But the book addresses only the demand side. It explains why so many Romanians wanted radical change. It does not fully explain why that demand took the specific form it did instead of calling for a stronger welfare state or a more redistributive economic policy, but an indirect call for multivectorial foreign policy, strategic neutrality, and a fundamental reorientation of Romania’s place in the world. Călin Georgescu, who won the annulled first round in November 2024 with 22.9%, and George Simion, who won the repeat first round in May 2025 with nearly 41%, did not campaign primarily on economic redistribution. They campaigned against the foundational foreign policy orientation that has defined Romania since 1995. And since then the Romanian political debate has become structurally divided between ‘pro-europeans’ (or pro-Westerners, to use President Nicușor Dan’s preferred term) and the ‘pro-Russians’.

That orientation has a name: the Snagov consensus. Signed on 21 June 1995, by the leaders of all parliamentary parties at a villa on the shore of Lake Snagov, the declaration committed Romania’s political class to EU as the country’s strategic destiny. It was reaffirmed in 2001 with a NATO-specific commitment. For nearly three decades, this Snagov consensus broadly held. It survived government alternations between left and right, economic crises, the slow grind of accession negotiations, and the disappointments of post-accession politics as well as the constant struggles between different clientelistic gangs dominating Romanian politics. Nicușor Dan ultimately won the presidency on a pro-European platform in May 2025, but the structural fact remains that over 40% of Romanian voters in 2025 chose a candidate running against this consensus.

What Pop, Alexandrescu, and Anghel document so well is the demand for radical change, for revolution in the domestic and international politics of the country. What I want to offer here is a theoretical framework that explains why these socio-economic grievances in Romania necessarily feed into foreign policy contestation. The framework I want to develop comes from a much appreciated political economist that studied the interwar period named Karl Polanyi.

Prior

Pop, Alexandrescu, and Anghel give us the socio-economic foundation. I want to build on it with a structural argument about why this socio-economic grievance expresses itself as foreign policy contestation rather than, say, demands for a stronger welfare state.

Polanyi, writing in 1944, argued that the expansion of market society produces a predictable counter-movement. When economies are “disembedded” from social relations, when labor, land, and money are treated as commodities to be allocated by market forces, the social costs accumulate until society mobilizes to protect itself. Polanyi called this the “double movement”, understood sequentially as the first movement of market liberalization, followed by a second movement of social self-protection. The counter-movement is driven by the structural dislocation produced by market liberalization, not necessarily by ideological factors, as some cultural explanations of populism would claim. The ideology comes later, as a vehicle.

My claim is that Romania’s Snagov consensus was a political economy commitment that Romanian elites took under the veil of foreign and security policy. This is also visible if one reads the report produced by the ‘Snagov Committee’ preceding the Snagov Declaration, which was tasked only with developing a new strategy for economic development for the country. “Joining the West” bundled together a strategic orientation (NATO, EU, the United States), an economic model (market liberalization, privatization, FDI-driven growth, labor market opening), and a social contract (the promise that integration would deliver broadly shared prosperity). The bundle was what we now call the ‘Snagov consensus’. And because the bundle was inseparable, contesting the economic model eventually meant contesting the foreign policy that enabled that model.

Evidence

The first movement, Romania’s disembedding into the Western liberal order, delivered real gains. GDP per capita rose from roughly 26% of the EU average at the time of accession to over 75% of the EU average by 2024. In this regard, Romania became one of Europe’s fastest-growing economies. Bucharest, Cluj, Timișoara, and Iași have been transformed into dynamic urban centers integrated into the European single market.

But the first movement also generated severe social costs.

The most dramatic is emigration. Between Romania’s political transition and 2021, the country lost approximately 17% of its population. More than four million Romanians, roughly one in five, left to live and work in Western Europe — me included 🇨🇭. In 2023, Romania had 3.15 million citizens living in other EU member states, the highest absolute number of any EU country. As a consequence, entire regions of the country aged rapidly. Towns lost their working-age populations. The country developed what the many studies have called one of the fastest depopulation rates in the European Union. EU accession in 2007 accelerated the trend by removing the remaining barriers to labor mobility.

This is Polanyi’s commodification of labor in its purest form. Romanian workers became inputs in Western European labor markets. Construction workers in Germany, care workers in Italy, agricultural laborers in Spain. The Romanian economy received remittances that sustained the families that remained home and contributed to (rather shallow) domestic demand. Western European economies received cheap, skilled labor. The human cost was borne by the communities left behind, particularly by quasi-abandoned children.

Beyond emigration, Romania also experienced the deindustrialization of non-competitive sectors, a widening urban-rural inequality, the concentration of economic gains from integration in the single and world markets among the educated, mobile, and urban. And a pervasive perception, not entirely wrong, that the political class had captured the management of European integration for its own benefit.

Here is the critical structural point I would like to make, drawing on Polanyi. In most Western European countries, distributional grievances generated by economic integration can be separated from foreign policy. A French voter can oppose austerity without opposing NATO membership. A Greek voter can contest the terms of the eurozone bailout without demanding a strategic reorientation toward Russia. The economic model and the foreign policy alignment are distinct commitments that can be contested independently.

In Romania, they cannot. The Snagov consensus bundled them together. EU accession was simultaneously the economic model and the foreign policy. There is no way to contest the distributional consequences of integration without contesting the integration project itself in some form. This is why the counter-movement in Romania takes a stronger foreign policy-related form rather than a purely domestic one, as we see in Germany or France. Georgescu does not merely demand redistribution within the existing alignment. What he demands is a new alignment, a multivectorial foreign policy, because the economic model and the strategic orientation are the same project.

