Seasoned NATO heads will snicker at the idea that squabbles over defense spending are new. As long ago as 1970, then U.S. ambassador to NATO Harlan Cleveland described the alliance as “an organized controversy about who is going to do how much.” Since NATO’s inception, the United States has carried most of the organization’s burden. As historian Lawrence Kaplan wrote, the alliance at its birth in 1949 was a one-sided affair, with Washington insisting on European contributions not because it expected anything in return but mainly as a sweetener to induce a skeptical U.S. Senate to abandon a tradition of nonentanglement in Europe.

Given that history, it may be tempting to dismiss President Donald Trump’s on-and-off admonishments of NATO as inconsequential. But the alliance is not immune to changing political moods; it can and will wither when it stops serving its members’ interests. So a periodic reassessment is in order—particularly when a presence as disruptive as the new U.S. president arrives on the scene. European allies should respond to Trump in two ways: in the short term, focus on preventing the president from abandoning the alliance and, in the long term, prepare to assume a bigger role in defending the European continent.

Enter Trump

To understand Trump’s impact, one needs to begin with what makes NATO tick. At its heart, the alliance is a compact between the United States and Europe to provide security services to each other. The nature of these services has changed over time: At first, Washington expected Europe to simply help the United States rebuild the continent, but then it looked to NATO allies to help defend against the Soviets and then to help manage crises from Kosovo to Afghanistan. European expectations of the United States have evolved similarly.

But three beliefs have not changed: that insecurity is best handled through active engagement, that engagement works best with allies on one’s side, and that there are no allies better than those in NATO. As long as these beliefs held true, it mattered little that the Soviet threat disappeared or that the Yugoslav Wars broke out in the 1990s: the allies simply adapted NATO to deal with each successive new challenge.

This could be changing, however. Trump has challenged the idea that active engagement in Europe is a core U.S. interest. He appears to regard all foreign relations as zero-sum transactions, in which each contribution to someone else’s security represents a net loss to the United States. That is a dramatically different view from anything NATO allies have seen before. As U.S. academic Walter Russell Mead pointed out, one would have to go back long before the alliance was born—to the early years of the Franklin Roosevelt administration—for a precedent. NATO has had its share of disagreements, such as over the 2003 U.S.-led war in Iraq. But these arguments revolved around ways to manage threats. These were passionate and divisive fights, but they took place in the context of a shared preference for engagement. Trump does not appear to believe in such an approach.

U.S. allies likely draw comfort from the fact that as the administration has settled into power, the zero-sum attitude has partly given way to a more traditional U.S. foreign policy. This is mostly due to the counsel of the secretary of defense, the national security adviser, and most other senior appointees since the early days of the Trump administration. They understand that a policy of global engagement is not a charity act but that it serves broader U.S. interests. They have forestalled a break in U.S. policy by converting the president’s dislike of what alliances represent into a dissatisfaction with what the North Atlantic alliance does and by channeling this sentiment into a set of familiar demands easily recognizable by allies, such as higher defense spending or a greater focus on terrorism. Under their auspices, U.S. troops have continued to arrive in Central Europe to bolster NATO’s ability to deter Russia, a move that appears inconsistent with Trump’s desire to lessen U.S. obligations abroad.

The trouble with relying on senior U.S. officials to maintain the transatlantic defense relationship is that, in times of crisis, they can be overruled by the president. U.S. decisionmaking is highly centralized, and presidents have on previous occasions ignored their senior staff. In 2013, then president Barack Obama chose to disregard advice to launch strikes against Syria after the government of President Bashar al-Assad had dropped sarin gas on the suburbs of Damascus. One assumes that the national security establishment would, in a crisis, urge Trump to honor U.S. obligations to its allies. Whether a president who regards alliances as a burden would order U.S. troops into action is anyone’s guess.

What Is Europe to Do?

