The Age of Forever Wars

Why military strategy no longer delivers victory

I wrote this piece for the May/June 2025 edition of Foreign Affairs. I am very grateful to the journal for kindly giving me permission to distribute it to subscribers.

---

Coalition forces in Bagramm Afghanistan (2002)

In Operation Desert Storm, the 1991 campaign to liberate Kuwait from Iraqi occupation, the United States and its coalition allies unleashed massive land, air, and sea power. It was over in a matter of weeks. The contrast between the United States’ grueling and unsuccessful war in Vietnam and the Soviet Union’s in Afghanistan could not have been more stark, and the speedy victory even led to talk of a new era of warfare—a so-called revolution in military affairs. From now on, the theory went, enemies would be defeated through speed and maneuver, with real-time intelligence provided by smart sensors guiding immediate attacks using smart weapons.

Those hopes proved short-lived. The West’s counterinsurgency campaigns of the early decades of this century, which came to be labeled “forever wars,” were not notable for their rapidity. Washington’s military campaign in Afghanistan was the longest in U.S. history, and in the end it was unsuccessful: despite being pushed out at the start of the U.S. invasion, the Taliban eventually came back.

Nor is this problem limited to the United States and its allies. In February 2022, Russia launched a full-scale invasion of Ukraine that was supposed to overrun the country in a matter of days. Now, even if a cease-fire can be reached, the war will have lasted for more than three years, during which it was dominated by grinding, attritional fighting rather than bold and audacious actions. Similarly, when Israel launched its invasion of Gaza in retaliation for Hamas’s October 7, 2023, assault and hostage taking, U.S. President Joe Biden urged that the Israeli operation should be “swift, decisive, and overwhelming.” Instead, it continued for 15 months, in the process expanding to other fronts in Lebanon, Syria, and Yemen, before a fragile cease-fire was reached in January 2025. By mid-March, the war had reignited. And this leaves out numerous conflicts in Africa, including in Sudan and the Sahel, that have no end in sight.

The idea that surprise offensives could produce decisive victories began to be embedded in military thinking in the nineteenth century. But again and again, forces that undertake them have shown how difficult it is to bring a war to an early and satisfactory conclusion. European military leaders were confident that the war that began in the summer of 1914 could be “over by Christmas”—a phrase that is still invoked whenever generals sound too optimistic; instead, the fighting would last until November 1918, concluding with fast offensives but only after years of devastating trench warfare along almost static frontlines.

In 1940, Germany overran much of western Europe in a matter of weeks by means of a blitzkrieg, bringing together armor and airpower. But it could not finish the job, and after initial rapid advances against the Soviet Union in 1941, it was drawn into a brutal war with enormous casualties on both sides that would only end nearly four years later with the total collapse of the Third Reich. Similarly, the decision by Japan’s military leadership to launch a surprise attack on the United States in December 1941 ended in the catastrophic defeat of the Japanese empire in August 1945. In both world wars, the key to victory was not so much military prowess as unbeatable stamina.

Yet despite this long history of protracted conflict, military strategists continue to shape their thinking around short wars, in which all is supposed to be decided in the first days, or even hours, of combat. According to this model, strategies can still be devised that will leave the enemy surprised by the speed, direction, and ruthlessness of the initial attack. With the constant possibility that the United States could be drawn into a war with China over Taiwan, the viability of such strategies has become a pressing issue: Can China quickly seize the island, using lightning force, or will Taiwan, supported by the United States, be able to stop such an attack in its tracks?

What is clear is that amid rising tensions between the United States and a variety of antagonists, there is a critical misalignment in defense planning. In recognition of the tendency of wars to drag on, some strategists have begun to warn about the dangers of falling into the “short war” fallacy. By emphasizing short wars, strategists rely too much on initial battle plans that may not play out in practice—with bitter consequences. Andrew Krepinevich has argued that a protracted U.S. war with China would “involve kinds of warfare with which the belligerents have little experience” and that it could pose “the decisive military test of our time.” Moreover, failure to prepare for long wars creates vulnerabilities of its own. To transition from a short war to a protracted one, countries must impose different demands on their military and on society as a whole. They also will need to reappraise their objectives and what they are prepared to commit to achieve them.

