On 1 April 2026, Donald Trump told the Daily Telegraph that he was strongly considering withdrawing the United States from NATO — when asked whether he would reconsider after the war against Iran, he replied: “beyond reconsideration”. I’m not reconsidering. The day before, on 31 March, Marco Rubio on Fox News had framed the question with the bluntness of someone who already considers it answered: when allies deny their bases, what is NATO for? Why are we in it? The implicit reply, unstated but legible in every syllable, is this: we are in NATO to protect allies who do not behave like allies. And if they do not behave like allies, we will find other ways to defend our interests.

The statement came after weeks of tensions with European allies who had refused to make their bases and airspace available for American operations against Iran. France, Germany, Spain, Italy — with different degrees, different languages, different motivations — all resisted, in one way or another, the pressure from Washington. And Washington responded as those unaccustomed to refusal respond: with the threat of abandoning the game.

It is worth pausing on this sequence with the analytical calm it deserves. Not to defend NATO as a sacred institution — it is not, and those who treat it as such forfeit the capacity to think. But to understand what actually happened, who violated what, and why Trump’s narrative — the ungrateful allies who betrayed America — is, in the precise terms of the facts, turned a hundred and eighty degrees around.

II. The Structure of the Paradox

Washington’s reasoning, in its most elementary form, runs as follows: the United States protected Europe for seventy years, maintained military bases on European soil at its own expense, guaranteed the security of the continent under the Atlantic nuclear umbrella. In return, it asked for loyalty. When America needed that loyalty — when it asked allies to open their bases for a necessary and justified military campaign — the allies said no. That no is a betrayal. And betrayal has consequences.

It is a line of reasoning with its own internal consistency, and it would be dishonest to deny it. It would also be dishonest, however, to overlook what is missing from this reasoning. Missing is a detail of some significance: Washington did not ask. It used.

On 27 March 2026, as I reconstructed in the previous piece, American bombers were already in flight toward Sigonella when someone remembered to communicate the flight plan to the Italian Air Force. Not a prior request. A notification after the fact was already underway. The planes were in the air, the route was calculated, the mission had, in practice, already begun. Only at that point did someone consider that it might be worth informing the government of the country over which those planes were about to land. Minister Crosetto said no. He applied the procedures set out in the bilateral treaties in force. He did exactly what those treaties provide that an Italian government should do when the procedure is not followed. And Washington was furious.

This is the paradox worth pressing on, because it has a precise logical structure that deserves to be made visible: the United States was using an ally’s soil without authorisation, in violation of agreed procedures, to conduct a war that ally had not chosen. When the ally enforced its own rules — the same rules Washington had signed, the same rules on which the system of American bases in Europe is founded — the response was: traitors. Those who do not respect the rules accuse those who do of having betrayed the alliance. This is an argumentative structure with a precise name in formal logic. In international relations it is called, more simply, bullying.

III. What the North Atlantic Treaty Says

The North Atlantic Treaty, signed in Washington on 4 April 1949, has an Article 5 that everyone cites and an Article 1 that almost no one remembers. Article 5, the collective defence clause, has become the symbol of the alliance: an attack against one is an attack against all. Rightly cited on 12 September 2001, rightly invoked every time European security is discussed. But Article 1 says something equally fundamental, and says things that sit poorly with American conduct in recent months: the parties undertake to settle any international dispute in which they may be involved by peaceful means, and to refrain from the threat or use of force in any manner inconsistent with the purposes of the United Nations.

Article 1, note, is not a preamble of intent. It is a contractual obligation. An obligation that the United States, by launching Operation Epic Fury eleven days after reaching an agreement in Geneva, while negotiations were still formally under way, at the very least evaded. Diplomacy did not fail: it was interrupted. That distinction is not rhetorical — it has precise legal and political consequences. An alliance whose very first article commits parties to the peaceful resolution of disputes cannot ask its members to be complicit in a military campaign launched while a diplomatic solution was being negotiated.

No one says this in these terms, of course. The language of foreign policy has its conventions, and among those conventions is the refusal to apply to the most powerful allies the same standards applied to everyone else. But conventions do not change facts. The facts are that Washington launched a war while it was negotiating, used allied bases without following procedures, and now threatens to abandon the alliance because the allies followed procedures. The retaliation is against those who enforced the rules Washington itself had signed.