Europe’s answer to Trump depends on whether the president’s aversion to NATO is an anomaly or a sign of lasting change in U.S. attitudes toward Europe. If it is the former, the right response is to avoid giving the White House a pretext to break with NATO and await the next president. This, broadly speaking, is the policy of most European allies. The latter possibility—an enduring change lasting beyond the current U.S. president—has not been considered seriously enough by European governments.

However, it should be—the United States has been cooling on NATO for a number of years, since long before Trump’s election. Former U.S. secretary of defense Robert Gates warned in his 2011 farewell speech of a generational change afoot in the United States: “Future U.S. political leaders—those for whom the Cold War was not the formative experience that it was for me—may not consider the return on America’s investment in NATO worth the cost.” The current secretary of defense appears to agree. In February 2017, in an apparent reference to Trump, James Mattis told his NATO peers that “the impatience Secretary Gates predicted is now a governmental reality.”

These views are not necessarily reflected in U.S. public opinion polls, which have remained broadly supportive of alliances. But what Gates and Mattis seem to be saying is that those in Washington tasked with making policy find it increasingly difficult to defend the policy of engagement in Europe. Perhaps they also find it increasingly hard to convince themselves. After all, as the historian Robert Kagan observed in an interview, “What the U.S. has done over the past seventy years is unique. No one, not even the Roman Empire, has tried to sustain a global order. It is not a normal or natural state of things.”

The challenge before European allies therefore goes beyond a president who thinks poorly of alliances (and who may eventually be succeeded by a more pro-engagement figure). Whoever comes next will face the increasingly difficult task of explaining to U.S. audiences why Europe’s stability matters to them. This trend could yet change. A number of factors—a crisis in Europe that grips Americans’ imagination, an articulate pro-European leader in Washington, a crisis in the United States that the European allies help resolve—could revive America’s flagging interest in the alliance it created nearly seventy years ago. But for now, the passage of time and memories work against NATO.

A Two-Step Response to Trump

A policy that seeks to maximize European security would therefore consist of two strands. The first is a concerted attempt to avert a near-term crisis by denying Trump a reason to walk away from U.S. obligations. To this end, the allies should continue to address his criticisms of the alliance, as they have been for the past several months. The second strand is to reflect on what to do if U.S. interest in the alliance wanes to the point that Washington’s commitment to European defense can no longer be taken for granted.

Trump has named two things that he expects from allies: that they spend at least 2 percent of GDP on defense and that they do more to counter terrorism. The first demand is not new. Even before the 2016 U.S. presidential election, NATO countries, spurred by Russia’s aggression against Ukraine, had agreed to aim to reach 2 percent by 2024. The extra money would not go to the United States, as Trump occasionally and erroneously suggests; it would pay for new weapons and other kit needed mainly to defend Europe. So it makes strong sense for European allies to start raising defense budgets for their own security, and they have—by some $10 billion in 2016 alone.

The counterterrorism ask is harder to meet. The United States shows no signs of wanting to reassign the leadership of the fight against the self-proclaimed Islamic State from the Global Coalition Against Daesh to NATO (nor would all European allies agree). That leaves the alliance with few ways to contribute in a meaningful and consequential manner. Two options could be to expand its military training program in Iraq and do more to strengthen Libya’s flagging security forces. Announcements along these lines are expected at the May 25 meeting of NATO heads of state and government in Brussels—the first with Trump at the table.

The trouble with the above options is that they may underestimate the depth of change in U.S. thinking and may not hold enough power to convince a skeptical mind such as Trump’s. He regards the 2 percent target as a past debt and may give little credit to allies for doing something that he says they should have been doing all along. These measures will be successful only if the U.S. president is prepared to change his mind on NATO and is merely looking for tweetable successes to justify the shift.

To really get Washington’s attention, one former senior U.S. official said in an interview, Europe should propose to the United States a new security project of transformative significance. This could come in the shape of a joint Marshall Plan–like effort to shore up the fledgling democracies of North Africa, for example. The idea is to generate a security benefit—such as fewer terrorists with the potential to harm the United States—at a scale that would firmly imprint on American minds that a joint effort with Europe makes America safer.