Multipolarity supplies the content for this demand. In a unipolar world, the counter-movement had no foreign policy program to offer. There was nowhere else to go. The autochthonist current in Romanian politics, which Georgescu embodies, the tradition of resistance to external integration that runs from Eminescu through Ceaușescu, could offer cultural grievance but not a viable strategic alternative. The return of great power competition changes this, however. Once China, Russia, and other poles are visible on the horizon as credible and feasible sources of alternative benefits and opportunities, multivectorialism becomes articulable as a policy program that can also respond to social dislocation problems and economic development problems. The international system provides the menu of options that makes Georgescu and Simion’s counter-movement’s foreign policy demand coherent, even if, in Romania’s case, the options are largely illusory.

Romania has been here before. In the 1920s, an elite-driven liberal consensus anchored Romania in the French alliance system, the Little Entente, and the League of Nations. That consensus, too, bundled strategic alignment with a liberal economic model. And it, too, eroded when the social costs of that model accumulated and an alternative pole appeared. France and Britain’s indifference to Romania’s economic plight during the Great Depression undermined domestic faith in the Western guarantors. Nazi Germany’s growing economic weight in Southeastern Europe offered an alternative. Romanian politicians, as Rebecca Haynes has documented, did not merely submit to German pressure. They actively pursued rapprochement with Berlin. The counter-movement found its pole.

The structural homology between today’s political developments and those of the 1920-30s is instructive. In both periods, an elite consensus bundled foreign policy with political economy. In both periods, the social costs of the economic model accumulated over time. In both periods, the consensus held as long as the international system offered no alternative pole. And in both periods, the appearance of a credible, or seemingly credible, alternative enabled the counter-movement to acquire a foreign policy program and engage in domestic partisan contestation.

Posterior

If this analysis is correct, then the standard prescriptions for defending the pro-Western consensus are treating symptoms rather than causes.

Securitization of the external threat, the argument that Romania faces Russian aggression and therefore must maintain Western alignment, addresses the permissive conditions (multipolarity) but not the productive conditions (distributional dislocation). It asks the populations who bore the costs of integration into the single and world markets to accept those costs indefinitely because the geopolitical environment is dangerous. This may work in the short term. Nicușor Dan’s election suggests it did. But it does not resolve the underlying structural tension that exists and which Alina Pop, Filip Alexandrescu, and Ionuț-Marian Anghel new book demonstrates.

‘Democratic resilience’ programs, meaning counter-disinformation, media literacy, and electoral integrity measures, addresses the transmission mechanism (how external actors exploit domestic grievances) but not the grievances themselves. It is the equivalent of treating a fever without diagnosing the infection but instead just putting a blanket over the patient with the hope that he’ll get better.

A Polanyian diagnosis points to a different prescription. If the counter-movement is generated by the social costs of the integration model that Romania pursued over the past three decades, then the only durable way to defend the consensus is to renegotiate the social contract it rests on. The gains of integration must be redistributed more broadly. The communities that lost their populations to emigration, their industries to competition, and their young people to Western European labor markets need a material stake in the continuation of Romania’s Western alignment.

This is not a romantic appeal. Everyone who knows me knows that I do not make emotional appeals. The Snagov consensus was a cross-class bargain. It held for so long because it promised that European integration and the pro-Western orientation of the country would benefit everyone. The post-accession reality concentrated the benefits in the hands of few. The social base that bargain eroded as the reality of the asymmetrical costs and social dislocation became unavoidable. Either the bargain is renewed on terms that address the distributional failure, or the counter-movement will continue to find its audience, regardless of who occupies the Cotroceni Palace or what Russia or other malign actors may do or not.

Nicușor Dan’s victory buys time. It does not buy a solution. The 41% who voted for Simion in the first round are not going away. They are the counter-movement’s social base, and they have structural reasons to be there. And Dan’s first year in office has done nothing to convince us that things are going in the right direction.

Prior

Two theoretical frameworks generate competing expectations about regional security cooperation that are useful for the case of the Black Sea region.

The dominant framework is Stephen Walt’s balance-of-threat theory. Walt argues that states balance not simply against power but against threat, which he defines as a combination of aggregate capability, geographic proximity, offensive capacity, and perceived aggressive intent. The core prediction of Walt’s theory is that when a group of states faces a common and proximate threat, they have strong incentives to balance against it through enhanced cooperation. The higher the perceived threat, the stronger the incentive. When the balance-of-threat prediction fails to materialize, analysts working within this framework tend to reach for a residual explanation. The problem must be one of political will. The structural incentives are there, but leaders have failed to act on them. Align the incentives, convene the right people, and cooperation will follow.

The challenger framework I want to introduce here is Barry Buzan and Ole Waever’s regional security complex theory. Buzan and Waever argue that security dynamics are not uniformly distributed across the international system. They cluster into regional security complexes, durable patterns of amity and enmity among geographically proximate states whose security concerns are sufficiently interconnected that they cannot be analyzed or resolved independently of one another. Within a structured security complex, states share linked threat perceptions. What happens to one directly affects the security calculus of the others. But not all geographic spaces constitute structured complexes. Some areas are what Buzan and Waever call ‘unstructured security spaces’, zones where the security dynamics of adjacent regional complexes meet without generating their own autonomous pattern of cooperation among the states located there.