IV. The Alliance as Protection and as Constraint

There is one way of reading NATO, widespread in Atlanticist circles of every political persuasion, that reduces it to an insurance policy: the Americans protect us, we pay the premium with our bases and political loyalty, and in return no one invades us. It is a reading with its own logic during the Cold War, when the Soviet threat was real and the American nuclear umbrella was the only concrete guarantee of survival for Western Europe. De Gasperi in 1949 knew perfectly well that Italy was signing an asymmetric pact: security in exchange for presence, protection in exchange for subordination. It was a deliberate choice, made in a specific context, with a specific enemy at the door.

There is another way of reading NATO, far less common, that treats it as a pact founded on mutual obligations, shared procedures, a system of rules that applies to everyone. In this reading, the alliance is not a hierarchy with Washington at the top and the others below — it is a federation of sovereignties that have agreed on certain rules and expect those rules to be respected by all, including the most powerful member. This reading corresponds to the text of the Treaty, to the letter of its articles, to the legal form chosen by the alliance’s founders. The problem is that practice has systematically betrayed form. America has acted within NATO like a majority shareholder who tolerates minority shareholders as long as they do not obstruct its plans.

The two readings coexisted for decades in an undeclared tension. Europeans preferred not to choose between them, collecting the benefits of American protection without ever confronting the political cost of explicitly defining the limits of that dependence. Washington maintained the ambiguity, which allowed it to invoke the alliance when it needed it and ignore it when inconvenient. The system worked as long as interests were sufficiently aligned. When they ceased to be — as in 2003 with Iraq, as today with Iran — the tension became visible and the ambiguity did not hold.

Trump has made explicit, with a frankness one can in some measure appreciate, the first reading. NATO is an instrument serving American interests. If allies are not useful, the alliance has no reason to exist. That is not a surprising logic — it is the logic of any great power in any historical era. What is surprising is that there are still, in Europe, political leaders who pretend not to recognise it or who are shocked when it is stated without euphemism.

The alternative is not naïveté. The alternative is to acknowledge that an alliance founded exclusively on dependence upon the strongest member is not an alliance — it is vassalage with benefits. And that the correct response to a threat of abandonment is not supplication, but building the conditions to no longer need it. That construction requires time, resources, political will. It requires above all the willingness to ask uncomfortable questions about what has been accepted for decades without discussion — a willingness that in Italy remains largely absent.

V. The Precedent Nobody Wants to See

In 2003, when the United States invaded Iraq, France and Germany said no. Chirac and Schröder opposed it publicly, openly, in the Security Council. Donald Rumsfeld coined the expression “Old Europe” to brand this resistance with contempt. George W. Bush built the coalition of the willing — the coalition of the willing — bypassing the Security Council and ignoring the allies who refused to follow. The consequences of that war — the destabilisation of Iraq, the rise of ISIS, the regional chaos that continues to this day — are well known.

Who was right in that case is not a rhetorical question. It is a question with a fairly clear historical answer, even if few state it with this bluntness. France and Germany were right. Not because Saddam Hussein was an acceptable leader — he was not, for the same reasons the Islamic Republic is not acceptable today. But because the war was wrong, was illegal under international law, was founded on false information, and its consequences were foreseeable by anyone who had read the UN inspectors’ reports attentively.

There is, however, an element of that crisis that is systematically forgotten when 2026 is discussed. In 2003, despite the open political conflict between Washington and Paris, no one expected France to allow American military aircraft to overfly its territory without authorisation, or for NATO bases in Europe to be used for the Iraq campaign without prior consultation with the host allies. The principle of consultation was respected, even within the conflict. What distinguishes today from then is not the quarrelsomeness of allies — that has always existed. It is the fact that Washington decided it could dispense with consultation entirely, as if the European bases were operational appendages of a command accountable to no one.

That precedent matters, because it establishes that European allies’ resistance to American military ventures is not a novelty of 2026, is not anti-Americanism, is not the caprice of left-wing European governments. It is a position with historical roots, defended by centre-right governments like Chirac’s, that subsequently found retrospective justification in the Iraqi catastrophe. And that today resurfaces with an analogous logic, even if the context is different and the stakes are higher still.

VI. What Trump Really Wants

The threat to leave NATO must be read on two levels. The first is the tactical level: a negotiating pressure, a raised voice designed to produce concessions before the crisis stabilises. Trump has used this technique systematically — with trading allies, with diplomatic partners, with his own collaborators. The threat as an opening move. “Beyond reconsideration” as a starting point, not an endpoint.

The second level is more structural, and more troubling. Trump represents a current in American strategic thinking — not new, not marginal — that holds that American engagement in Europe is costly, unproductive, and historically obsolete. That current has deep roots. Robert Taft, senator from Ohio and leader of Republican conservatives in the forties and fifties, opposed American entry into NATO with arguments Trump could endorse word for word: permanent alliances commit America to European wars that are not its concern, limit Congress’s freedom of action, and serve the interests of the East Coast establishment more than those of the real country. Taft lost that battle. But his current never disappeared — it remained latent, resurfacing at regular intervals whenever America felt overwhelmed by the weight of its global commitments.