Once military planners accept that any major contemporary war might not end quickly, they are required to adopt a different mindset. Short wars are fought with whatever resources are available at the time; long wars require the development of capabilities that are geared to changing operational imperatives, as demonstrated by the continual transformation of drone warfare in Ukraine. Short wars may present only temporary disruptions to a country’s economy and society and do not require extensive supply lines; long wars demand strategies for maintaining popular support, functioning economies, and secure ways to rearm, restock, and replenish troops. Long wars also require constant adaptation and evolution: the longer a conflict lasts, the more pressure there is for innovations in tactics and technologies that might yield a breakthrough. Even for a great power, failure to prepare for and then rise to meet these challenges could be disastrous.

Yet it is also fair to ask how realistic it is to plan for wars that do not have a clear endpoint. It is one thing to sustain a protracted counterinsurgency campaign but quite another to prepare for a conflict that would involve continuing and substantial losses of people, equipment, and ammunition over an extended period. For defense strategists, there may also be significant obstacles to this kind of planning: the militaries they serve may lack the resources to prepare for a long war. The answer to this dilemma is not to prepare for wars of indefinite duration but to develop theories of victory that are realistic in their political objectives and flexible in how they might be achieved.

The short war fallacy

The advantages of short wars—immediate success at a tolerable cost—are so obvious that no case can be made for knowingly embarking on a long one. By contrast, even admitting the possibility that a war could become protracted may seem to betray doubts about the ability of one’s military to triumph over an adversary. If strategists have little or no confidence that a prospective war can be kept short, then arguably the only prudent policy is not to fight it at all. Still, for a country such as the United States, it might not be possible to rule out a conflict with another great power of similar strength, even if rapid victory is not assured. Although Western leaders have an understandable aversion to intervening in civil wars, it is also possible that the actions of a nonstate adversary could become so persistent and harmful that direct action to deal with the threat becomes imperative, regardless of how long that may take.

This is why military strategists continue to shape their plans around short wars, even when a protracted conflict cannot be excluded. During the Cold War, the main reason the two sides did not devote extensive resources to preparing for a long war was the assumption that nuclear weapons would be used sooner rather than later. In the current era, that threat remains. But the prospect of a great-power conflict turning into something like the cataclysmic world wars of the last century is frightening—adding urgency to plans that are designed to produce a quick victory with conventional forces.

Strategies for carrying out this ideal type of war are geared above all toward moving fast, with some element of surprise and with sufficient force, to overwhelm enemies before they can mount an adequate response. New warfighting technologies tend to be assessed according to how much they might help achieve rapid battlefield success rather than how well they might help secure a durable peace. Take artificial intelligence. By harnessing AI, the thinking goes, militaries will be able to assess battlefield situations, identify options, and then choose and implement those options in a matter of seconds. Vital decisions may soon be made so fast that those in charge, let alone the enemy, will barely appreciate what is happening.

So ingrained is the fixation with speed that generations of U.S. military commanders have learned to shudder at the mention of attritional warfare, embracing decisive maneuver as the route to quick victories. Long slogs of the sort now taking place in Ukraine—where both sides seek to degrade each other’s capabilities, and progress is measured by body counts, destroyed equipment, and depleted stocks of ammunition—are not only dispiriting to the belligerent countries but also hugely time-consuming and expensive. In Ukraine, both sides have already expended extraordinary resources, and neither is close to anything that resembles a victory. Not all wars are conducted at such a high intensity as the Russian-Ukrainian war, but even prolonged irregular warfare can take a heavy toll, resulting in a growing sense of futility in addition to mounting costs.

Although it is known that audacious surprise attacks often deliver far less than promised and that it is much easier to start wars than to end them, strategists still worry that potential enemies may be more confident in their own plans for rapid victory and will act accordingly. This means that they are required to concentrate on the likely opening phase of war. It may be assumed, for example, that China has a strategy for taking Taiwan that aims to catch the United States unprepared, leaving Washington to respond in ways that either have no hope of success or are likely to make matters much worse.

