's sole concession to defensive realism, though he does not quite admit this theoretical point.

The only security concern of the insular power, therefore, is the threat of domination of the Eurasian continental land mass by a single hegemon. This could come about, as balance-of-power theorists have often suggested, if a continental empire could combine the surpluses of many conquered continental states and use it to produce a naval challenge to the insular state. Thus, the insular states buckpass to continental balancers unless it appears likely that an aspiring hegemon might succeed; they then seek to restore the balance by helping the surviving competitors of the continental hegemon to cut it down to size. Afterwards, insular states withdraw from the continent, their faith in the stopping power of water restored. This occurs, ironically, just at the moment that they have overcome that stopping power by projecting massive land forces across the water into the heart of the continent, a wrinkle that Mearsheimer overlooks. The British went home after the Napoleonic Wars and World War I. The United States went home after World War I and seemed on the verge of doing so after World War II, until the risk of Soviet hegemony over Europe caused it to reconsider.

The Problems

Mearsheimer has made a significant contribution to our understanding of the behavior of great powers, especially great land powers. For the most part, their security concerns are not easily satisfied. Their basic tendency is to expand to dominate their immediate environment, meaning, for the most part, the territories around them. They do this for security reasons: one does not need a hyperactive genius like Napoleon, an imaginative but adolescent atavist like Kaiser Wilhelm II, or a monster like Adolf Hitler to get hegemonic wars. One only needs unbalanced multipolarity amid the massive power of the states these men led. Stated baldly this argument is overdrawn, but it serves an important theoretical purpose: to show that all aspiring hegemons have in common a significant military advantage over their neighbors. Mearsheimer successfully demonstrates that the great land powers do bid for hegemony and that “land power is the main instrument for winning wars. He is twice correct, but it would have been helpful had he wrestled theoretically with the fact that the land powers seldom win their great hegemonic wars. Napoleonic France, Wilhelmine Germany, Nazi Germany and the Soviet Union in the Cold War are 0 for 4, and the great sea powers have been central to bringing them down.

Indeed, Mearsheimer's treatment of sea powers is full of unsolved puzzles. He gives vast causal weight to the geographic fact of large bodies of water. In this, he has much in common with the better geo-politicians such as Halford Mackinder and Nicholas Spykman. Britain and the United States have indeed foregone direct efforts to dominate the Eurasian landmass. Even when the temptations must have been high, at the end of great balance-of-power wars that left Britain and the United States on the continent in positions of strength, the stopping power of water overcome, these two have packed up and gone home, or nearly done so. Surely this is more than an accident of geography. It seems a prediction of defensive, not offensive, realism. With the continental hegemons eliminated, the insular powers felt safe enough to leave.

The insular powers are guilty of other kinds of satisficing behavior, too, a fact that also seems better explained by defensive realism. How are we to understand why the British stayed out of the American Civil War? Mearsheimer handles this in a brief and unsatisfying footnote: the British calculated that the Union would win and that intervention was fruitless. But given the duration and scale of the North American bloodletting, it is peculiar that the British would have been so resigned to the notion that they could not defend, much less improve, their power position by helping the South.

Similarly, it is peculiar that the United States, having expanded across North America, drew back from the direct conquest of Canada and Mexico, not to mention other countries in Central and South America. Mearsheimer credits the United States with understanding that by the time the lands encompassing the lower 48 states were under U.S. control, Mexico and Canada were recognized as full-fledged nations and were thus too difficult for the United States to conquer and govern. Such restraint seems remarkably sagacious, indeed un-tragic, given offensive realism's predictions for the behavior of land powers. Presumably, offensive realism would argue in reply that massive power disparities made the United States the hegemon of the Western Hemisphere to such a degree that there was no need to directly conquer neighboring Canada and Mexico.

Given that the United States did indeed establish and defend a kind of hegemony across the entire Western Hemisphere, and that the British established and defended their hegemony across the then-less-developed world, it appears that at least some of the behavior of the insular powers can be explained by offensive realism. Such powers expand their influence wherever others are weak, but they endeavor to do so without direct occupation by large armies. Water did not stop the United States or Britain; rather, it was the medium by which they expanded. The stopping power of water that Mearsheimer observes was in truth an artifact of the presence of several organized great powers on the Eurasian landmass. Because they had to defend themselves from each other with armies, they had limited resources left for navies. Insular powers preferred to project their own defense outward with navies; it was eminently reasonable for them to want to fight at sea, away from their homelands, rather than on their own shores. Insular powers were free to put the bulk of their resources into navies and were rewarded for doing so by no longer having to worry about invasion.