In other words, Walt expects cooperation to track threat intensity. The higher the threat, the more cooperation we should observe. Buzan and Waever expect cooperation to track structural linkage. Even under conditions of high threat, an unstructured security space will produce low levels of cooperation because the states within it do not share structurally linked security dynamics. Their threat perceptions may be similar in content, but they are not functionally interdependent.

For the Black Sea, Walt predicts robust cooperation. The threat is high and proximate. Buzan and Waever predict weak cooperation, if the Black Sea is an unstructured security space where multiple regional complexes overlap without generating their own coherent pattern. The opposite is the case for the Baltic Sea. Because the threat in the Baltic Sea is lower and there is no active naval warfare, cooperation should be weaker. Conversely, Buzan and Waever predict strong cooperation because the Baltic Sea is deeply embeded in the European security complex and thus it is a highly structured security environment.

Evidence

The Black Sea region experiences objectively higher levels of threat than the Baltic. There is an active naval war in these waters. Russia has deployed its fleet aggressively, mined shipping lanes, attacked civilian port infrastructure, and maintained a posture of coercive escalation against NATO littoral states. NATO’s Sea Shield 2026 exercise, which launched from Constanta just days before my Varna conference with 2,500 personnel and 48 ships from 13 nations, is a direct response to this threat environment.

The Baltic states face lower levels of active military threat. No naval warfare, no mine contamination, no direct attacks on civilian shipping.

And yet the Baltic has produced significantly higher levels of regional security cooperation. The Nordic-Baltic Eight format has evolved into what some analysts now call Europe’s most reliable security engine, jointly financing a 500 million dollars Ukraine military support package in 2025, constructing the Baltic Defence Line along their borders with Russia and Belarus, and approaching defense spending of 5% of GDP. NATO’s Operation Baltic Sentry has turned the Baltic Sea into one of the most monitored maritime spaces in Europe.

The Black Sea, by contrast, remains a space of strategic fragmentation. The EU published a Black Sea strategic approach in May 2025, but it lacks a concrete action plan, a clear timeline, and a dedicated budget. Previous frameworks like the Black Sea Synergy and the Eastern Partnership failed to address regional security challenges that systematically emerged over the past three decades.

This pattern is anomalous for Walt but is quite expected by Buzan and Waever.

The Baltic fits the model of a structured security subcomplex. Estonia, Latvia, and Lithuania share the same strategic axis, the Russian western military district. They share the same geographic vulnerability, as small states on NATO’s northeastern flank. And they share the same institutional embedding, through Nordic-Baltic cooperation formats accumulated over decades. Their threat perceptions are structurally linked. What happens to one directly and immediately affects the security of the others. This structural linkage, not the level of threat per se, is what enables the dense cooperation we observe.

The Black Sea displays the characteristics of an unstructured security space. It sits at the intersection of at least three regional security complexes, the European, the Middle Eastern, and the post-Soviet Central Asian. Turkey illustrates this most clearly. Ankara has security commitments and strategic relationships spanning the Eastern Mediterranean, the Middle East, and the Caucasus. Its strategic attention is structurally dispersed across these spaces. This is a geographic reality that any Turkish government would face regardless of ideology or alliance preferences. Turkey’s relationship with Russia in the Black Sea cannot be isolated from its relationship with Russia in Syria, in Libya, in the South Caucasus, or in the energy transit corridors connecting Central Asia to Europe. Russia understands this. As a recent Chatham House analysis observed, Moscow’s preferred vision for the Black Sea is a de facto condominium with Turkey, with Russia asserting control over the northern basin and Ankara over the south.

Romania and Bulgaria face a different version of this structural problem. Both are embedded in the European regional security complex, but their strategic orientations diverge. Romania’s security axis runs northeast, toward Moldova and Ukraine. Bulgaria’s pulls toward the Western Balkans and the Eastern Mediterranean. They share a coastline, but not a single strategic depth.

Moreover, in the Baltic the limits of indigenous cooperation among smaller states have been partially compensated by external force projection. The United Kingdom, Germany, Canada, and the United States all maintain enhanced forward presence battlegroups on NATO’s northeastern flank. The Black Sea has no equivalent option. The 1936 Montreux Convention gives Turkey sovereign control over the Straits and limits the tonnage and duration of stay for non-littoral warships. Since February 2022, Turkey has invoked the Convention’s wartime provisions to close the Straits to belligerent warships altogether. Non-littoral NATO allies cannot sustain a meaningful naval presence in the Black Sea. The Baltic can import balancing from outside. The Black Sea cannot. This legal-institutional constraint further isolates the Black Sea from the compensatory mechanisms available to other NATO flanks and reinforces the structural fragmentation that Buzan and Waever’s framework predicts.

Posterior

The comparative evidence we have favors Buzan and Waever over Walt. Walt’s balance-of-threat prediction fails the Black Sea-Baltic test. The region with higher threat has produced less cooperation. The region with lower threat has produced more. Political will is not the missing variable. The Baltic states do not cooperate more because their leaders are more willing. Instead, they cooperate more because they inhabit a structured security subcomplex in which their security concerns are functionally interdependent. The structure generated the cooperation, not the other way around.

The Black Sea’s cooperation deficit is structural. Romania, Bulgaria, and Turkey may share a similar view of Russia as a threat, but they do not inhabit a common security system. Their threat perceptions converge in content but diverge in structure. Each state faces Russia through a different strategic axis, with different stakes and different constraints. The Black Sea is an unstructured security space, and it behaves like one.