Trump is the most fully realised product of that tradition in contemporary American politics. He is not a pure isolationist — he is willing to use military force, as the campaign against Iran demonstrates. But he is a committed unilateralist: America acts for its own interests, alone or with whomever it chooses to involve, without asking permission and without accounting to alliances it regards as cages rather than resources. In this vision, NATO is not a value — it is a tool. And a tool that does not work is changed or discarded.

Knowing which of the two levels Trump’s words are operating on at this moment is not simple, and he probably does not know with precision himself. But the European response cannot depend on this uncertainty. It must be built on the assumption of the worst case — that the threat is serious, that the alliance is genuinely in question, that the European security architecture built over seventy years is less solid than was always believed.

One element of the 1 April address to the nation deserves specific attention. In the hours preceding the speech, Trump had told reporters he would express his disgust for NATO before the country. In the formal address, lasting nineteen minutes, the Atlantic Alliance was not mentioned once. That is not a change of mind — it is a directorial choice. The aggressive message about NATO was intended for European allies, through the international media. The address to the nation was intended for Americans, and Americans are not interested in NATO. Different audiences, different messages, calibrated with the precision of someone who knows exactly what he wants to obtain from each. Trump’s unpredictability is not chaos — it is a method. Understanding it is the first step toward not being overwhelmed by it.

And here the most uncomfortable question opens: if this is the situation, what does it mean for Europe, and for Italy in particular, to keep discussing sovereignty over bases as though it were a technical matter of administrative procedure? What does it mean to assert the right not to participate in a specific war, without confronting the deeper question of the system of strategic dependence within which that war is situated?

VII. The Italian Knot

Italy, in this crisis, has a specific problem that other European countries do not have to the same degree. The problem is not the scale of American bases on its territory — even if that scale is significant, with Aviano, Sigonella, Vicenza, Camp Darby between Pisa and Livorno, Gaeta, Naples. The problem is the quality of the political debate about what those bases mean, about what relationship we want with them, about what conditions we are prepared to accept in exchange for the American presence.

That debate does not exist. Or rather: it exists in the form of the episode. It exists when something happens — when bombers are turned away, when a base is caught up in a news story, when the tension becomes impossible to ignore. And then it disappears, reabsorbed by the inertia of a system that prefers not to ask questions because the answers would be uncomfortable.

The comparison with other major European countries is instructive. Germany, since the end of the Cold War, has built an explicit and recurring parliamentary debate on the American military presence on its territory, on the limits of its use, on the compatibility between those agreements and German foreign policy. An imperfect debate, not always conclusive, often instrumental — but existing, visible, capable of producing real constraints. France resolved the problem at the root by leaving NATO’s integrated military command in 1966 under De Gaulle, and rejoining only in 2009 under Sarkozy — a choice that maintained for decades a margin of autonomy no other European ally has ever had. Italy did neither. It accepted the presence, collected the economic and political benefits, and preferred not to discuss the conditions.

The result is that today, when the crisis makes those conditions suddenly relevant, Italy has no position. It has a gesture — Crosetto’s no, courageous and correct — but no policy behind that gesture. It has no parliamentary framework, no public debate worthy of the name. It has a minister who did the right thing for the right reasons, expressing the will of a government with its own democratic mandate — but in almost total institutional isolation, without anyone ever having asked Parliament, and through it the citizens, whether that position corresponds to a deliberate choice or merely an emergency reflex.

Meanwhile the bills arrive. On 2 April a rocket struck the Shama base in Lebanon, headquarters of the Italian Unifil contingent — the second attack on Italian personnel after the Erbil strike on 12 March. No casualties, fortunately. But Italian soldiers are being hit in a conflict the Italian Parliament never voted on, in a theatre no one democratically decided to staff under these conditions. And on the economic front: S&P Global has cut Italy’s growth projections for 2026 to 0.4%, halved from the previous forecast. The cost of the closure of Hormuz — a fifth of global oil supplies blocked in the Persian Gulf — is paid by Italian families at the petrol pump, by Italian businesses along their supply chains, by the Italian economy in the year-end figures. This is not a distant war. It is already here.

Trump’s threat to NATO changes the coordinates of this deficit radically. Not because it makes an American departure from the alliance imminent or certain — it is not, and probably will not be. But because it makes evident, for anyone willing to see it, that the certainty on which the entire architecture of European security has been built — the American guarantee as a stable, predictable, non-negotiable element — is no longer such. Not from today, not suddenly: for years. But today it is stated openly, without euphemism, by a sitting American president.