Clearly, then, major sea powers strove for much more than regional naval superiority - they built navies of global reach, and sought a margin of superiority over any, and sometimes all, possible challengers. Having done so, they had a potent instrument of power projection at hand, and they used it to expand their power where others were weak or disorganized. Given their continental security problems, the land powers would have had a difficult time building a navy sufficient to challenge an insular power endowed with roughly equivalent resources, a problem the Wilhelmines tried and failed to solve. Likewise, the insular powers would have had a difficult time building an army to challenge the great land powers, but they still managed to challenge the not-so-great.

The case of Imperial Japan, however, undermines Mearsheimer's argument that the stopping power of water predicts continental restraint for insular powers. Japan was an insular power, yet starting at the end of the 19th century it struggled to dominate the peninsula and islands in its immediate neighborhood. Not many decades into the 20th century Japan tried to conquer the proximate Eurasian land mass. Japan was undeterred by the stopping power of water, and also unreassured by it. Mearsheimer treats the facts of the Japanese case extensively, but devotes only a single paragraph to its status as a theoretical anomaly. Japan expanded because there were no great powers in Northeast Asia to prevent it from so doing. Opportunity mattered much more to its behavior than did the defensive barrier provided by nature. Japan succumbed to the temptation to expand its power by asserting control over weaker players, as did the United States and the British before it. The United States and the British relied more on strategies of indirect influence, though they did not forego direct military intervention and control when it was necessary or feasible. Even when British and U.S. intervention was direct, these two states seem to have been adept at local tactics of divide, co-opt and conquer. Japan relied on direct military conquest and stern repression. Perhaps the power disparity between Japan and these weaker states was too small to permit less heavy-handed tactics.

In contrast to what Mearsheimer asserts, his theory of offensive realism predicts that the sea powers should have the same general expansive orientation as the land powers. Japan is the clearest and most brutal example of land power-like behavior, but instead of playing it up theoretically, Mearsheimer plays it down. The tactics and strategies of the two great insular powers, Britain and the United States, have usually differed from those of the great land powers, it is true, but their expansive orientation could just as easily have been stressed. Overall, Mearsheimer would have been wise to take up this point and offer an explanation for how and why the expansive strategies of land and sea powers differ, and what kinds of success each kind of power permits. Such an explanation, even if imperfect, would have sat better with the theory of offensive realism than the somewhat mechanical invocation of “the stopping power of water.

The Future

Mearsheimer's theory of offensive realism explains even more about international politics than he himself suggests. It predicts that all great powers will be grabby land powers and sea powers. Defensive realists predict too much cooperation. Anarchy does provide strong incentives for states to improve their power position, and they will attempt to do so with greater energy than would generally make sense to the disinterested observer. Nevertheless, Mearsheimer's account of the stopping power of water and its influence on the insular powers does suggest that physical facts may make a state feel relatively safe, with the result that it will compete with less energy, maybe less brutality, and perhaps with less self-destructiveness than will the land powers. If both observations are correct, then perhaps the divide between offensive and defensive realists is not really so great. Anarchy invites states to compete intensively for power for security reasons, and they do. The purpose of power is security, and sometimes states achieve enough security to temper, though not terminate, their pursuit of more power.

On the basis of his historical account of the influence of the stopping power of water, Mearsheimer predicts that the United States will soon come home. It will return to its historical "off-shore balancer" role. But holding geography out of the equation, the world today is quite similar to his system of unbalanced multipolarity. A consistent theory of offensive realism should predict that once the international system becomes one of unbalanced multipolarity, even a disproportionately powerful insular state should succumb to the temptation to try to dominate the entire system, and knock potential peer competitors out of the game. Mearsheimer predicts otherwise: the permanent defensive advantage offered by the ocean barrier will somehow always bring even the greatest of sea powers back home. Thus his prediction that the United States will soon come home, though this outcome is nowhere in sight.