This does not mean cooperation is impossible, however. It only means that the pathway to cooperation runs through structure-building rather than through appeals to political will. If the Black Sea is to develop higher levels of security cooperation, it must first be transformed from an unstructured security space into something resembling a structured security subcomplex linked to the European security architecture.

Implications

Three policy consequences follow directly from this diagnostic.

First, comprehensive top-down security frameworks will continue to fail. The EU’s May 2025 Black Sea strategic approach is the latest in a series of attempts to impose coherence on a structurally fragmented space. Like its predecessors, it will underdeliver, not because the EU’s intentions are wrong but because it assumes a level of strategic alignment among the littoral states that the underlying structure defining inter-actions in the region simply do not support. A framework that requires Turkey, Romania, and Bulgaria to coordinate across their divergent strategic axes will collapse under the weight of those divergences.

Second, the political will diagnosis that permeates analyses of Black Sea security cooperation produces, in my view, the wrong prescriptions. If the obstacle were volitional, then the answer would be more high-level engagement, more diplomatic energy, more joint declarations. But a decade of shared threat perception has not produced meaningful trilateral security cooperation. The fact that cooperation has not emerged despite strong incentives is itself the evidence that the obstacle is structural.

Third, Montreux forecloses the most common compensatory mechanism available to other NATO flanks. If external balancing is legally impossible in the Black Sea, then whatever security cooperation exists must be generated from within, by and among the littoral states themselves. This makes the pathway to cooperation narrower and more demanding, but it also clarifies the task at hand for security planners and political leaders.

Recommendations

If the problem is structural, the response must be structural. The Black Sea needs to be transformed from an unstructured security space into a structured security subcomplex. And the most promising way to do this is from the bottom up.

I am advocating for a Schuman-style functional logic applied to Black Sea security. Monnet and Schuman understood in 1950 that Franco-German reconciliation could not be achieved through grand political declarations. It had to be built through concrete, limited cooperation in specific sectors that would create material interdependencies and generate spillover pressures toward broader integration. The design principle was accretion, not comprehensiveness. Start small. Let functional spillover do the work of expanding cooperation over time.

The encouraging news is that this logic is already at work, even if it has not been recognized as such. The Mine Countermeasures Naval Group (MCM Black Sea), established through a trilateral memorandum of understanding signed in January 2024 and operational since July of that year, is precisely the kind of narrow, technically focused cooperation that can serve as a foundation for something larger. The group has completed four activation cycles under rotating Bulgarian, Romanian, and Turkish command, conducting joint mine reconnaissance and surveillance in the western Black Sea. Romania’s former defense minister Ionut Mosteanu has already signaled that this format should expand to include patrol forces protecting energy facilities and trade routes.

This trajectory follows the functional logic I am describing. Once officers from three navies share a common operating picture, they develop habits of communication that did not previously exist. Once mine countermeasure crews train together and develop shared protocols, they create a baseline of interoperability that can be extended to other domains. Once shared operations generate after-action reviews, they create institutional feedback loops that require further coordination. Each functional step generates pressures for the next.

The critical insight from this is that the design should aim for accretion rather than comprehensiveness from the start. A comprehensive trilateral security framework would collapse under the weight of the divergent strategic priorities described above. A narrow functional arrangement focused on mine countermeasures does not require Turkey, Romania, and Bulgaria to agree on their respective relationships with Russia across multiple theaters. It only requires them to agree that drifting mines are a shared problem. And they are.

Beyond mine countermeasures, two further domains offer immediate potential for functional deepening. Joint maritime domain awareness, meaning shared surveillance and intelligence-sharing for monitoring Russian naval activity and environmental hazards. And joint exercises focused on the protection of undersea infrastructure, which build interoperability without requiring agreement on the more politically sensitive questions of force posture and threat response.

None of these measures are dramatic. None will make headlines. But that is precisely the point. The objective should be to transform the Black Sea from a space of security liminality into a structured security subcomplex linked to the European security architecture. This transformation cannot be decreed. It must be grown from below, through the slow accumulation of shared practices, institutional habits, and operational trust.

For decades, we have lived in the warm, if occasionally stifling, embrace of the American security guarantee. In Bucharest, Warsaw, and Tallinn, NATO has always played the role of a civilizational anchor, but also that of the ultimate reassurance that the tragedies of the twentieth century would not be replayed. But as the European Commission President spoke in Davos, warning that “nostalgia will not bring back the old order,” a chill must have ran through the room that had little to do with the winter outside.

Von der Leyen’s speech was intended to empower the rich and powerful in Davos but also the millions watching online. Her speech was a call for a “new independent Europe” capable of navigating a world fractured by war in Ukraine and the return of Donald Trump to the White House. She spoke of defense spending surging toward €800 billion, of industrial mobilization, and of a Europe that finally pays its own way. But for a Romanian observer like myself, the subtext was slightly different. It was an admission that the era of outsourcing our security is over. The Pax Americana is fraying, and in the vacuum, a fierce contest has erupted over what should replace it.

This is also a debate over the nature of the Western alliance itself. Since the shocking return to power of Trump in January 2025, three distinct visions for the “Europeanization” of NATO seem to have coalesced in Europe. We might call them Consolidation, Complementarity, and Substitution. Each offers a different theory of power, a different assessment of American reliability, and, crucially, a different future for the frontline states of Eastern Europe.