In this context, Italy’s posture of calculated ambiguity — not too close, not too distant, never say anything that might be read as a clear choice — is no longer sustainable as a strategic stance. It is sustainable as a short-term tactic, to buy time while the crisis evolves. But it is not a policy. And a country with no policy on where it stands, what it defends, in which Europe it wants to live, is a country that leaves others to decide for it.

VIII. The Spanish Lesson, Revisited

I wrote, in the previous piece, about the Spanish response: the closure of airspace to American military aircraft involved in Iran, a gesture that was not a reaction to a procedural episode but the assertion of a political position. That response deserves to be re-examined in the light of Trump’s threat to NATO, because its logic becomes even clearer.

Spain chose. It did not wait for Washington to define the terms of the relationship. It defined its own terms, publicly, with the consequences that entailed in terms of diplomatic pressure and the risk of retaliation. It did so because an explicit position, however uncomfortable, is more defensible than an ambiguity that leaves every interpretation open. It did so because a country that knows where it stands is in a stronger negotiating position than one that does not.

Trump’s response to European resistance was not that of an offended ally demanding explanations. It was that of someone who has understood that protection is no longer perceived as indispensable in the way it once was. And that those who do not perceive it as indispensable are not easily blackmailed with the threat of withdrawing it.

This is the true paradigm shift the 2026 crisis is producing, slowly, through the pressure of events: Europe is discovering — or is being forced to discover — that strategic dependence on America is not a given of nature. It is a choice. A choice made seventy years ago in very different circumstances, which today can be revised, renegotiated, redefined. Not through rupture — not immediately, not rashly. But through a deliberate choice, made democratically, with eyes open to the consequences.

IX. The Question That Remains

Trump says that the American withdrawal from NATO is “beyond reconsideration” — I’m not reconsidering. Rubio asks why the United States is in NATO if allies deny their bases. These are destabilising statements, made in bad faith or good faith, it does not matter: they produce a precise effect in the European interlocutor. They force a response. They force the question: and us? Why are we in NATO? What does this alliance defend, in the Europe of 2026? On what terms do we want to remain?

These are questions that provoke fear, because the answers are difficult and the consequences of wrong answers are enormous. But they are questions that must be asked, precisely because they provoke fear. An alliance that cannot be discussed is not an alliance — it is a dogma. And dogmas, in politics as in religion, survive as long as no one puts them to the test. When they are put to the test — as Trump is doing, with the brutality of someone with no interest in preserving diplomatic conventions — they either withstand scrutiny, or they reveal their fragility.

Does NATO withstand scrutiny? The honest answer is: it depends which NATO. The alliance as a system of collective defence founded on mutual obligations, shared procedures, a negotiated equilibrium among the sovereignties of its members — that NATO is defensible, and worth defending. The alliance as a hierarchy with Washington at the top, as a mechanism through which European bases are used without permission for wars Europeans did not choose — that NATO does not withstand scrutiny, and would not deserve to.

The 2026 crisis offers an unexpected opportunity, and like all unexpected opportunities it risks passing uncaptured. The opportunity is this: to reformulate, explicitly and without euphemism, what alliance we want. Not the one we inherited from the Cold War. Not the one Trump would reduce to a pure relation of force. The one we could build, if we had the clarity to understand what we want to defend and the courage to say so.

The paradox of demanding loyalty after violating the rules has a simple solution, in theory: one reminds those who demand it that the rules apply to everyone. One notes that the alliance was not a blank cheque, that the bases were not American property, that allied consent was not implicit and assumed but explicit and negotiated. One says, with the steady voice of those who know they are right, that those who violate the procedures have no standing to complain about those who enforce them.

But there is a question more uncomfortable than all the rest, which the 2026 crisis has made impossible to evade: are Americans still our allies? Or are we simply vassals who had the temerity to remember we exist? The difference is not rhetorical. An ally discusses, negotiates, respects procedures even when it is the strongest party. A vassal is consulted when convenient and ignored when not. Washington’s anger at Crosetto’s no — and the tone of Trump in the Telegraph — suggests that someone, on the other side of the Atlantic, has forgotten which of the two is our role in the pact.

Craxi said it in 1985, on the tarmac at Sigonella, with the carabinieri facing down Delta Force. Crosetto did it in 2026, with procedures and signatures. The next step — taking this position beyond the episode, into a policy, before Parliament, before the people who have the right to choose — remains to be taken.