Time for Extra Insurance

There is always the possibility that no argument and no new transformative idea will reverse the gradual drift between the United States and Europe. Such an outcome is neither near nor inevitable, so the allies’ focus must be on making sure it will never come to this point. But given the flow of U.S. politics, it may be time to start reflecting on how to maintain Europe’s security without the United States.

The alternatives are all worse than the present arrangement, but some are marginally better than others. For a start, some security challenges in Europe appear simply unmanageable without the United States. The possibility of a conflict with Russia falls into this category. In this case, U.S. troops and weapons, including nuclear weapons, would make the crucial difference between putting up a fight and actually winning. This realization—rather than a desire to please Trump—is the reason why every one of the allies on NATO’s eastern border has raised defense spending. They have little choice but to double down on the relationship with the United States and hope that it works.

Other European allies seem more concerned with mass inflows of refugees and migrants and with terrorism. Both of these challenges are better tackled in cooperation with other allies, but if need be, it is fully within European states’ means to manage them alone. One option is to build European defenses entirely without NATO or the United States, under the guise of the EU. Politicians including European Commission President Jean-Claude Juncker have proposed such an option in recent months. Their idea carries the added attraction of raising the EU’s profile at a time of fluctuating support for the institution. For a minority in Europe, the idea of the continent without NATO also satisfies a craving to show the United States the boot.

Europe’s politicians are right to explore a backup to NATO. But a credible defense plan must deal with the full spectrum of threats if it is to win the support of all European allies. Deterrence of and defense against Russia is a big challenge that requires high-end military skills such as the ability to suppress air defenses, detailed military plans, and a command structure to guide operations in times of war. Even these do not guarantee success: Russia may still conclude that its nuclear arsenal and willingness to take risks—unmatched in Europe—gives it the ability to threaten European countries into not defending their allies. But without plans, commands, and sophisticated weapons in meaningful numbers, the Europeans are certain not to impress Moscow—and may therefore be unable to deter it from misbehaving.

These capabilities do not currently exist outside NATO. Nor are they easy to replicate quickly. The EU has led a number of military operations, but nothing resembling a conventional war with a sophisticated foe; the member states have deliberately left this task to NATO. So Europe’s first choice as it reflects on its defenses in the Trump era must be to seek an arrangement with the alliance that preserves the recourse to NATO’s plans, commands, and assets (such as an airborne warning and control system or the fleet of C-17 transport aircraft based in Hungary)—even if the United States stays out of the conflict.

This would be a new challenge, but one in keeping with the overall trend: Europe has increasingly played a leading role in military operations in or near Europe. One would have to go back to the 1999 air war against Yugoslavia to find the last operation on the continent or in its immediate neighborhood that the United States initiated, launched, and commanded. All subsequent operations were conducted by the EU or at least at Europe’s initiation, such as the 2011 NATO operation in Libya. The United States under Obama even started to contribute forces to EU-flagged operations—two in the Democratic Republic of the Congo and one in Kosovo. So the idea of European countries leading and the United States supporting is already an established reality.

However, there is a substantial difference between Europe commanding relatively small missions with U.S. support and Europe leading a potential high-intensity war with little to no U.S. backing. Such an approach would bring a number of untested firsts and would need to be vigorously exercised. It would also likely require European allies to raise defense spending above what they have already pledged—though they could reduce the bill by making use of existing NATO assets.

No Walking Away

There will be opposition in some quarters of Europe to the idea of linking the continent’s security to NATO assets. Trump is unpopular in Europe, so it is easier to make the case for the EU to distance itself from the United States than for greater reliance on NATO commands or aircraft. But the task before European allies, should the United States disengage, is to build a defense arrangement that is capable of deterring foes and defending the continent—not one that scores political points. That leaves few options other than preserving access to NATO assets for an operation without the United States.

After all, European allies have heavily cofinanced and staffed the alliance’s commands, cowritten its defense plans, and paid for much of its commonly owned hardware. Now would be the wrong time for the Europeans to walk away from their own investments.