The View from the Edge

To understand these visions, one must first understand the political geography of the continent. In Western Europe, strategic autonomy is often an intellectual exercise. Often time nothing more than a project of identity building or industrial policy. In the East, however, it can be a matter of life and death. We do not have the luxury of viewing the Atlantic Alliance as a “negotiable” relationship. For us, NATO is the only thing standing between sovereignty and subjugation.

This is why the first vision, which I dubbed Consolidation, finds its most fervent champions among the nations of the Bucharest Nine. This vision treats Europeanization not as a path to independence, but as a desperate bid to keep the Americans interested. The logic under which it operates is rather transactional and realistic: if we want the United States to stay, we must make it cost-effective for them to do so.

Mark Rutte, the new NATO Secretary General, has become the high priest of this orthodoxy. His quip from just a few days ago, telling those dreaming of EU defense to “keep dreaming”, should not be seen, I think, as a slip of the tongue. It was, I believe, above all an expression of a particular vision. For the consolidationists, be them Donald Tusk in Poland or Mark Rutte in Brussels, the sheer mechanics of modern warfare make European autonomy a dangerous fantasy. We lack the strategic airlift, the intelligence, surveillance and reconnaissance (ISR) capabilities, and, most critically, the nuclear depth to deter Moscow alone.

From the perspective of a planner in the Romanian Ministry of National Defense, for instance, the math seems rather simple. Duplicating NATO’s command structures within the EU is not just wasteful; it is actively dangerous because it risks decoupling Europe from the U.S. nuclear umbrella. The consolidationist vision, therefore, advocates for massive increases in European defense spending, but strictly within a particular understanding of the NATO framework. We buy American kit to ensure interoperability, as Romania has done and will likely continue to do; we host American troops to ensure they serve as a “tripwire”; and we avoid any political rhetoric that might give an isolationist Washington an excuse to get away. It is about doing more, but not about leading.

This, to me, looks very much like a strategy of “locking in” the American leadership and commitment. It seems to me that it represents, above all else, an attempt to bind Gulliver with thousands of threads of logistical and financial integration. However, I think that it also has a tragic flaw, one that gnaws at us in the quiet hours of the night. What this vision assumes is that a potential American withdrawal is a rational calculation that can be influenced by better behavior on the side of the Europeans. In doing so, it completly ignores the possibility that the shift in Washington is not about money, but about a fundamental reorientation of American national interest away from Europe and toward the Indo-Pacific, as the recently published National Security Strategy and the National Defense Strategy jointly show. We may polish the silver and cook the finest meal, but if the guest has decided to leave, he will leave. Nothing that the Eastern or Western Europeans will do has any meaningful chance of changing that.

The Brussels Compromise

If Consolidation is the strategy of the fearful East, Complementarity is the strategy of the pragmatic Center. This is the vision of Berlin, of the European Commission, and of the moderate Atlanticists who populate the think tanks of Brussels. It is a hedging strategy, born of the realization that while the marriage with America might be failing, a divorce may be too expensive to contemplate.

Those supporting a vision of Europeanizing NATO in terms of complementarity argue that Europe must build its own capacity not to replace NATO, but to save it. Or at least to keep the lights on if the landlord stops paying the bills. This was the essence of von der Leyen’s Davos address. She did not call for a rupture. Instead, she sketched a future where the EU acts as a “European Pillar” within the alliance: capable of independent action in its near abroad (the Balkans, North Africa, perhaps a limited crisis in the East) while relying on the U.S. for ultimate collective defense.

This vision appeals to the European bureaucratic mind because it offers a technical solution to a political problem. It focuses on initiatives like the European Defence Fund (EDF) and Permanent Structured Cooperation (PESCO) to fix the fragmentation of Europe’s defense industry. I need to recognize that this logic is rather compelling. A Europe that can build its own tanks, coordinate its own logistics, and command its own battle groups is a better partner for the U.S., and a more resilient actor if the U.S. is absent.

However, from a Romanian perspective, complementarity often feels like a rhetorical sleight of hand. It tries to have it both ways. It claims autonomy while relying on American protection. The “European Pillar” may sound robust in a press release, but what does it actually mean in a high-intensity conflict on the shores of the Black Sea? The complementarist perspective seems to want to build redundancy, which is good, but it seems to me that it also tends to gloss over the hard question of political will. It creates the machinery of a superpower without the nervous system to direct it.

Moreover, many in the East may suspect that a vision of complementarity is a cover for industrial protectionism. When French or German leaders talk about “European sovereignty,” we often hear “buy French jets and German tanks.” From this perspective, Eastern Europeans may very well wary about swapping a security dependency on Washington (which has a proven track record) for an industrial dependency on Paris and Berlin (which have a history of hesitation regarding Russia). At the same time, it is hard to see what alternative there may exist. It is not like the Eastern Europeans are not already very much dependent on the industries, manufacturing powerhouses, and value chains controlled and directed from Berlin and Paris.

The French Bet

Then there is the third vision, the one that causes the most anxiety in Warsaw and Bucharest, even as it seduces the intellectuals of the West: Substitution.

This is the radical option. Gaining ground rapidly in Paris and among the “sovereigntist” wings of the European strategic community, a substitutionist view posits that the Atlantic era is effectively over. The argument most frequently heard is that the U.S. is structurally unreliable, that it is distracted by China, polarized domestically, and increasingly transactional. All are obviously true! Therefore, Europe must prepare to replace the U.S. entirely.