To anticipate such a surprise attack, U.S. strategists have devoted much time to assessing how the United States and other allies can help Taiwan thwart China’s opening moves—as Ukraine did with Russia in February 2022—and then make it hard for China to sustain a complex operation some distance from the mainland. But even this scenario could easily lead to protraction: if the first countermoves by Taiwanese forces and their Western allies are successful, and China gets bogged down but does not withdraw, Taiwan and the United States would still face the problem of coping with a situation in which Chinese forces have a presence on the island. As Ukraine has learned, it is possible to get stuck in a protracted war because an incautious adversary has miscalculated the risks.

This is not to say that modern armed conflicts never end in quick victories. In June 1967, it took Israel less than a week to decisively vanquish a coalition of Arab states in the Six-Day War; three years later, when India intervened in the Bangladesh war for independence, it took Indian forces just 13 days to defeat Pakistan. The United Kingdom’s 1982 victory over Argentina in the Falklands War unfolded fairly quickly. But since the end of the Cold War, there have been many more wars in which early successes faltered, lost momentum, or didn’t quite achieve enough, transforming the conflicts into something far more intractable.

Indeed, for some kinds of belligerents, the pervasive problem of long wars may provide an important advantage. Insurgents, terrorists, rebels, and secessionists may embark on their campaigns knowing that it will take time to undermine established power structures and assuming that they will simply outlast their more powerful enemies. A group that knows it is unlikely to triumph in a rapid confrontation may recognize that it has greater chances of success in a long and arduous struggle, as the enemy is worn down and loses morale. Thus, in the last century, anticolonial movements, and more recently, jihadist groups, embarked on decades-long wars not because of poor strategy but because they had no other choice. Especially when confronted by a military intervention from a powerful foreign army, the best option for such organizations is often to let the enemy tire of an inconclusive fight and then return when the time is right, as the Taliban have done in Afghanistan.

By contrast, great powers tend to assume that their significant military superiority will quickly overwhelm opponents. This overconfidence means that they fail to appreciate the limits of military power and so set objectives that can be achieved, if at all, only through a prolonged struggle. A larger problem is that by emphasizing immediate battlefield results, they may neglect the broader elements necessary for success, such as achieving the conditions for a durable peace, or effectively managing an occupied country in which a hostile regime has been toppled but a legitimate government has yet to be installed. In practice, therefore, the challenge is not simply planning for long wars rather than short ones but planning for wars that have a workable theory of victory with realistic objectives, however long they may take to realize.

Not losing is winning

Effective warfighting strategy is a matter of not just military method but also political purpose. Evidently, military moves are more successful when combined with limited political ambition. The 1991 Gulf War succeeded because the George H. W. Bush administration aimed only to expel Iraq from Kuwait and not to overthrow Iraqi dictator Saddam Hussein. Russia’s 2022 invasion of Ukraine might have had more success if it had concentrated on the Donbas rather than trying to take political control of the entire country.

With limited ambition, it is also easier to compromise. A workable theory of victory requires a strategy in which military and political objectives are aligned. It may be that the only way to resolve a dispute is through the total defeat of the enemy, in which case sufficient resources must be allocated to the task. At other times a military initiative may be taken in the firm expectation that it will lead to early negotiations. That was Argentina’s view in April 1982 when it seized the Falkland Islands. When Egyptian President Anwar al-Sadat ordered his armed forces to cross the Suez Canal in October 1973, he did so to create the conditions for direct talks with Israel. His armed forces were pushed back, but he got his political wish.

Underestimating the enemy’s political as well as military resources is one of the main reasons that short-war strategies fail. Argentina assumed that the United Kingdom would accept a fait accompli when it seized the Falklands and did not imagine that the British would send a task force to liberate the islands. Wars are often launched in the misguided belief that the population of the opposing power will soon buckle under an attack. Invaders may assume that a section of the population will embrace them, as could be seen in Iraq’s invasion of Iran in 1980 and, for that matter, in Iran’s counterinvasion of Iraq. Russia based its full-scale attack on Ukraine on a similar misreading: it assumed there was a beleaguered minority—in this case, Russian speakers—who would welcome its forces; that the government in Kyiv lacked legitimacy and could easily be toppled; and that the West’s promises of support to Ukraine would not amount to much. None of these assumptions survived the first days of the war.