How unbalanced is the current international system? Very much so. If Mearsheimer's definition of power would allow us simply to count GDPs, he might agree. One has to treat the EU as a unitary great power to find an economy and population roughly equal to the United States; otherwise it takes more than four nation-states in combination to approximate America's economic prowess. Mearsheimer's way of measuring power demands that we also assess current ground force strengths, and he would note that the U.S. Army and Marines are small by the historical standard of continental great powers. But even Mearsheimer measures land power by the combination of ground forces and their supporting air and naval forces. U.S. air and naval forces now provide an historically unprecedented augmentation of American ground power. We do not know how much, because there has not yet been a good test (not even in Afghanistan), but most agree that the augmentation is large. It could be larger still if the U.S. Air Force were more interested in the land battle.

It is also worth noting that though a great sea power could formerly hope to command only one global commonsâ€"the seaâ€"in many parts of the world the United States now commands the sky above 10,000 feet. Although the United States does not now truly command space (it probably cannot immediately deny space to others and simultaneously fully protect American assets), it uses space for many more military purposes than does any other power, which provides the United States with an extraordinary intelligence, communications and navigation advantage. If it came to it, the United States would probably win an extended armed struggle for command of space. The United States has gone beyond command of the sea to command of the planetary commons. Counting economic power, and military power properly measured, the current distribution of power globally is probably as unbalanced as it was when it took more than two decades, six coalitions, and the ultimate military cooperation of Britain, Austria, Prussia and Russia to bring down Napoleon.

The United States is, however, an insular power protected by a large body of water. Mearsheimer and most of his defensive realist foils are right to predict that the United States is unlikely to embark on a campaign of continental conquest to consolidate its global hegemony. The United States will probably not be tempted to risk all on a single throw of the iron dice, because the loss of its world hegemonic position would but incrementally reduce its security. The oceans do provide an extra margin of defensive safety. The U.S. nuclear arsenal makes a direct great power challenge to the United States exceedingly difficult, even if the United States were no longer the global hegemon, and makes equally self-destructive a direct U.S. military challenge to another power with sufficient nuclear weapons. Indeed, in a somewhat murky discussion, even Mearsheimer concedes that states with secure nuclear retaliatory forces are very unlikely to attempt the conquest of their nuclear peers. Nuclear deterrence will likely moderate great power competition, making old-style hegemonic wars unlikely.

Nevertheless, since the Cold War ended, the United States has not only held its positions across the globe, but is trying energetically to expand its power and influence. As the last Defense Guidance of the elder Bush Administration recommended, it has been U.S. policy to try to forestall the rise of peer competitors. The United States has not brought its legions home as it did after World War I and nearly did after World War II, or as the British did after the Napoleonic Wars and after World War I. The post-Cold War debate on foreign policy between Democrats and Republicans has not been fought out over the issue of whether the United States should have an activist foreign policy: the two parties could only disagree on how much military management of the internal politics of strategically insignificant failed states the United States should undertake. The two parties agreed that the U.S. military's Unified and Specified Commands and Commanders should continue to organize the entire globe for war; that NATO should expand; that the U.S. military should be substantially more capable than that of any other state or combination of states; and even that the U.S. military should be forward deployed, still a quarter million troops as of June 2001. They have also agreed, more or less, that the United States should preserve some degree of nuclear superiority over other nuclear powers, and that it would be better if no other nuclear powers emerged; both policies that Mearsheimer would predict. And the terrorist attacks of September 11, 2001, which portend a significant increase in the cost of hegemony to the United States, seem to have injected still more energy and commitment into an expansive U.S. foreign policy, not less.

I happen to view U.S. foreign policy activism since the collapse of the Soviet Union to be excessive relative to U.S. security requirements. So, I think, for most practical purposes, does John Mearsheimer. But it is peculiar that Mearsheimer should find this policy of U.S. primacy at odds with his theory. After all, his theory, offensive realism, predicts it.

Barry Posen is a professor of political science and a member of the Security Studies Program at the Massachusetts Institute of Technology. He is most recently the author of The Struggle against Terrorism: Grand Strategy, Strategy, and Tactics, in the Winter issue of International Security.