Charles Michel’s snapback at Rutte, telling NATO’s Secretary General that “Donald Trump is not my daddy,” could be seen as the emotional high-water mark of this movement. But the intellectual foundation was laid by Emmanuel Macron long before. The substitutionist vision demands a full spectrum of sovereign capabilities, from autonomous command and control, independent satellite intelligence, and, most controversially, a Europeanized nuclear deterrent.

Macron’s offer to open a dialogue on the French nuclear deterrent’s role in European security is the most significant strategic development in decades. For the first time, a European power is seriously suggesting that it could extend a nuclear umbrella over the continent. For the substitutionists, this is the only logical path. If Article 5 is conditional on the whims of a U.S. president, then it is not a guarantee at all. Europe must possess, therefore, the ultimate sanction itself.

For a Romanian, this vision is terrifyingly seductive. It is seductive because it speaks to a dignity and maturity that Europe has long lacked. Why should 450 million wealthy, educated Europeans depend on voters in Pennsylvania for their safety? There is a certain nobility in the idea that we should stand on our own feet.

But the flaws are rather paralyzing. First, there is the issue of trust. To trade an American nuclear umbrella for a French one requires a leap of faith that Eastern Europe is not ready to make. The United States has spent eighty years demonstrating its commitment to European security. France has spent much of that time jealously guarding its independence and flirting with “strategic dialogue” with Moscow. Would a French president really trade Lyon for Vilnius? I am not that sure. And in the grim calculus of nuclear deterrence, uncertainty can be fatal.

Second, the timeline is all wrong. Building the capabilities envisioned by the substitutionists, the airlift, the deep-strike capacity, the integrated air and missile defense, would take two decades and trillions of euros. We do not have two decades. The Russians are rearming now. The crisis is immediate. To signal a move toward substitution today is to dismantle the existing security architecture before the new one is built. It is akin to demolishing your roof in the middle of a thunderstorm because you have blueprints for a better one.

The Friction of Transition

These three visions are colliding in real-time in Brussels and every other major European capital, creating friction within the alliance that our adversaries are keen to exploit.

The friction is visible in the fights over the “Third Country Rule” in EU defense procurement, where Eastern states fight to keep access to U.S., British, and South Korean weaponry while France pushes for “Buy European” clauses. And it is visible in the command structure debates at SHAPE, where officers struggle to reconcile NATO planning with nascent EU military staffs.

What we are witnessing is a chaotic interregnum. As an Eastern European, I find myself torn. My heart is with Consolidation, as I want the Americans to stay, the old maps to remain valid, and the certitudes of the post-Cold War world to hold. I want to believe that if we just buy enough F-35s and meet the 5% GDP spending targets, Washington will love us again.

But my head tells me that the substitutionists, for all their arrogance, have identified the correct historical trajectory. The United States is changing. The world is changing. A Europe that cannot defend itself is a historical anomaly that is being corrected by the return of great power politics.

My fear is that we are likely to end up with the worst of all worlds. We may pursue a muddled version of complementarity that alienates the Americans enough to accelerate their withdrawal, but fails to build the sufficient muscle to replace them. We might annoy the U.S. with rhetoric of autonomy while remaining militarily impotent.

As the Davos crowds dispersed and the black limousines wound their way down the mountain, the message was clear. The holiday from history is over. And for the first time in eighty years, Europe is realizing that no one else is going to pay for its security.

The next day, they turned to more serious business. The protocol, prepared by Soviet Foreign Minister Vyacheslav Molotov and his Romanian counterpart Ana Pauker, specified which Danube islands belonged to which state. At the end, almost as an afterthought, the document stated that the “Serpent Island, situated in the Black Sea east of the Danube mouths, becomes part of the USSR.”

There was no serious debate. No negotiations over the island’s strategic value or its resources. Ana Pauker belonged to the Muscovite faction of the party. The Romanian delegation just signed. Three months later, on May 23, 1948, officials met on the island itself for the formal handover. In this document, the language quietly changed to declare that the island “was returned to the USSR.” Returned, as if it had always belonged to Moscow and Romania had merely been holding it temporarily.

The island was economically insignificant, but of some strategic importance for military purposes. It enabled Moscow to survey and project force across the western part of the Black Sea and into the Eastern Mediterranean. But what really mattered politically was the Romanian ‘consent’ as a ritual formalizing an outcome already determined by power. The smiles in Moscow between Stalin and Groza, the warm greetings, the jokes about staying in shape of the Romanian Premier were just cheesy and performative ways of covering the hard reality that Romania was formally sovereign but functionally subordinate, and this was how subordination — or, as David Lake would put it, “indirect rule” — worked when you maintained the fiction of sovereign equality and partnership.

In 2026, that logic has become uncomfortably relevant again. Not only in Moscow, however. In Washington, too.

Two Bounded Orders

The Cold War produced two competing hegemonic projects. John Mearsheimer called them “bounded orders” — regional hierarchical orders where great powers structure political and economic relations within their spheres of influence. Both the United States and the Soviet Union used mixtures of inducement and coercion to manage their orders and shape their security, political, and economic architectures. Both provided public goods and extracted, to various degrees, resources. But the dominant logic between these two orders differed fundamentally, and that difference defined everything in the lives of billions of people.