When a short-war plan does not produce the anticipated victory, the challenge for military leaders is to achieve a new alignment between means and ends. By September 2022, President Vladimir Putin realized that Russia risked a humiliating defeat unless it could bring more soldiers to the front and put its economy on a comprehensive war footing. As the leader of an authoritarian state, Putin could quash domestic opposition and keep control of the media and did not have to worry too much about public opinion. Nonetheless, he needed a new narrative. Having asserted before the war that Ukraine was not a real country and that its “neo-Nazi” leaders had seized power through a coup in 2014, he could not explain why the country failed to collapse when hit by a superior Russian force. So Putin changed his story: Ukraine, he alleged, was being used by NATO countries, in particular the United States and the United Kingdom, to pursue their own Russophobic objectives.

Despite having initially presented the invasion as a limited “special military operation,” the Kremlin now portrayed it as an existential struggle. This meant that instead of simply stopping Ukraine from being so troublesome, Russia now sought to demonstrate to NATO countries that it could not be broken by economic sanctions or the alliance’s weapons supplies to Ukraine. By describing the war as defensive, the Russian government was telling its people how much was at stake while also warning that they could not now expect a quick victory. Instead of scaling back its objectives to acknowledge the difficulties of defeating the Ukrainians in battle, the Kremlin scaled them up to justify the extra effort. By annexing four Ukrainian provinces in addition to Crimea, and by continuing to demand a supine government in Kyiv, Russia has made the war tougher, not easier, to end. This situation illustrates the difficulty of ending wars that are not going well: the possibility of failure often adds a political objective—the desire to avoid the appearance of weakness and incompetence. Reputational concerns were one reason why the U.S. government hung on in Vietnam long after it was clear that victory was out of reach.

Replacing a failed theory of victory with one that is more promising requires not only reappraising the enemy’s actual strengths but also recognizing the flaws in the political assumptions that underlay the opening moves. Suppose that U.S. President Donald Trump’s push for a cease-fire bears fruit, leaving the war frozen along current frontlines. Moscow could portray its territorial gains as a success of sorts, but it could not truly claim victory as long as Ukraine has a functioning independent and pro-Western government. If Ukraine temporarily accepted its territorial losses but could still build up its forces and obtain some form of security guarantees with the help of its Western partners, the outcome would still be a far cry from Russia’s oft-stated demand for a demilitarized neutral Ukraine. Russia would be left administering and subsidizing wrecked territory with a resentful population while having to defend the long cease-fire lines.

Yet although Russia has not been able to win the war, so far it has not lost. It has been forced to withdraw from some territory conquered early in the war, but since late 2023 it has made slow but continued gains in the east. On the other hand, Ukraine has also not lost, for it has successfully resisted Russian attempts at subjugation and has forced Russia to pay a heavy price for every square mile taken. Most important, it remains a functioning state.

No end in sight

In commentary on contemporary warfare, the distinction between “winning” and “not losing” is vital yet hard to grasp. The difference is not intuitive because of the assumption that there will always be a victor in war and because, at any time, one side can appear to be winning even if it has not actually won. The situation of “not losing” is not quite captured by terms such as stalemate and deadlock since these imply little military movement. Both sides can be “not losing” when neither can impose a victory on the other, even if one or both are on occasion able to improve their positions. This is why proposals to end protracted wars normally take the form of calls for a cease-fire.

The problem with cease-fires, however, is that the parties to the conflict tend to regard them as no more than pauses in the fighting. They may have little effect on the underlying disputes and may simply offer both sides the opportunity to recover and reconstitute for the next round. The cease-fire that ended the Korean War in 1953 has lasted for over 70 years, but the conflict remains unresolved and both sides continue to prepare for a future war.

Most models of warfare continue to assume the interaction between two regular armed forces. According to this framing, a decisive military victory comes when the enemy’s forces can no longer function, and such an outcome should then translate into a political victory, as well, since the defeated side has little choice but to accept the victor’s terms. After years of tension and intermittent fighting, one side may get into a position in which it can claim an unequivocal victory. One example is Azerbaijan’s offensive in Nagorno-Karabakh in 2023, possibly ending a three-decade war with Armenia.