The United States built its order on restraint as the primary mechanism. This was for opportunistic reasons, not pure liberal altruism. As the celebrated liberal theorist G. John Ikenberry put it, Washington bound itself through institutions that constrained American freedom of action and this enabled the persistence of American postwar order. It was about the perpetuation of American global status. The Bretton Woods system subjected U.S. monetary policy to international rules, at least until the Nixon Shock of 1973. The Marshall Plan transferred wealth outward to European nations when tribute could have flowed inward. GATT, the precursor of the World Trade Organization, opened American markets to competitors. The American army secured the key maritime routes and choke points. These choices cost real money, reasonably constrained the arbitrary use of power, and imposed some limits on what Washington could do.

Why accept these constraints? Because restraint purchased consent. By limiting its own arbitrary power, by providing global quasi-public and club goods, by creating genuine benefits for smaller states, the United States made its hegemony voluntary and attractive. States joined the American order because joining offered advantages in terms of security guarantees, market access, an institutional voice in multilateral fora, and a reasonable level of protection from various types threats. These are meaningful and useful quasi-public or club goods that most weaker states crave for. The system was hierarchical, but it was not purely coercive. It has legitimacy derived from voluntary association and subordination in exchange for prized goods.

This legitimacy proved durable. When American relative power started to decline, partners did not defect. As Robert Keohane has famously argued, the lesser powers increased their investment in the system. They built more institutions, deepened transnational and inter-national integration, and expanded collective action. The order survived the gradual erosion of American material preeminence because it was built on something more than fear. It was built on opportunistic, material interests that made lesser powers invested in the bounded order led by the United States. All that because of the benefits and advantages it offered.

The United States still used coercion extensively. It overthrew governments, intervened militarily, applied economic pressure, violated sovereignty when convenient. Iran 1953, Guatemala 1954, Chile 1973, Iraq 2003… the list is long. But these were deviations from stated principles and provoked domestic and international backlash precisely because they contradicted the order’s legitimating logic. When coercion became too visible, it generated domestic and international resistance that imposed symbolic and material costs on decision-makers in Washington. The system’s institutional architecture created friction against arbitrary use of power even when it did not prevent it. Failing to obtain a favorable resolution of the UN Security Council, for instance, did not stop a military intervention, but it did impose political and symbolic costs that made leaders in Washington put significant efforts in gaining the support of the international community.

The Soviet Union built its bounded order differently. Moscow also provided some public goods within its bloc, particularly through energy subsidies to key allies, and technological transfers. But the dominant logic was extractive and coercive. Eastern Europe became an economic hinterland for the Soviet Union. Reparations flowed to Moscow after World War II, when the U.S. was investing billions in Western Europe through the Marshall Plan. Trade terms through COMECON favored the Soviet center. Satellites sold raw materials cheaply and bought Soviet finished goods expensively — the same trade relationship that characterized the dying colonial empires. The Warsaw Pact was less an alliance than a military occupation with institutional veneer.

At least in the first few decades and particularly in Eastern Europe, Moscow did not need to attract voluntary participation in its bounded order. That happened later and particularly vis-a-vis countries from the Global South. What it needed was to make exit from its bounded order impossible. This required constant demonstration of capacity and willingness to use force. Hungary 1956, Czechoslovakia 1968, the Polish crisis of 1980. Communist regimes across Eastern Europe understood that challenging Soviet dominance meant catastrophe. The Soviet bounded order worked by foreclosing any exit options through military presence, economic integration structured to create dependence, and a demonstrated willingness to crush any defiance.

There were no meaningful constraints on Soviet power within the bloc. No institutions that could limit Moscow’s freedom of action. No real reciprocal obligations. Compliance was secured not through buy-in and the provision of inducements but through prohibitive exit costs and the arbitrary use of political and, if needed, military coercion. The USSR maintained the fiction that satellite states were sovereign and partnerships fully voluntary — the rhetoric of socialist fraternity, sovereign equality and mutual benefit. But everyone understood that all these were smokes and mirrors. When Romanian officials signed protocols in Moscow in the cold winter of 1948, when Polish leaders implemented policies they had not chosen, or when Czech reformers were removed by Soviet tanks, sovereignty was shown to be fully performative. Subordination and indirect rule was reality.

But this was not to last. When the Soviet capacity and willingness to coerce weakened under Gorbachev, the entire structure collapsed within a few years. States did not sought to negotiate better terms within the Soviet bounded order. They left entirely. The Warsaw Pact dissolved. COMECON evaporated. Four decades of extraction had produced no loyalty, no investment in the preservation of the order, no sense that the Communist system was worth defending. When the option to leave appeared, no one wanted to stay. Not even the Russian people.

American alliances, by contrast, survived tests that Soviet alliances did not. NATO persisted through Suez, through Vietnam, through Iraq, and through the rise of China. Partners disagreed with Washington, sometimes bitterly. They resisted American pressure, as the French and the Germans did before the American invasion of Iraq, in 2003. But they remained part of the American order. Not because their exit from that order was impossible, but because the hegemonic order led by the U.S. provided genuine value to their societies and economies. Restraint had created buy-in. Voluntary association and the provision of sought-after quasi-public and club goods gave legitimacy. And legitimacy ensured the resilience of the American-led order.

America’s Soviet Turn

Donald Trump’s second term in the White House has systematically reversed this logic. The United States has not withdrawn from the world. Donald Trump is not isolationist, as many claimed by repackaging old pejorative labels used decades ago.1 What has happened, in my view, is that Trump inverted the logic by which it engages with and in the world. The provision of public and club goods has been replaced by extraction. Consent is no longer bought with carrots but forced through coercion. America has stopped acting like the caretaker of its postwar order and started acting like the Soviet Union in post-war Eastern Europe.

Trump’s approach seems to rest on an assumption that structural dependencies have created sufficient lock-in effects that coercion can replace inducements without losing alliance cohesion. He may think that decades of integration around dollar dominance, American military presence, and economic interdependence make exit prohibitively costly. That allies dependent on American security guarantee and market access, both in Europe and Asia, cannot credibly threaten to leave. To some extent, he may even believe that structural power has made consent optional. We see it in how Trump treats the Europeans and the East Asian powers when it comes to trade, to security obligations, or even territorial rights. It is visible in the way the U.S. treats Canada or Denmark when it comes to Greenland. It is not surprising that it is the historical U.S. allies and clients that face the brunt of Trump’s trade policies, not stronger rivals, such as China, who have successfully forced the U.S. to abandon its coercive measures.

This was the Soviet bet, too. Moscow maintained its bounded order not through attraction but through prohibitive exit costs. Military occupation made leaving unthinkable. Economic integration was structured to create asymmetric dependence. Trade denominated in rubles, infrastructure incompatible with Western systems, supply chains oriented toward Soviet production, all made autonomous development or swift switching difficult. The 1981-89 Romanian economic history points in this direction.2 Demonstrated willingness to use force established the credibility of Soviet threats: defiance would be crushed. The Soviet bounded order functioned not because its satellites wished to remain within it, but because attempts to exit entailed catastrophe, as Hungary and Czechoslovakia learned firsthand.

This assumption held for decades. Most Eastern European governments complied under duress. They signed protocols in Moscow they had not negotiated. They implemented policies most of their people opposed. They provided resources they preferred to keep. The system persisted because the costs of defection exceeded the costs of compliance.

Until it did not. When Gorbachev signaled that Moscow would no longer use force to maintain the bloc, the entire structure collapsed within a couple of years. Poland held free elections in June 1989. Hungary opened its border in September. The Berlin Wall fell in November. Czechoslovakia’s Velvet Revolution came in December. The same month, the Romanian dictator was executed by firing squad on Christmas day. By 1991, the Soviet Union itself had dissolved. States did not negotiate better terms within the Soviet sphere. They left entirely, immediately, completely.

Why? Because four decades of extraction had produced no loyalty, no investment in system preservation, no sense that the order was worth defending. Compliance had been purely a function of fear. Remove the fear, and compliance disappeared. The Soviet order had no reservoir of legitimacy to draw on when coercion capacity weakened. When the option to leave appeared, no one wanted to stay. Orders built on voluntary consent bought through the provision of public goods and restraint bend under pressure. Orders built on pure coercion break.

Implications

In light of all this, what should we expect going forward? First, if Trump’s Soviet turn becomes institutionalized, we should expect accelerating alliance erosion — not from Chinese or Russian pressure, but from the leader of the order itself. Alexander Cooley and Dan Nexon called this “exit from below.” When the United States treats allies as subordinates to coerce, uses military force to acquire “access to wealth and resources,” frames sovereignty as conditional on compliance, alliance cohesion becomes unsustainable. The primary threat to American alliances may no longer external wedging. It may very well be internal extraction making exit attractive.

Second, the durability of America’s bounded order now depends almost entirely on whether Washington can impose structural lock-in mechanisms built on coercion and fear that can substitute for restraint and the provision of quasi-public and club goods. Trump’s bet may prove correct in the short-to-medium term if he can identify and impose these lock-in mechanisms. The costs of leaving the dollar-based financial system, reconstructing security arrangements without U.S. inputs, and reorienting trade and value-added chains are genuinely prohibitive in the short-to-medium term. But prohibitive exit costs do not ensure a sustainable order. The Soviet Union maintained its own bounded order for for decades through a diverse set of lock-in mechanisms. However, when the option to leave emerged, or when the benefits of exit exceeded what previously were seen as prohibitive costs of staying in, the system collapsed completely and irreversibly. If American capacity to coerce weakens, if a future major recession or economic crisis undermines the U.S. financial system or American dominance on trade and technology, even temporarily, there will be no reservoir of legitimacy or genuine interest-based desire to sustain the order through crisis.

Third, my bet is that this transformation is not unique to Trump personally. Instead, I think that it reflects a deeper frustration within American domestic politics about the costs of postwar leadership and hegemony. The bipartisan consensus that sustained restraint and the provision of global quasi-public and club goods at the expense of a more generous provision of public goods internally, to the American people, has fractured. The case for bearing asymmetric costs has lost political purchase. Trump is simply articulating what others felt, but the sentiment precedes him and will outlast him. Future administrations may attempt to restore restraint and may even be willing to maintain the provision of key club goods, but the demonstration that extraction is possible makes reversal difficult. Partners now understand that American commitments are conditional, that alliances can function as protection rackets. Trust, once broken, is expensive to rebuild. The damage may be irreversible. And future American presidents, Republican or Democrats, may be a bit too tempted to continue the arbitrariness enabled and legitimized by Trump.

The United States spent seven decades building something different than the Soviets. It wagered that restraint would prove more durable than the arbitrary use of power, that consent bought through inducement would outlast that forced through coercion. Now it seems to have reversed that wager. America’s Soviet turn risks producing the Soviet outcome: an order that functions until it does not, and when it fails, it fails completely — because there is nothing else holding it together.

Subscribe now