Alternatively, even if a country’s armed forces are still largely intact, pressures may build up on its government to find a way out of the conflict because of the cumulative human and economic costs. Or there may be no prospect of a true victory, as Serbia came to recognize in its war against NATO in Kosovo in 1999. When one of the parties to a conflict experiences regime change at home, that can also lead to the abrupt end of hostilities. When they do end, however, long wars are likely to leave legacies that are bitter and lasting.

Even in cases in which a political settlement, and not just a cease-fire, can be reached, a conflict may not be resolved. Territorial adjustments, and perhaps substantial economic and political concessions by the losing side, may produce resentment and a desire for redress among the defeated population. A defeated country may remain determined to find ways to recover what it has lost. This was France’s position after forfeiting Alsace-Lorraine to Germany in 1871 after the Franco-Prussian War. In the Falklands War, Argentina claimed to be recovering territory it had lost a century and a half earlier. Moreover, for the victor, enemy territory that has been taken and annexed will still need to be governed and policed. If the population cannot be subdued, what may initially appear as a successful land grab may end up a volatile situation of terrorism and insurgency.

In contrast to standard models of war, in which hostilities usually have a clear starting point and an equally clear end date, contemporary conflicts often have blurred edges. They tend to pass through stages, which can include war and periods of relative calm. Take the United States’ conflict with Iraq. In 1991, Iraqi forces were quickly defeated by a U.S.-led coalition, in what was ostensibly a short, decisive war. But because the United States decided not to occupy the country, the war left Saddam in charge, and his continuing defiance created a sense of unfinished business. In 2003, under President George W. Bush, the United States reinvaded Iraq and achieved another speedy victory, and this time Saddam’s Baathist dictatorship was toppled. But the process of replacing it with something new precipitated years of devastating intercommunal violence that at times approached full-blown civil war. Some of that instability has continued to this day.

Because civil wars and counterinsurgency operations are fought in and among populations, civilians bear the brunt of the harm from these wars, not only by being caught up in deliberate sectarian violence or crossfire but also because they are forced to flee their homes. This is one reason why these wars tend to lead to prolonged conflict and chaos. Even when an intervening power decides to walk away, as both the Soviet Union and, much later, the U.S.-led coalition did in Afghanistan, it does not mean that conflict ends—only that it takes on new forms.

In 2001, the United States had a clear “short war” plan for overthrowing the Taliban, which it implemented successfully and relatively efficiently using regular forces combined with the Afghan-led Northern Alliance. But there was no clear strategy for the next stage. The problems Washington faced were caused not by a stubborn opponent fighting with regular forces but by endemic violence, in which the threats were irregular and emerged out of civil society and in which any satisfactory outcome depended on the elusive goals of bringing decent governance and security to the population. Without external forces to prop up the government, the Taliban was able to return, and Afghanistan’s history of conflict continued.

Israel’s triumph in 1967—a paradigmatic case of quick victory—also left it occupying a large territory with resentful populations. It created the conditions for many wars that followed, including the Middle East wars that erupted with Hamas’s October 7, 2023, attacks. Since then, Israel has fought campaigns against the group in the Gaza strip, from which Israel had withdrawn in 2005, and against Hezbollah in Lebanon, where Israel had fought a mismanaged operation in 1982. The two campaigns have taken similar forms, combining ground operations to destroy enemy facilities, including tunnel networks, with strikes against weapons stocks, rocket launchers, and enemy commanders.

Both conflicts have caused huge numbers of civilian casualties and widespread destruction of civilian areas and infrastructure. Yet Lebanon could be considered a success because Hezbollah agreed to a cease-fire while the war in Gaza was still underway, which is something it had said it would refuse to do. By contrast, the short-lived cease-fire in Gaza was not a victory, because the Israeli government had set as its objective the complete elimination of Hamas, which it did not achieve. In March, after a breakdown of negotiations, Israel resumed the war, still without a clear strategy to bring the conflict to a definitive end. Although severely depleted, Hamas continues to function, and without an agreed plan for the future governance of Gaza or a viable Palestinian alternative, it will remain an influential movement.

In Africa, protracted conflicts appear endemic. Here the best predictor of future violence is past violence. Across the continent, civil wars flare and then abate. These often reflect deep ethnic and social cleavages, aggravated by external interventions, as well as cruder forms of power struggle. The underlying instability ensures constant conflict in which individuals and groups can have a stake, perhaps because the fighting provides both a stimulus to and a cover for trafficking in arms, people, and illicit goods. The current war in Sudan involves civil strife and shifting allegiances, in which one oppressive regime was toppled by a coalition, which then turned in on itself, leading to an even more vicious war. It also involves external actors such as Egypt and the United Arab Emirates, which are more concerned with preventing opponents from gaining an advantage than with ending the violence and creating the conditions for recovery and reconstruction.

Proving the rule, cease-fires and peace treaties, when they do occur, often turn out to be short-lived. Sudanese parties have signed more than 46 peace treaties since the country achieved independence in 1956. Wars tend to be identified when they boil over into direct military confrontation, but the pre- and postwar simmering is part of the same process. Rather than discrete events with a beginning, a middle, and an end, wars might be better understood as the result of poor and dysfunctional political relations that are difficult to manage by nonviolent means.

A different kind of deterrent

The main lesson the United States and its allies can draw from their considerable experience of lengthy wars is that they are best avoided. Should the United States become involved in a protracted great-power conflict, the country’s whole economy and society will need to be put on a war footing. Even if such a war ends with something approximating a victory, the population would likely be shattered and the state drained of all spare capacity. Moreover, given the intensity of contemporary warfare, the speed of attrition, and the costs of modern weaponry, ramping up investment in new equipment and ammunition might still be insufficient to sustain a future war for long. At a minimum, the United States and its partners would need to procure sufficient stocks in advance to stay in the fight long enough for a much more drastic, full-scale mobilization to be set in motion.

And then, of course, there is the risk of nuclear war. At some point during a protracted war involving either Russia or China, the temptation to use nuclear weapons might prove irresistible. Such a scenario would probably bring a long conventional war to an abrupt conclusion. After seven decades of debate about nuclear strategy, a credible theory of nuclear victory over an adversary able to retaliate in kind has yet to be found. As with conventional war strategists, nuclear planners have focused on speed and brilliantly executed opening moves, with the aim of taking out the enemy’s means of retaliation and eliminating its leadership, or at least alarming and confusing it to generate a paralysis of indecision.

All such theories, however, have appeared to be unreliable and speculative since any first strikes would have to contend with the risk of an enemy launch on warning as well as sufficient systems surviving for a devastating riposte. Fortunately, these theories have never been tested in practice. A nuclear offensive that does not produce immediate victory and instead results in more nuclear exchanges might not be protracted, but it would undoubtedly be bleak. This is why the condition has been described as one of “mutually assured destruction.”

It is worth recalling that one reason the U.S. defense establishment embraced the nuclear age so enthusiastically was that it offered an alternative to the devastating world wars of the early twentieth century. Strategists were already keenly aware that fights to the finish between great powers could be exceptionally long, bloody, and costly. As with nuclear deterrence, however, great powers may now need to prepare more conspicuously for longer conventional wars than current plans assume—if only to help ensure that they don’t happen. And as the war in Ukraine has painfully shown, great powers can be implicated in long wars even when they are not directly involved in the fighting. The United States and its allies will need to improve their defense industrial bases and build stocks to better prepare for these contingencies in the future.

The conceptual challenge this kind of preparation poses, however, is different from what would be required to prepare for a titanic confrontation between superpowers. Although the prospect may be unpalatable, military planners need to think about managing a conflict that risks protraction in the same way that they have thought about managing nuclear escalation. By preparing for protraction and reducing any potential aggressor’s confidence in being able to wage a successful short war, defense strategists could provide another kind of deterrent: they would be warning adversaries that any victory, even if it could be achieved, would come with an unacceptably high cost to their military, economy, and society.

Wars start and end through political decisions. The political decision to initiate armed conflict is likely to assume a short war; the political decision to bring the fighting to an end will likely reflect the inescapable costs and consequences of a long war. For any military power, the prospect of drawn-out or unending hostilities and significant economic and political instability is a good reason to hesitate before embarking on a major war and to seek other means to achieve desired goals. But it also means that when wars cannot be avoided, their military and political objectives must be realistic and attainable and set in ways that can be achieved by the military resources available. One of the great allures of military power is that it promises to bring conflicts to a quick and decisive conclusion. In practice, it rarely does.